#molevolvr

2024-10-02

Maria Doyle (10:46:41): > @Maria Doyle has joined the channel

Janani Ravi (10:46:49): > @Janani Ravi has joined the channel

Mildred Anashie (10:52:10): > @Mildred Anashie has joined the channel

Chioma Onyido (10:53:36): > @Chioma Onyido has joined the channel

Aleru Divine (10:54:48): > @Aleru Divine has joined the channel

USLACKBOT (10:58:15): > JRaviLabhas joined this channel by invitation fromcommunity-bioc.

UAPRC38U8 (10:58:15): > @UAPRC38U8 has joined the channel

Andres Wokaty (10:58:39): > @Andres Wokaty has joined the channel

James Efienemokwu (11:02:06): > @James Efienemokwu has joined the channel

Ekunola Paul O. (11:18:49): > @Ekunola Paul O. has joined the channel

Liliana Garmendia (11:39:43): > @Liliana Garmendia has joined the channel

KritikaVerma (13:06:45): > @KritikaVerma has joined the channel

Leenaa Al-Amin (13:06:45): > @Leenaa Al-Amin has joined the channel

Rebecca Makinde (13:06:45): > @Rebecca Makinde has joined the channel

inyang john (13:06:45): > @inyang john has joined the channel

Earl-Millen Toh Akwe (13:06:45): > @Earl-Millen Toh Akwe has joined the channel

Valentina Buoro (13:06:45): > @Valentina Buoro has joined the channel

Cate Ouma (13:14:19): > @Cate Ouma has joined the channel

inyang john (13:15:28): > Hello Good day everyone > I’m Jay. A pharmacist and also a software engineer that loves to research and improve human lives > I am really excited to commence this journey with you guys and also improve myself.

Rebecca Makinde (13:27:41): > Hello everyone, I’m Rebecca, a data enthusiast. I’m interested in contributing to the MolEvolvR project. I look forward to contributing immensely to this project whilst also learning.

Ekunola Paul O. (13:31:38): > Hello Everyone! > I’m Ekunola Paul, excited to contribute to the MolEvolvR project. I have experience in coding:computer:, bioinformatics:dna:, and enjoy collaborating:handshake:. Looking forward to learning and working with everyone!:blush:

Valentina Buoro (13:33:32): > Hello Everyone, > > My name is Valentina Buoro, an Outreachy applicant. I’m from Nigeria. I have a background in pharmacy and I’m a self-taught software developer. I have a deep interest in using software to aid healthcare and drug research. > > I’m really excited about this project because it aligns with my background and future career path. I look forward to working with the team and making significant contributions during this contribution period:blush:.

Valentina Buoro (13:34:24) (in thread): > Nice to meet you John.Fellow pharmacist turned software developer here:face_with_hand_over_mouth:

Valentina Buoro (13:34:57) (in thread): > Nice to meet you Rebecca:blush:

Valentina Buoro (13:35:39) (in thread): > Nice to meet you Paul:blush:

U03FSJFJF63 (13:36:08): > @U03FSJFJF63 has joined the channel

U050WBC101L (13:36:08): > @U050WBC101L has joined the channel

U05RD2U4H29 (13:36:08): > @U05RD2U4H29 has joined the channel

Ekunola Paul O. (13:37:51) (in thread): > Hi Valentina! > Nice to meet you:blush:. Can’t wait to collaborate with you!:handshake:

Janani Ravi (13:41:57): > set the channel description: Discussion related to MolEvolvR | Developing MolEvolvR for protein analysis: extending the web app, and creating a companion R package | https://github.com/molevolvr and https://github.com/molevolvr2.0

Janani Ravi (13:42:17): > set the channel topic: 2024-10 | Outreachy MolEvolvR contributions

Rebecca Makinde (13:43:55) (in thread): > Hello Valentina:wave:It’s nice to meet you.

Rebecca Makinde (13:44:16) (in thread): > Hi Paul:wave:It’s nice to meet you.

Rebecca Makinde (13:44:31) (in thread): > Hello Jay, I’m Rebecca. It’s nice to meet you.

Mildred Anashie (13:44:58): > Hello Everyone > > I’m Mildred, an outreachy applicant who has interest in the MolEvovlR project.I’m new tothis projectbut I’m ready to learn,contributeand collaborate.

Rebecca Makinde (13:46:40): > Please has anyone been able to successfully install the BiocManager Version?

Rebecca Makinde (13:47:45) (in thread): > Hi Mildred, it’s nice to meet you. > Welcome.

Leenaa Al-Amin (13:48:01): > Hello everyone! My name is Leenaa Al-Amin from Nigeria. I look forward to contributing and collaborating with you all on the MolEvolvR project.:blush:

Rebecca Makinde (13:51:15) (in thread): > Hi Leena, it’s nice to meet you. I’m Rebecca.

Leenaa Al-Amin (13:53:15) (in thread): > I have been able to, let me know ifthere’sanything I can help you with

Ekunola Paul O. (13:53:23) (in thread): > nice to meet you all

Leenaa Al-Amin (13:56:20) (in thread): > Welcome Mildred!

Aleru Divine (14:01:31): > Hello everyone, > My name is Divine Aleru. I’m a biochemist and a full-stack developer, and I enjoy writing from time to time. > I’m really excited to contribute to the MolEvolvR project! While I’m not yet proficient in R, I see this as a fantastic opportunity to learn and grow. > I look forward to collaborating with all of you!:sparkles:

Mildred Anashie (14:02:27) (in thread): > Thank you@Rebecca Makinde

Mildred Anashie (14:02:36) (in thread): > Thank you@Leenaa Al-Amin

Rebecca Makinde (14:02:49) (in thread): > Alright. Thank you so much. > So, I tried running the commands on R, and they worked, and then I installed R studio and tried running the “BiocManager::install(version =”3.19”)” command but I have been getting this error: > > Error: Bioconductor version cannot be validated; no internet connection? See > #troubleshooting section in vignette > > I have been troubleshooting for a long while now but I’m stuck.

Rebecca Makinde (14:10:03) (in thread): > @Leenaa Al-AminThank you so much for wanting to help. I just tried confirming the version, and I saw that it’s 3.19

Leenaa Al-Amin (14:24:58) (in thread): > I can’t seem to figure out what the issue might be, maybe you could ask on the channel or you could try writing the error message inchatgpt.comand try the fixes it suggests

Leenaa Al-Amin (14:29:30) (in thread): > Welcome Aleru!

Valentina Buoro (14:37:40) (in thread): > Thank you all:blush:

Aleru Divine (14:48:48) (in thread): > Thank you@Leenaa Al-Amin

Valentina Buoro (14:48:50) (in thread): > Hi@Rebecca Makinde, I saw this thread on bioconductor’s support page, someone had the same issue you’re facing. You might find it usefulhttps://support.bioconductor.org/p/9153043/

Valentina Buoro (14:49:48) (in thread): > Welcome Leenaa. Nice to meet you

Valentina Buoro (14:50:16) (in thread): > Welcome Divine, nice to meet you

Valentina Buoro (14:51:41) (in thread): > also try to check your internet strength. You can usefast.comto confirm that your internet’s strength is strong

Rebecca Makinde (14:59:15) (in thread): > Thank you so much@Leenaa Al-Amin@Valentina BuoroI was trying to say the error was already fixed.

Valentina Buoro (14:59:54) (in thread): > awesome

Leenaa Al-Amin (15:00:10) (in thread): > Glad to hear that!

Rebecca Makinde (15:00:44) (in thread): > I really appreciate

Cate Ouma (15:59:49): > Hello Everyone. My name is Cateline from Nairobi, Kenya. I am an Outreachy Intern and can’t wait to challenge myself and make meaningful contributions to the MolEvolvR project:blush:

Mildred Anashie (16:00:32) (in thread): > Welcome@Aleru Divine

Valentina Buoro (16:01:25) (in thread): > Welcome Cateline, its nice to meet you

Valentina Buoro (16:05:18) (in thread): > This guide has been curated by the mentor to help us get started on the projecthttps://github.com/JRaviLab/MolEvolvR/blob/main/.github/outreachy_guidelines.md

Cate Ouma (16:08:33): > Thank you Valentina

Mildred Anashie (16:08:45) (in thread): > Welcome@Cate Ouma

Valentina Buoro (16:10:20) (in thread): > you’re welcome:blush:. feel free to ask on the channel if you encounter any issues

Cate Ouma (16:12:29) (in thread): > Thank you@Mildred Anashie

inyang john (16:46:26) (in thread): > @Valentina Buoronice to meet you here.Let’skeep joining Hands

inyang john (16:48:11) (in thread): > Welcome

UAPRC38U8 (17:16:00): > Good day,<!channel>! We are excited to see this overwhelming response!:raised_hands:Welcome to the MolEvolvR (pronounced, mol-evolver)!:writing_hand:**** Here are some useful links to get you startedWhere to contribute? > * Interested in R package development/function documentation/developing case studies/implementing new features?https://github.com/JRaviLab/MolEvolvR > * Interested in front-end/UI/UX? Web-app repo:https://github.com/JRaviLab/molevolvr2.0 > * Please check out issues/discussions in these repos (starting with MolEvolvR) > * Unsure? leave a note on the issue or share your interests via this channel & tag > Background on getting started + what’s MolEvolvR > * Start w/README&outreachy guidelines| able to install? if not, start with revising those! > * MolEvolvR Webapp (1.0, R/Shiny version):jravilab.org/molevolvr > * Preprint:doi.org/10.1101/2022.02.18.461833 > * :lock:Outreachy page(w/ your login) > Other useful resources > * :computer:Getting startedw/:r-logo:,:github:, and other coding basics > * :saluting_face:Code of conduct > :alphabet-white-question:Need help? Start by posting your Qs here > * Post in this channel (in place of DM | any of us can respond & I am not your bottleneck) > * Most Q&A or not sure who to ask – tag us:<@UAPRC38U8><@U03FSJFJF63>(<@U06S82GCD35>) > * Outreachy & Bioconductor Qs:<@U01MW62UDQR>@Maria Doyleon#outreachy > * R package dev:<@U06S82GCD35> > * Back-end:<@U06S82GCD35><@U05RD2U4H29> > * Front-end, UI/UX:<@U05RD2TTT53> > * Data engineering/management:<@U05R1E35XHD> > * In Asia/Pacific time zones diametrically opposite to mine and need to connect asap?@KritikaVermawill be your go-to person! > Most importantly, please help each other outgenerously&have fun**!:raised_hands:

Janani Ravi (17:17:22) (in thread): > Welcome aboard,@inyang john!

Janani Ravi (17:17:32) (in thread): > Welcome aboard,@Rebecca Makinde!

Janani Ravi (17:18:10) (in thread): > Welcome aboard,@Valentina Buoro! and thanks for starting to help out others as well!:raised_hands:

Janani Ravi (17:18:27) (in thread): > Welcome aboard,@Mildred Anashie!

Janani Ravi (17:20:55) (in thread): > Excellent to see problem-solving in action already! thanks to everyone who helped.:raised_hands:

Janani Ravi (17:21:30) (in thread): > Welcome aboard,@Leenaa Al-Amin!

Janani Ravi (17:22:32) (in thread): > Welcome aboard,@Aleru Divine!

Janani Ravi (17:22:55) (in thread): > Welcome aboard,@Cate Ouma! I saw you on a github issue & then the msg disappeared. Let me know if you have any Qs.

Cate Ouma (17:24:43) (in thread): > Hi, I wanted to contribute to the project, but I’m confused since it doesn’t have the “good first issue” tag like the rest

Cate Ouma (17:25:48) (in thread): > I’m not really sure if its a good place for me to start

Seyi Kuforiji (17:48:38): > Hello everyone, my name is Seyi Kuforiji from Lagos, Nigeria. I am skilled with statistical data analytics tools as well as some software engineering tools and practices. I’m currently working towards becoming a well-rounded data analytics engineer. > > I am excited to learn more about R programming language and bioinformatics and to use these tools to solve real-world problems > > The idea of playing a part in improving the development of computational biology and bioinformatics with MolEvolvR is something I really look forward to. > > Outside of work, I like to play chess or watch football. I look forward to collaborating with you all and sharing ideas, it’s nice to meet you!

Mildred Anashie (18:09:01) (in thread): > Thank you@Janani Ravi

Mildred Anashie (18:11:12) (in thread): > It’s nice to meet you@Seyi KuforijiWelcome

Seyi Kuforiji (18:16:00) (in thread): > Nice to meet you too@Mildred Anashie

UAPRC38U8 (18:29:18) (in thread): > Welcome back,@Seyi Kuforiji!

Seyi Kuforiji (18:36:11) (in thread): > Thank you@Janani Ravi, I have a question concerning the “Tagging function to be renamed”. Does this mean we note down these functions to be changed rather than make the changes ourselves?

Valentina Buoro (18:39:55) (in thread): > Thank you@Janani Ravi:blush:. My pleasure:grin:

UAPRC38U8 (19:33:05) (in thread): > Quick Day 2 update: I found most of you on:github:– check your emails for invites. But I’m not:100:sure – so please > 1. confirm that I have the correct detailshere in this doc– otherwise, feel free to edit your details > 2. :star:MolEvolvRso we can check and ensure we have invited the right:github:person! > :speech_balloon:We are updatinglabelson both GitHubissues&discussionsto help you find the correct matches | feel free to take a peek & ask clarifying Qs if you still need to learn what it takes to resolve that issue. If you have any suggestions on what might help you find the right issues, let us know. Also, there are a few larger issues/discussions that multiple folks can work on.:writing_hand:Please assign yourself to the issues/discussions as you see fit by leaving us a comment.

UAPRC38U8 (19:35:29): > :mag_right:I see a few full stack/web developers – if you would like to switch to R-pkg tasks, check out:github:MolEvolvR. If you want to find appropriate web app development tasks, pleasealsocheck out MolEvolvR2.0 or send us a note here by tagging<@U05RD2TTT53><@U05RD2U4H29>.

UAPRC38U8 (19:36:53) (in thread): > cc:@Aleru Divine@Daud Ahmed@Liliana Garmendia@Rhaenyra Liang@Valentina Buoro[those who have indicated this to me]. I can update this list based onthis doctomorrow.

adekemi balogun (20:54:01): > Hello everyone, my name is Adekemi a new outreachy applicant.I’ma data analysis enthusiast looking forward to contributing to this project and working with you all.

UAPRC38U8 (21:42:13) (in thread): > I’veresponded on GitHub.Hope that helps.

Rhaenyra Liang (21:58:32): > Hello everyone:wave:My name is Rhaenyra Liang from Taiwan. I have a background in dance and am a self-taught software developer.Over the past two years, I’ve developed web development skills and gained experience through software developer internships. I also have a strong interest in web app, R, data analysis, and biotechnology. > > I’m very excited to participate in the MolEvolvR project, happy to meet everyone, and look forward to learning and collaborating with the team.:grin:

Rhaenyra Liang (22:02:47) (in thread): > Nice to meet you Jay!

Rhaenyra Liang (22:03:06) (in thread): > Nice to meet you Rebecca!

Rhaenyra Liang (22:03:45) (in thread): > Nice to meet you Paul!

Rhaenyra Liang (22:04:53) (in thread): > Nice to meet you Mildred!

Rhaenyra Liang (22:05:56) (in thread): > Nice to meet you Leenaa!

Rhaenyra Liang (22:06:41) (in thread): > Nice to meet you Aleru!

Rhaenyra Liang (22:14:05) (in thread): > Nice to meet you Adekemi!

Janani Ravi (22:28:12) (in thread): > Welcome aboard,@Ekunola Paul O.!

Cate Ouma (22:54:07): > Hello Team:wave:I have a question for the Automate case study creation Project. From what I’ve understood, I am supposed to identify the proteins to run through MolEvolvR, do multiple alignments and then a phylogenetic analysis to get the evolutionary relationship. I just wanted to confirm if I am in the right track and if this the case, are the case studies and additional visualizations based on my inferred evolutionary relatinship?

UAPRC38U8 (23:07:32) (in thread): > Yes! If youdon’thave your favorite proteins to start your molevolvr analysis with, we can provide you with a few sets of proteins to get going.Let us know whichyou’dprefer.There are a few additional analyses/visualizations related features indiscussions.You can take a look at that later, too, to draw on some inspiration for ‘additionalanalyses’.

Cate Ouma (23:12:42) (in thread): > Well noted and thank you. I’d appreciate getting a set of proteins to work with. I have already identified 2 proteins I can start with, but I’m open to other suggestions that would be better:face_with_cowboy_hat:

UAPRC38U8 (23:18:09) (in thread): > Try yours. Try ours, too. You’ll learn very different things based on where you start! Some proteins may have no domain architectures at all while others may have several interesting ones.Similarly, a few might be species specific while others are very broadly conserved.Start with yours today.You can input multiple proteins at once.Check out additional options, too, to see what works best.We can go with our protein sets tomorrow.

Janani Ravi (23:18:59) (in thread): > Welcome back,@Rhaenyra Liang!

Cate Ouma (23:19:21) (in thread): > Okay:+1:. Thank you

2024-10-03

Seyi Kuforiji (00:16:20) (in thread): > Welcome Adekemi,it’snice to meet you!

Aleru Divine (01:06:47) (in thread): > Thank you so much@Valentina Buoro@Mildred Anashie@Janani Ravi@Rhaenyra Liang

Aleru Divine (01:07:55) (in thread): > Got it!<@UAPRC38U8>

PATIENCE ONAH (01:32:19): > Hello everyone, my name is Patience Onah. I am entry level backend software developer. I look forward to learning and contributing to this project.

Ekunola Paul O. (02:53:53) (in thread): > Thank you@Janani Ravi

Aleru Divine (02:56:41) (in thread): > @Janani RaviI checked MolEvolvR2.0 and didn’t find any issues to attempt. At least none described. > > I would keep an eye out for one.I’llwork together with@Seyi Kuforijion this issue for nowhttps://github.com/JRaviLab/MolEvolvR/issues/7Thank you so much@Seyi Kuforiji:pleading_face: - Attachment: #7 Tagging functions to be renamed > • Identify user-facing functions > • Identify non-conforming functions | _, . –> camelCase names > • Tag internal_use functions to start with .

Aleru Divine (03:07:39) (in thread): > Welcome@adekemi balogun

Aleru Divine (03:07:54) (in thread): > Hi@Rhaenyra Liang, welcome

Aleru Divine (03:08:15) (in thread): > Hi@PATIENCE ONAH, welcome:hugging_face:

Aleru Divine (03:09:00) (in thread): > Hi@Seyi Kuforiji, looking forward to collaborating with you too

Seyi Kuforiji (03:15:29) (in thread): > Hi@Aleru Divine, I left you a message in the issue comment section. Please let me know how I can help you get started!

Mildred Anashie (03:17:04) (in thread): > Hi@PATIENCE ONAHWelcome

Mildred Anashie (03:20:56) (in thread): > Hi@Cate OumaIs it okay to learn and collaborate with you?:pray:

Leenaa Al-Amin (03:27:34) (in thread): > hello@Janani RaviI noticed my GitHub handle “leenaaize” wasn’t added to the google doc, should I edit it in ?

Aleru Divine (03:27:58) (in thread): > Thank you so much!:woman-running::dash:On my way!

Leenaa Al-Amin (03:28:05) (in thread): > Welcome@PATIENCE ONAH

adekemi balogun (03:32:14) (in thread): > Hi@PATIENCE ONAH, welcome

Seyi Kuforiji (03:38:51) (in thread): > Welcome@PATIENCE ONAH:face_with_cowboy_hat:

Komuhangi Tumuhairwe (03:39:57): > Hello everyone, > My name is komuhangi, komu is okay. > I am a software engineer and outreachy applicant, I am excited to be part and to contribute to the MolEvolvR project. > I look forward to learning from you all. > > Cheers!!

adekemi balogun (03:40:01) (in thread): > Thank you so much@Aleru Divine@Rhaenyra Liang@Seyi Kuforiji

inyang john (03:41:19): > Making mistakes is an inevitable aspect of learning. Continue pushing and don’t give up. - File (JPEG): IMG_6335

Seyi Kuforiji (03:53:12) (in thread): > Don’t mention Aleru:muscle::face_with_cowboy_hat:

Seyi Kuforiji (03:54:47): > Hi@Janani Ravi, there are cases where the function names appear as arguments in other functions, should this be noted for change as well? or be left as is?

Rhaenyra Liang (04:28:24) (in thread): > Nice to meet you Komu!

PATIENCE ONAH (04:29:05) (in thread): > Thanks to you all

Rhaenyra Liang (04:30:27) (in thread): > Thank you so much<@UAPRC38U8>@Aleru Divine!

Valentina Buoro (04:51:13) (in thread): > Gotten the invite:white_check_mark:

Valentina Buoro (04:52:35) (in thread): > Noted

George Opondi (05:18:44): > Hello everyone. My name is George Opondi an outreachy applicant from Kenya and I am excited to collaborate with you all on the MolEvolvR project

Cate Ouma (05:32:40) (in thread): > Yeah of course. You should comment on the task in GitHub so that it’ll also be assigned to you

Aniebiet Afia (05:41:26): > Hi everyone, > I am Aniebiet Afia, an Outreachy applicant. I am excited to be here to learn and make positive contributions to this amazing Open Source project. I have always been enthusiastic about contributing to open source projects and I am really happy for this opportunity.

Mildred Anashie (05:47:27) (in thread): > Welcome@George OpondiExcited to collaborate with you too > You can go through the pinned messages to get started

Mildred Anashie (05:48:25) (in thread): > Welcome@Aniebiet Afia:blush:You can go through the pinned messages to get started

UAPRC38U8 (05:49:11) (in thread): > If function names change, all instances/function calls need to change throughout.Also, the same functions are used in the web-app repo as well.That’sthe reason I recommended doing this in phases so the web-app/package don’tend up with breaking changes.cc:<@U06S82GCD35>

Samuel Ogunleye (06:22:04) (in thread): > Hello@Janani RaviI just joined the channel. > > I’m interested in the frontend aspect — are there issues to pickup, please?

Seyi Kuforiji (06:26:24) (in thread): > Right! I have a record of some of the function names ready to be changed and new function names suggestions.I’llattach a text file containing the old and new names and with your approvalI’llimplement these changes on every file and folder that they may be present.Is the web-app/package also in the Molevolvr repository? And because of the recordI have and I can run search on any other folder/repo and replace them with the new names.

Seyi Kuforiji (06:31:40) (in thread): > I’m going on a file-to-file basis and then recording the function names that don’t conform with the required naming conventions. Please look through the text file@Janani Ravi - File (Plain Text): function names update.txt

Seyi Kuforiji (06:32:14) (in thread): > Please let me know what you think.

Aleru Divine (07:01:03) (in thread): > Welcome@George Opondi

Aleru Divine (07:01:16) (in thread): > Welcome@Aniebiet Afia

Valentina Buoro (07:27:04) (in thread): > Welcome Rhaenyra

Valentina Buoro (07:29:48) (in thread): > Welcome Patience.It’snice to have you here. You can go through the pined messages to know how to get started with this projecthttps://community-bioc.slack.com/archives/C07Q4QH8JFN/p1727903760738299 - Attachment: Attachment > Good day, <!channel>! We are excited to see this overwhelming response! :raised_hands::skin-tone-3: Welcome to the MolEvolvR (pronounced, mol-evolver)! > > :writing_hand::skin-tone-3: Here are some useful links to get you started > Where to contribute? > • Interested in R package development/function documentation/developing case studies/implementing new features? https://github.com/JRaviLab/MolEvolvR > • Interested in front-end/UI/UX? Web-app repo: https://github.com/JRaviLab/molevolvr2.0 > • Please check out issues/discussions in these repos (starting with MolEvolvR) > • Unsure? leave a note on the issue or share your interests via this channel & tag > Background on getting started + what’s MolEvolvR > • Start w/ README & outreachy guidelines | able to install? if not, start with revising those! > • MolEvolvR Webapp (1.0, R/Shiny version): jravilab.org/molevolvr > • Preprint: doi.org/10.1101/2022.02.18.461833 > • :lock: <https://www.outreachy.org/outreachy-dec-2024-internship-cohort/communities/bioconductor/#developing-molevolvr-for-protein-analysis-extendin|Outreachy page> (w/ your login) > Other useful resources > • :computer: Getting started w/ :r-logo:, :github:, and other coding basics > • :saluting_face: <https://jravilab.github.io/lab_docs/culture.html#code-of-conduct|Code of conduct> > :alphabet-white-question: Need help? Start by posting your Qs here > • Post in this channel (in place of DM | any of us can respond & I am not your bottleneck) > • Most Q&A or not sure who to ask – tag us: @UAPRC38U8 @U03FSJFJF63 (@U06S82GCD35) > • Outreachy & Bioconductor Qs: @U01MW62UDQR @Maria Doyle on #outreachy > • R package dev: @U06S82GCD35 > • Back-end: @U06S82GCD35 @U05RD2U4H29 > • Front-end, UI/UX: @U05RD2TTT53 > • Data engineering/management: @U05R1E35XHD > • In Asia/Pacific time zones diametrically opposite to mine and need to connect asap? @KritikaVerma will be your go-to person! > Most importantly, please help each other out generously & have fun! :raised_hands::skin-tone-3:

Valentina Buoro (07:31:06) (in thread): > Hi Komu:smile:.It’snice to have you here.You can go through the pined messages to know how to get started with this projecthttps://community-bioc.slack.com/archives/C07Q4QH8JFN/p1727903760738299

Valentina Buoro (07:31:52) (in thread): > Welcome George:hugging_face:

Valentina Buoro (07:32:53) (in thread): > Welcome Aniebiet:hugging_face:

Valentina Buoro (07:49:43): > > Who else is taking the coursera’s intro to r programming course? I found this statement really funny:joy:If the first line of your R script is > > > setwd("C:\Users\jenny\path\that\only\I\have") > > > > I will come into your office and SET YOUR COMPUTER ON FIRE:fire:. > – Jenny Bryan

Valentina Buoro (07:52:11) (in thread): > Apparently the best way to start a new r studio project is to use the ‘usethis’ package ieusethis::use_course("bit.ly/rmd4sci-materials")

Joy (07:57:07): > Helloeveryone, > > My name is Joy, and I’m an Outreachy applicant for the December 2024 to March 2025 round. I’m excited to contribute to the MolEvolvR project and to collaborate with such a talented and diverse group! > > I have experience with R programming, bioinformatics, literature review, and I am eager to learn and grow alongside this community. I’m especially looking forward to contributing to the project’s development and learning more about bioinformatics tools. > > Looking forward to working with you all and learning from this experience!:blush:

Joy (08:28:26) (in thread): > Hello<@UAPRC38U8>, I just joined the channel, I am yet to get the invite on github. I have updated my information on the Google doc. I am interested in the R package dev

UAPRC38U8 (08:50:33) (in thread): > Yes! Definitely recommendusethis, Rproj files, andhere.

UAPRC38U8 (08:52:55) (in thread): > Can you add this to the issue? David and I will look at it and respond there?Thanks!

UAPRC38U8 (08:53:03) (in thread): > Easier to keep track via GitHub.

U05RD2TTT53 (09:13:37): > hi all, I’ll be working on finding some things in the 2.0 repo that folks could contribute to.please stand by, andbear with meas this is a pretty new repo

Matthew Idungafa (09:34:49): > Hello everyone, i am Matthew Idungafa, a fullstack web developer, I hold a BSc in Human Anatomy and also have literature review skills, Looking forward to contributing to this community and having an amazing learning experience here

Seyi Kuforiji (09:40:37) (in thread): > Oh okay,I’vejust done that now.

Seyi Kuforiji (09:46:15) (in thread): > Welcome@Matthew Idungafa! Quick info to get you started:https://community-bioc.slack.com/archives/C07Q4QH8JFN/p1727903760738299 - Attachment: Attachment > Good day, <!channel>! We are excited to see this overwhelming response! :raised_hands::skin-tone-3: Welcome to the MolEvolvR (pronounced, mol-evolver)! > > :writing_hand::skin-tone-3: Here are some useful links to get you started > Where to contribute? > • Interested in R package development/function documentation/developing case studies/implementing new features? https://github.com/JRaviLab/MolEvolvR > • Interested in front-end/UI/UX? Web-app repo: https://github.com/JRaviLab/molevolvr2.0 > • Please check out issues/discussions in these repos (starting with MolEvolvR) > • Unsure? leave a note on the issue or share your interests via this channel & tag > Background on getting started + what’s MolEvolvR > • Start w/ README & outreachy guidelines | able to install? if not, start with revising those! > • MolEvolvR Webapp (1.0, R/Shiny version): jravilab.org/molevolvr > • Preprint: doi.org/10.1101/2022.02.18.461833 > • :lock: <https://www.outreachy.org/outreachy-dec-2024-internship-cohort/communities/bioconductor/#developing-molevolvr-for-protein-analysis-extendin|Outreachy page> (w/ your login) > Other useful resources > • :computer: Getting started w/ :r-logo:, :github:, and other coding basics > • :saluting_face: <https://jravilab.github.io/lab_docs/culture.html#code-of-conduct|Code of conduct> > :alphabet-white-question: Need help? Start by posting your Qs here > • Post in this channel (in place of DM | any of us can respond & I am not your bottleneck) > • Most Q&A or not sure who to ask – tag us: @UAPRC38U8 @U03FSJFJF63 (@U06S82GCD35) > • Outreachy & Bioconductor Qs: @U01MW62UDQR @Maria Doyle on #outreachy > • R package dev: @U06S82GCD35 > • Back-end: @U06S82GCD35 @U05RD2U4H29 > • Front-end, UI/UX: @U05RD2TTT53 > • Data engineering/management: @U05R1E35XHD > • In Asia/Pacific time zones diametrically opposite to mine and need to connect asap? @KritikaVerma will be your go-to person! > Most importantly, please help each other out generously & have fun! :raised_hands::skin-tone-3:

Matthew Idungafa (09:47:19) (in thread): > Thank you@Seyi Kuforiji

Seyi Kuforiji (09:50:06) (in thread): > Don’tmention!Feel free to reach out ifyou’rehaving any trouble getting started!

Matthew Idungafa (11:13:04) (in thread): > @Seyi Kuforiji, please is there any other slack channel we are supposed to join for this project? I’m seeing another link in the readme and it seems i don’t have access

Seyi Kuforiji (11:14:15) (in thread): > What’s the link?

U05RD2TTT53 (12:34:43): > okay, i’m in the process of adding some issues that people can contribute to here. please watch this and claim issues that you’d like to work on.https://github.com/JRaviLab/molevolvr2.0/issues?q=is%3Aissue+is%3Aopen+label%3Aoutreachy

UAPRC38U8 (12:47:16) (in thread): > Seyi, thanks for posting these links. > Welcome aboard,@Matthew Idungafa!

UAPRC38U8 (12:47:39) (in thread): > Welcome aboard,@Joy!

UAPRC38U8 (12:48:17) (in thread): > welcome aboard,@Aniebiet Afia!

UAPRC38U8 (12:48:43) (in thread): > Please add your detailshere& let us know if you have any Qs.

UAPRC38U8 (12:49:10) (in thread): > Welcome aboard,@George Opondi! > Please add your detailshere& let us know if you have any Qs.

UAPRC38U8 (12:50:27) (in thread): > Welcome aboard,@PATIENCE ONAH! > Please add your detailshere& let us know if you have any Qs.

U05RD2TTT53 (12:51:23): > also, i think for now, it’d be good to only claim one issue to work on, to give other people a chance to participate

UAPRC38U8 (12:52:37) (in thread): > <@U05RD2TTT53>, in addition to folks above,@Matthew Idungafa@Joyare also interested in front-end, ui/ux issues. > everyone, check out Vince’s latest notes on new open issues on molevolvr2.0:github:repo.

Matthew Idungafa (13:20:03): > <@U05RD2TTT53>when going through the readme file, i felt some interns may not have experience using docker , so i thought to improve it a bit, kindly review this pr when you have the time:https://github.com/JRaviLab/molevolvr2.0/pull/23 - Attachment: #23 updated readme file with docker commands and made some edits

Seyi Kuforiji (14:38:45) (in thread): > Don’tmention Janani:face_with_cowboy_hat:

Valentina Buoro (15:22:10) (in thread): > Thanks!@Janani Ravi

Valentina Buoro (15:25:04) (in thread): > Welcome Joy:blush:, nice to have you here. You might want to go through the pinned messages on this channel to get started.Looking forward to collaborating with youhttps://community-bioc.slack.com/archives/C07Q4QH8JFN/p1727903760738299 - Attachment: Attachment > Good day, <!channel>! We are excited to see this overwhelming response! :raised_hands::skin-tone-3: Welcome to the MolEvolvR (pronounced, mol-evolver)! > > :writing_hand::skin-tone-3: Here are some useful links to get you started > Where to contribute? > • Interested in R package development/function documentation/developing case studies/implementing new features? https://github.com/JRaviLab/MolEvolvR > • Interested in front-end/UI/UX? Web-app repo: https://github.com/JRaviLab/molevolvr2.0 > • Please check out issues/discussions in these repos (starting with MolEvolvR) > • Unsure? leave a note on the issue or share your interests via this channel & tag > Background on getting started + what’s MolEvolvR > • Start w/ README & outreachy guidelines | able to install? if not, start with revising those! > • MolEvolvR Webapp (1.0, R/Shiny version): jravilab.org/molevolvr > • Preprint: doi.org/10.1101/2022.02.18.461833 > • :lock: <https://www.outreachy.org/outreachy-dec-2024-internship-cohort/communities/bioconductor/#developing-molevolvr-for-protein-analysis-extendin|Outreachy page> (w/ your login) > Other useful resources > • :computer: Getting started w/ :r-logo:, :github:, and other coding basics > • :saluting_face: <https://jravilab.github.io/lab_docs/culture.html#code-of-conduct|Code of conduct> > :alphabet-white-question: Need help? Start by posting your Qs here > • Post in this channel (in place of DM | any of us can respond & I am not your bottleneck) > • Most Q&A or not sure who to ask – tag us: @UAPRC38U8 @U03FSJFJF63 (@U06S82GCD35) > • Outreachy & Bioconductor Qs: @U01MW62UDQR @Maria Doyle on #outreachy > • R package dev: @U06S82GCD35 > • Back-end: @U06S82GCD35 @U05RD2U4H29 > • Front-end, UI/UX: @U05RD2TTT53 > • Data engineering/management: @U05R1E35XHD > • In Asia/Pacific time zones diametrically opposite to mine and need to connect asap? @KritikaVerma will be your go-to person! > Most importantly, please help each other out generously & have fun! :raised_hands::skin-tone-3:

Valentina Buoro (17:54:21): > Hi<@U05RD2TTT53>, I’d like to know if our PRs should be to the main branch in our forked repo or to the main molevolvr2.0 repo

UAPRC38U8 (17:55:25) (in thread): > See if you have permission to create branches. If not, we will invite you as a collaborator. that’s what we’re doing with molevolvr repo.

U05RD2TTT53 (18:14:26) (in thread): > ifyou’vealready created a fork, you can keep it that way. when submitting a PR you should submit it against main of the 2.0 repo, not main of your fork

Valentina Buoro (18:50:00) (in thread): > noted. thanks<@UAPRC38U8><@U05RD2TTT53>

Valentina Buoro (19:26:52): > Hi@KritikaVerma, pls I’d like to know if you have started work on thisissue, will love to collaborate with you on it

KritikaVerma (22:41:01) (in thread): > Yes sure@Valentina Buorofeel free to DM me if you have any questions or doubts.

Joy (23:11:32): > Guys, I need help, while trying to load the package:library(MolEvolvR), I get an error message, how do I get out of this situation?

UAPRC38U8 (23:21:36) (in thread): > Do you have all the other dependencies in place—did installation go through without issues?

UAPRC38U8 (23:24:39): > Ifany of you areunclear about which issue to start with based on the instructions above, please respond to this thread.Also, if there are people interested in multiple issues/discussions, that’s great, but encourage new issue-free folks to take a few over. That will make sure everyone has something to start with.We will populate more front-end, back-end/r-package, and documentation/case-study issues over the weekend for week 2!So ifyou’reready to move on to bigger tasks, stay tuned!

UAPRC38U8 (23:32:47): > It’s great to see the flurry of activity here and the many contributions so far on GitHub — keep up the good work! We’re also loving the camaraderie and how people are readily collaborating with each other!:raised_hands:

Teresia (23:53:50): > Hello, I’m Teresia. Excited to work on the MolEvolvr project amongst yourselves.<@U05RD2TTT53>@Janani Ravikindly check my request on github

2024-10-04

Seyi Kuforiji (00:01:02) (in thread): > What error message are you getting?@Joy

Seyi Kuforiji (00:14:28) (in thread): > I have a suggestion to improve collaboration, but first a question@Janani Ravi. Is there continuous integration set up for the repos given the number of people collaborating? If there’s none I could take up the task or create an automated testing and automated code integration.

Seyi Kuforiji (00:15:58) (in thread): > Welcome Teresia!It’snice to meet you!

Teresia (00:38:28) (in thread): > Nice to meet you too Seyi

Teresia (00:38:48) (in thread): > What part of the project are you working on?

PATIENCE ONAH (00:46:53) (in thread): > Thank you so much@Valentina Buoroand@Janani Ravi

UAPRC38U8 (01:58:18) (in thread): > Excellent Qn.I don’t thinkthat’sset up for these repos yet.Yes, you can get started on that! Wecancheck with<@U06S82GCD35><@U05RD2U4H29>to make sureit’snot in the works already.Another useful addition we were thinking of is unit tests and error handling for these functions.If you get a chance, can you create issues for these too and add the outreachy tag? We will populate it tomorrow and maybe others can hop on those, too!

Valentina Buoro (02:24:47) (in thread): > Thanks!

Joy (03:35:56) (in thread): > Yes<@UAPRC38U8>I have all the other dependencies installed without issues.

Joy (03:40:47) (in thread): > @Seyi KuforijiThis is the message I got below: library(MolEvolvR) > Error in library(MolEvolvR) : there is no package called ‘MolEvolvR’ > > install.packages(“MolEvolvR”) > Installing package into ‘C:/Users/User/AppData/Local/R/win-library/4.4’ > (as ‘lib’ is unspecified) > Warning message: > package ‘MolEvolvR’ is not available for this version of R > > A version of this package for your version of R might be available elsewhere, > see the ideas athttps://cran.r-project.org/doc/manuals/r-patched/R-admin.html#Installing-packages> remotes::install_github(“JRaviLab/MolEvolvR”) > Error in loadNamespace(x) : there is no package called ‘remotes’ - Attachment (cran.r-project.org): R Installation and Administration > R Installation and Administration

Seyi Kuforiji (03:55:16) (in thread): > Hi@Joy, try running this commandinstall.packages("remotes"), before continuing with the rest of your installation. Let me know if this helps!

Valentina Buoro (04:07:22) (in thread): > Nice to meet you Teresia:hugging_face:. Do check the pinned messages on this channel for more information on this project.

Seyi Kuforiji (04:47:32) (in thread): > The MolEvolvR application:face_with_cowboy_hat:

Seyi Kuforiji (05:17:26) (in thread): > Alright!I’vecreated the issues on Github. Please let me know when you have confirmed with David and Faisal!

Valentina Buoro (07:36:33) (in thread): > Side note: anytime you come across an error saying that : there is no package called ‘name-of-package’ , what you’d need to do is to install the package using the install.packages(“name-of-package”) command.

Valentina Buoro (10:29:41): > @inyang johnhttps://community-bioc.slack.com/archives/C07Q4QH8JFN/p1727903760738299 - Attachment: Attachment > Good day, <!channel>! We are excited to see this overwhelming response! :raised_hands::skin-tone-3: Welcome to the MolEvolvR (pronounced, mol-evolver)! > > :writing_hand::skin-tone-3: Here are some useful links to get you started > Where to contribute? > • Interested in R package development/function documentation/developing case studies/implementing new features? https://github.com/JRaviLab/MolEvolvR > • Interested in front-end/UI/UX? Web-app repo: https://github.com/JRaviLab/molevolvr2.0 > • Please check out issues/discussions in these repos (starting with MolEvolvR) > • Unsure? leave a note on the issue or share your interests via this channel & tag > Background on getting started + what’s MolEvolvR > • Start w/ README & outreachy guidelines | able to install? if not, start with revising those! > • MolEvolvR Webapp (1.0, R/Shiny version): jravilab.org/molevolvr > • Preprint: doi.org/10.1101/2022.02.18.461833 > • :lock: <https://www.outreachy.org/outreachy-dec-2024-internship-cohort/communities/bioconductor/#developing-molevolvr-for-protein-analysis-extendin|Outreachy page> (w/ your login) > Other useful resources > • :computer: Getting started w/ :r-logo:, :github:, and other coding basics > • :saluting_face: <https://jravilab.github.io/lab_docs/culture.html#code-of-conduct|Code of conduct> > :alphabet-white-question: Need help? Start by posting your Qs here > • Post in this channel (in place of DM | any of us can respond & I am not your bottleneck) > • Most Q&A or not sure who to ask – tag us: @UAPRC38U8 @U03FSJFJF63 (@U06S82GCD35) > • Outreachy & Bioconductor Qs: @U01MW62UDQR @Maria Doyle on #outreachy > • R package dev: @U06S82GCD35 > • Back-end: @U06S82GCD35 @U05RD2U4H29 > • Front-end, UI/UX: @U05RD2TTT53 > • Data engineering/management: @U05R1E35XHD > • In Asia/Pacific time zones diametrically opposite to mine and need to connect asap? @KritikaVerma will be your go-to person! > Most importantly, please help each other out generously & have fun! :raised_hands::skin-tone-3:

U050WBC101L (11:05:07) (in thread): > Did you follow the installation instruction in the package reporeadme?

Valentina Buoro (13:37:26) (in thread): > yes I did.

Awa (15:19:46): > Hi, I am Awa, an outreachy applicant and I’m thrilled to contribute to the MolEvolvR project specifically on the R package.

Awa (15:22:49): > Hi I am going through the issues on the package and i notice, the good first issues are claimed. Are we permitted to attempt other tasks for example adding unit tests or refactoring functions?

UAPRC38U8 (16:39:48) (in thread): > Hi Awa, what’s your experience and comfort level with coding, R, software dev – package or web-app? Please self-assign issues accordingly!

Awa (16:43:41) (in thread): > I am pretty comfortable with R coding(not expert) and have basics in software development. I am motivated to learn from you and grow and the community. Thanks will get started with some issues.

Zara Adwan (16:57:03): > @Zara Adwan has left the channel

Aleru Divine (17:03:57): > <@U05RD2TTT53>@Janani RaviPlease I need help starting the project. My OSTYPE is**** darwin24.0**, and whenever I try spinning up the project using the./run_stack.shexecutable, I run into issue(s) that I’m not sure what exactly it is. > > Please see attached screenshots - File (PNG): Screenshot 2024-10-04 at 21.45.56 - File (PNG): Screenshot 2024-10-04 at 21.46.24

Aleru Divine (17:33:07) (in thread): > Theaccounting-1container keeps restarting without actually starting at all. While**** db-1anddev-db-1**doesn’t start at all. > > Please see the docker UI screenshot.

Valentina Buoro (20:26:50): > Pls has anyone been able to open thelive web appfor molevolvr 1.0 ? I’ve been getting a connection time out when I try to open it on my browser

Valentina Buoro (20:27:58) (in thread): > welcome Awa:hugging_face:. It’s nice to meet you

UAPRC38U8 (20:56:47) (in thread): > It works for me. Can you try again? Which os/browser are you using?Check your firewall settings in case you thinkthat’san issue.

Samuel Ogunleye (21:56:13) (in thread): > First of all, are you sure you created an env file before running that command? > If no, the command will not likely run perfectly.

Samuel Ogunleye (21:56:42) (in thread): > We can have a sync together and look into the issue whenever you’re active.

UAPRC38U8 (23:43:07): > For those interested in bio/bioinfo, take a look at thecase study issue. I’ve added new instructions and there’s for at least 6-10 folks to work on them in parallel. Respond via GitHub to be assigned or assign yourselves. > > Reminder: those who’ve asked to be assigned to be multiple issues at once, try working on them linearly while others get a chance to contribute (post here to ask for other assignees).In the meantime,we’lladd more issues soon as promised.Thanks! - Attachment: #27 Automate case study creation | starting with rmd report > • Start w/ @the-mayer’s rmd report > • Identify proteins to run through MolEvolvR > • Create case studies > • Which additional summarization/visualizations would help with these case studies?

2024-10-05

Samuel Ogunleye (00:26:15): > I’m unable to make a PR, is there any access restriction?@Janani Ravi

Awa (00:34:00) (in thread): > can you share the error?

Samuel Ogunleye (00:36:08) (in thread): > Read access error. most likely it’s restricted.

Awa (00:37:15) (in thread): > Are you pushing to your fork or the original repo?

Awa (00:39:51) (in thread): > Others have been able to make PRs

Samuel Ogunleye (00:46:30) (in thread): > Cloned repo > set the upstream > pushing my branch

Awa (00:49:40) (in thread): > ok. you have to fork the repo. You can’t push toJRavLab/MolEvolvRyet. Push the changes to your fork and make the PR. > Also, it is a lot easier if you set up ssh with GitHub.

Samuel Ogunleye (00:51:19) (in thread): > It’s actually molevolvr2.0, but I will try forking first. Thanks’

Awa (00:54:05) (in thread): > ok that is the app. Yes fork first. Your welcome.:slightly_smiling_face:

Aleru Divine (01:53:32) (in thread): > Yes I did actually.I created an env file

Teresia (03:48:53) (in thread): > haha… dont be sly

Aleru Divine (03:50:42) (in thread): > What OS are you using@Samuel Ogunleye

Seyi Kuforiji (03:51:54) (in thread): > Idon’treally understand,butI’mworking on the MolEvolvR 1.0, the function issues and error handling.Do you want to be a part of it?:face_with_cowboy_hat:

Valentina Buoro (04:23:34) (in thread): > Google chrome

Valentina Buoro (04:23:46) (in thread): > I’llcheck my firewall settings

Teresia (05:20:09) (in thread): > Unless you are teaching me…am working on MolEvolvR 2.0 on the frontend. Still quite trying to figure out the files

Samuel Ogunleye (08:02:02) (in thread): > MacOS

Valentina Buoro (10:48:59) (in thread): > Resolved this:white_check_mark:

Valentina Buoro (10:54:29) (in thread): > My browser was the problem,changed it to safari

Cate Ouma (13:40:40): > Hi@Janani Ravi, @Team > I’m currently working on the case study automation project, and I have decided to focus on Klebsiella pneumoniae and have successfully filtered the taxonomy data for this species. However, I’m having a hard time finding a dataset that includes both the resistance genes and their drug classes associated with Klebsiella pneumoniae. > > I’ve already checked the Ontology Files in the CARD Database, but I’m still struggling to compile the necessary information. If anyone has suggestions on what I can do or any specific datasets you could recommend, I would greatly appreciate it

UAPRC38U8 (14:12:46) (in thread): > cc:[usergroup content]

Seyi Kuforiji (15:05:29) (in thread): > I’d be happy to help haha, just let me know what you need help with:grin:

U06D7706Y69 (16:32:45) (in thread): > Hi@Cate Oumadownload the CARD data version 3.3.0https://card.mcmaster.ca/download/0/broadstreet-v3.3.0.tar.bz2and go to the file ‘snps’ . You will see the bacteria name, gene (model type), drug (parameter type).

Cate Ouma (18:44:45) (in thread): > Thank you<@U06D7706Y69>:+1:

2024-10-06

Joy (04:59:04): > Hi<@UAPRC38U8>,@Team, I forked and cloned the repo, am I allowed to create a new branch apart from the main branch to work on it?

Seyi Kuforiji (05:09:39) (in thread): > Yes,you’reforked repodoesn’taffected the JRaviLab/MolEvolvR unlessyou’retrying to make a pull request.Basically you can do anything on a forked repo.

Matthew Idungafa (06:16:15) (in thread): > @Joy, yes, that’s the convention, you create a brach, push to it, then it eventually gets merged after your review

Joy (06:52:32) (in thread): > Welcome@George Opondi:hugging_face:

Samuel Ogunleye (06:55:01) (in thread): > Steps > * fork the repo > * clone the forked repo > * create a branch to work on > * push the branch

Joy (06:55:20) (in thread): > Thank you all. I truly appreciate it!

Cate Ouma (10:12:01): > Hi @Team@Janani Ravi. I’m I supposed to create a branch from the Issue page. The Development Section has no “create branch” option - File (PNG): image.png

Samuel Ogunleye (10:34:34) (in thread): > Not at all. Follow these steps > * fork the repo > * clone the forked repo > * create a branch to work on > * push the branch

Samuel Ogunleye (10:35:07) (in thread): > If you still find it difficult, I can jump on a call with you to guide you.

Samuel Ogunleye (10:36:56) (in thread): > First of all, are you familiar with git operations?

Cate Ouma (10:49:54): > My name is Cateline from Kenya. I am completely new to open source and I don’t know what to do. Can you kindly explain what I am supposed to do, especially on the creating branches part?

Samuel Ogunleye (11:06:31) (in thread): > Okay, I respondedtoyour first message but it seems you didn’t see it.Seems you don’t understand git operations — take your time to learn it before attempting this task. > Watch thisshort video here -https://youtu.be/CvUiKWv2-C0 - Attachment (YouTube): Git Tutorial for Absolute Beginners

Cate Ouma (11:09:01) (in thread): > I think I missed this. Let me view the video. > Thank you

Samuel Ogunleye (11:10:55) (in thread): > Alright. Drop a message here if you get stuck

Cate Ouma (11:16:10) (in thread): > Okay:+1:. Thank you > In my task #the case study automation, what should I upload on my created branch. Should I include the source of the data, the filtered data and the code used to filter it, or should I just include the code used to run my filtered data through MolEvolvR with them and visualisations generated?

Janani Ravi (17:41:16) (in thread): > If it’s molevolvr repo, you may have been added as a collaborator already — which means you might be able to create a branch and add a PR with me/<@U06S82GCD35>as reviewers.

Daud Ahmed (22:32:36) (in thread): > @Cate OumaI believe@Samuel Ogunleyeis pointing at the right direction. This is usually how you’re supposed to contribute to open-source. > > Once you will be done with your changes and push it to your remote branch and then when you will open your github usually it will automatically prompt you to create a pull request with the original repo (the one you forked from) or you can always go back to the original repo and go to pull request and then create a pull request against the forks. > > > In case if you want to go deep into it. This seems like a good video explaining it.https://www.youtube.com/watch?v=mqJIGIxRkEs - Attachment (YouTube): How to make Pull Request and Open Source contribution

2024-10-07

Awa (00:37:04): > Hi@Janani Ravi, Some new PRs in R package are dependent on current PRs, do we wait for merge or we can continue working and fix conflicts when necessary?

Samted Uche (teddyCodex) (04:04:00): > I’m getting build failures on some PRs while others involving the same type of changes (function name changes) did not have build failures. Does anyone have experience with this?<!here>

Awa (04:13:36) (in thread): > I think it is because you didn’t load the changes before committing. When you rename functions which are are exported, run > > devtools::load_all() > devtools::document() > > to update the NAMESPACE file.

Awa (04:16:47) (in thread): > Generally it is recommended to deprecate functions( specifically user facing functions) instead of renaming them to give user time to get used to the new functions before complete removal. I think this is also recommended by CRAN and Bioconductor

Samted Uche (teddyCodex) (04:27:05) (in thread): > @AwaJust figured it out before your message came in:smile:Thanks for the hint.

Awa (05:38:47): > hi team, i had a comment to make to those working on renaming/refactoring functions. Instead of simply renaming functions, I suggest a deprecation approach that clearly informs users about the changes. Use warning messages to guide them toward the new function and provide short documentation, including examples of how to adapt existing code. This approach would maintain backward compatibility while encouraging users to adopt improved practices over time. You can refer to thisPRthat deprecates functions in acc2lin - Attachment: #65 defunct functions in acc2lin > Description > > • #35 > > What kind of change(s) are included? > > • Bug fix (fixes an issue). > • Enhancement (updates functions). > > Checklist > > Please ensure that all boxes are checked before indicating that this pull request is ready for review. > > • I have read and followed the CONTRIBUTING.md guidelines. > • I have searched for existing content to ensure this is not a duplicate. > • I have performed a self-review of these additions (including spelling, grammar, and related). > • I have added comments to my code to help provide understanding. > • I have added a test which covers the code changes found within this PR. > • I have deleted all non-relevant text in this pull request template. > • Reviewer assignment: Tag a relevant team member to review and approve the changes.

Awa (05:41:44) (in thread): > In the same change, we can redirect all function definitions within a method to single man page.

Awa (05:44:56) (in thread): > Thehttps://github.com/JRaviLab/MolEvolvRpackage needs cleanup which can have inter-dependent tasks, let’s try to work only on assigned tasks to avoid overlapping work and merge conflicts.

Samted Uche (teddyCodex) (06:43:45) (in thread): > Thoughts on this?@Janani Ravi

UAPRC38U8 (09:40:32) (in thread): > We will revert with eta on merging PRs asap. cc:<@U06S82GCD35><@U03FSJFJF63>

U05RD2TTT53 (11:23:33): > hi all, just FYI, i have notifications for github so if you request a review or otherwise tag me on github, i’ll know immediately, no need to ping me on slack. also reviews may be delayed as we try not to work on weekends, and also we have other projects besides molevolvr that require our attention. thank you for your help and patience.

U05RD2TTT53 (11:32:16): > also, if you are just working on the frontend, and you’re having trouble running therun_stackscript, you can follow the/frontend/readme.mdto just run the frontend separately. though, we’d appreciate creating an issue for therun_stack.shproblem, where you provide all of the commands you’ve run, the log output, your OS version, and any other info you can provide

USLACKBOT (12:43:41): > This message was deleted.

Awa (12:45:45) (in thread): > Hi Cynthia. Welcome. For Microbiome study curation, join#bugsigdb, however if you meant the molEvolvR project, check the pinned messages for first steps.

Cynthia Iwuoha (12:57:11) (in thread): > My name is Cynthia Iwuoha, and I am an Outreachy applicant for the December 2024 to March 2025 program. I am excited to engage with such talented individuals and contribute to the Developing MolEvolvR for protein analysis: extending the web app, and creating a companion R package > > I am eager to make meaningful contributions and learn from the expertises present in this community. I look forward to collaborating with all of you and am grateful for any support and guidance you may provide as I embark on this journey.

Cynthia Iwuoha (12:59:45) (in thread): > @AwaThank you. There was a mixed up, corrected.

UAPRC38U8 (14:06:45) (in thread): > <@U06S82GCD35>/ I will revert asap.

Cynthia Iwuoha (14:52:06): > My name is Cynthia Iwuoha, and I am an Outreachy applicant for the December 2024 to March 2025 program. I am excited to engage with such talented individuals and contribute to the Developing MolEvolvR for protein analysis: extending the web app, and creating a companion R package > > I am eager to make meaningful contributions and learn from the expertises present in this community. I look forward to collaborating with all of you and am grateful for any support and guidance you may provide as I embark on this journey.

UAPRC38U8 (14:55:58) (in thread): > Hi Cynthia, please get started with the pinned posts in this channel & let us know which issues/discussions you want to start contributing to.

UAPRC38U8 (14:56:48) (in thread): > I just responded on theissuejust now. - Attachment (GitHub): Rename the listed function names in R/acc2lin.R · Issue #35 · JRaviLab/MolEvolvR > R/acc2lin.R Original Modified User Facing sink.reset sinkReset add_lins addLineage acc2lin acc2Lineage efetch_ipg efetchIPG ipg2lin IP…

UAPRC38U8 (14:58:00) (in thread): > You will commit your changes to the new branch and submit a PR (pull request) back to the main repo once you’re ready for us to look at it. Please add me/<@U03FSJFJF63>/<@U06S82GCD35>as reviewers when you’re ready to do so.

UAPRC38U8 (14:58:29) (in thread): > Also, please add a note to that issue as to which bacterial species and drugs you are starting with so others can start elsewhere without overlapping w/ you!

Janani Ravi (15:01:11) (in thread): > We’d given collaborator access to MolEvolvR so folks could create a branch or a fork, but yes, for MolEvolvR2.0, you’d be forking to submit PRs. I suppose you’ve been doing this already now. Thanks for stepping in,@Awa!

UAPRC38U8 (15:02:13) (in thread): > Welcome@Awa! [maybe you’ve been doing this already, but …] Please get started with the pinned posts in this channel & let us know which issues/discussions you want to start contributing to.

Samuel Ogunleye (15:02:24) (in thread): > Yeah, already done that.

UAPRC38U8 (15:03:29) (in thread): > Welcome aboard,@Teresia, and let us know if you have any Qs. Otherwise, if you’re communicating via GitHub, that’s great – we’re all set!

UAPRC38U8 (15:03:43) (in thread): > I hope you are unstuck now,@Joy!

Janani Ravi (15:04:49) (in thread): > I see you on GitHub figuring out issues to work on – I think you should be all set!

Samuel Ogunleye (15:07:52) (in thread): > I’m waiting for your feedback on an opened PR<@U05RD2TTT53>

Cynthia Iwuoha (16:55:21): > Thank you everyone for the warm welcome.

Joy (16:59:55) (in thread): > Welcome@Cynthia Iwuoha:hugging_face:

Joy (17:02:09) (in thread): > Yes I am@Janani Ravi. Thanks to you and the team.

Valentina Buoro (18:28:39) (in thread): > Noted

Valentina Buoro (18:29:27) (in thread): > Welcome Cynthia:hugging_face:

UAPRC38U8 (18:52:24): > Week 2 NoteDear MolEvolvR contributors<!here>, > I hope you have found your niche and started contributing to either MolEvolvR or MolEvolvR2.0 (or both)! If you haven’t, please take a careful look at the open issues/discussions (several have room for more contributors). When multiple people are contributing, spell out what you will be working on clearly to avoid duplicated effort. > > You will broadly see 5 kinds of issues: > * R package development – function renaming & documentation w/ roxygen2, function dependency map, error handling, unit testing, continuous integration, and anything else making it package-ready > * New features/enhancement – usually with [FEAT] tag in discussions/issues > * Documentation – help documentation for the R package vignette, web-app (2.0), or case studies (starting withall CARD genes/proteins) > * Front-end – new visualization, particular features within classes of dataviz > * Miscellaneous – belonging to 1 or more categories above, or something else > We appreciate your enthusiastic contributions and are looking through the comments left on issues/discussions/PRs regularly. We typically do not work over the weekends and rarely work evenings (likely overlapping w/ your daytime). So, please be patient once you post your messages and PRs. > > We would like to offer office hours for MolEvolvR starting this week | Wed/Thu, 8.30a MT (1h) |:zoom:link.

Matthew Idungafa (21:46:06): > Welcome@Cynthia Iwuoha, feel free to ask around if you face any challenges

Cate Ouma (22:12:30) (in thread): > Hi<@U06D7706Y69>, is there a way I can view this data from the API without having to download it?

Cate Ouma (22:19:26) (in thread): > I am thinking of extracting the data I need directly from the site instead of using R to filter the species and drug class.@Janani Ravi. Is this a right approach

UAPRC38U8 (22:53:19) (in thread): > Do you suspect that Rcan’thandle this or is there a different reason?

Cate Ouma (22:56:03) (in thread): > Yes, I’ve been trying to filter the data, but all my tables have 0 observations

2024-10-08

Samuel Ogunleye (00:54:06) (in thread): > Have you been able to spin up something? > We can collaborate on this if you don’t mind@Aleru Divine

Aleru Divine (01:01:08) (in thread): > Yesit’sworking now@Samuel OgunleyeThank you so much!

Valentina Buoro (11:47:05) (in thread): > Thanks<@UAPRC38U8>. Looking forward to the office hours.For those of us in WAT time zone,8:30am MT will be 15:30hrs(3:30pm)

2024-10-09

Valentina Buoro (10:31:09): > Hi guys.office hour’s starting now

U03FSJFJF63 (10:32:19) (in thread): > <@UAPRC38U8>will be a little late to join, but David and I are available now.

Seyi Kuforiji (10:32:29) (in thread): > That’stoday yeah?

UAPRC38U8 (10:40:12) (in thread): > Yeah,we’reon now.Same time tomorrow too.

UAPRC38U8 (11:26:41) (in thread): > Great to ‘see’ a few of you during office hours – we will be back again tomorrow morning (same time)! > If you have Qs for MolEvolvR2.0 front-end,<@U05RD2TTT53>may be around tomorrow, too!

U05RD2TTT53 (11:28:51) (in thread): > is this going to be a recurring meeting? if so, could i be invited to a calendar event?

Matthew Idungafa (11:30:04) (in thread): > Where did this meeting hold please?

UAPRC38U8 (11:30:51) (in thread): > We were going to premiere with these 2 and figure out the appropriate freq going forward. Will keep you all posted.

Valentina Buoro (12:09:50) (in thread): > Zoom@Matthew Idungafa. The link is at the end of the first post of this thread

UAPRC38U8 (17:38:36): > Hi all, if you’re interested indata visualizationfor the R-pkg, there are a fewnew issues(created by<@U05RD2U4H29>) &older discussionsfor you to work on. Let us know which ones you would like to contribute to & pitch your game plan – so we & others are in the loop, too!:muscle:

2024-10-10

Valentina Buoro (00:27:10): > For Mac users working on the frontend repo, if you’re unsure of how to run the run_stack.sh file, following these steps helped me.Culled from Apple support website > * Go to the Terminal app on your Mac. > * Use thecdcommand to move into the directory that contains the file you want to make executable. For example:cd molevolvr2.0 > * Enter thechmodcommand. For example:chmod 755 run_stack.sh > * After making the shell script file executable, you can run it by entering its pathname. For example:./run_stack.sh

Valentina Buoro (00:42:33): > Hi<@U05RD2TTT53>and everyone, I’m encountering this error=> ERROR [backend internal] load metadata fordocker.io/rocker/tidyverse:4.3while running the run_stack.sh file. I dont know how to fix this.

Valentina Buoro (00:54:17) (in thread): > I’llgo with your suggestion here. Thanks.https://community-bioc.slack.com/archives/C07Q4QH8JFN/p1728315136664029 - Attachment: Attachment > also, if you are just working on the frontend, and you’re having trouble running the run_stack script, you can follow the /frontend/readme.md to just run the frontend separately. though, we’d appreciate creating an issue for the run_stack.sh problem, where you provide all of the commands you’ve run, the log output, your OS version, and any other info you can provide

Komuhangi Tumuhairwe (04:44:55): > Hello everyone, I hope week two is going okay. > Is anyone using Linux for the MolEvolvR project? > I’ve been having a build issue; > > Bioconductor version 3.19 (BiocManager 1.30.25), R 4.4.1 (2024-06-14) > Installation paths not writeable, unable to update packages > path: /usr/lib/R/library > packages: > boot, codetools, foreign, lattice, Matrix, nlme, spatial > > Does anyone know how I can go about this?

Awa (04:55:04) (in thread): > Hi, what editor are you using? it is a common permission issue.

Komuhangi Tumuhairwe (04:56:11) (in thread): > Thank you for reaching out@Awa,I am using the R studio

Awa (05:00:20) (in thread): > I am guessing you installed RStudio as sudo. > 1. One way is to launch the IDE as sudo before installing the packages. > 2. another that would not require root is to create a writable directory (e.g.,~/R_packages) and set it as the installation path usingoptions(lib.loc = "~/R_packages").

Komuhangi Tumuhairwe (05:06:56) (in thread): > Interesting, let me look at both options and get back.

U05RD2TTT53 (10:30:50): > i’ll be in the office hours in 5

UAPRC38U8 (10:41:10) (in thread): > <@U05AUL5SESJ>:point_up_2:

Mildred Anashie (12:57:14): > Thank you so much for the office hour today@Janani Ravi:pray:I want to know ifit’spossible to request to be assigned to another even if you are currently assigned to one? > Or possibly switch tasks

UAPRC38U8 (12:57:43) (in thread): > sure, what did you have in mind?

Mildred Anashie (12:58:07) (in thread): > I intend to go through the new ones added and pick one > > One I can easily follow through but currently I am assigned to the Automate case study creation | Starting with rmd report

UAPRC38U8 (12:59:25) (in thread): > what have you been working on? what would you like to switch to? > > also, feel free to check in with@Cate Oumahere on#molevolvrif you have more Qs on the case study issue.

Mildred Anashie (13:00:59) (in thread): > YesI’mcurrently in communication with@Cate OumaWe are both trying to navigate the task

Matthew Idungafa (18:00:51) (in thread): > @Mildred Anashie, do you need a extra hand?

Mildred Anashie (18:07:34) (in thread): > Yes please@Matthew Idungafa

2024-10-11

Cate Ouma (02:09:08): > Hi guys, does anyone know how to solve this?@Janani Ravi<@U06D7706Y69><@U06S82GCD35> - File (PNG): Screenshot_20241011-090556~2.png

Awa (03:00:32) (in thread): > Hi@Cate Ouma, it is a workflow error from your PR. I checked it out and the error is as a result of missing dataset files. I also noticed you pushed the compressed dataset. I suggest you: > 1. unzip the dataset locally, place it in recommended directory for external data i.einst/extdatacreate the directory as they’re not yer available.

Awa (03:00:55) (in thread): > This is just one way to do it, others might have better ideas.

Cate Ouma (05:57:51) (in thread): > Thank you for this suggestion Awa. Let me try it out

Matthew Idungafa (08:07:41) (in thread): > @Mildred Anashie, alright, we can talk better in dm

Mildred Anashie (08:08:09) (in thread): > Thank you:pray:

2024-10-12

Mildred Anashie (07:35:16): > Hi<@U05RD2TTT53>all the issues on the front end repo have been claimed,I’lllike to find out if more would be added or I can be assigned to support someone. > > Thanks

Joy (10:57:07): > Thanks<@UAPRC38U8>for the office hours.

U05RD2TTT53 (10:57:28): > i’lltry to come up with some more issues on Monday, but the opportunities for contributing to the 2.0 frontend are limited at the moment unfortunately. you might want to look at the 1.0 repository for web app related things

Mildred Anashie (10:58:53) (in thread): > Thank you:pray:I’ll look at the 1.0 repo, and I look forward to more issues

Janani Ravi (17:58:51): > set the channel description: Discussion related to MolEvolvR | Developing MolEvolvR for protein analysis: extending the web app, and creating a companion R package | https://github.com/jravilab/molevolvr and https://github.com/jravilab/molevolvr2.0

2024-10-13

Cate Ouma (20:32:53): > Hi Team. I am trying to create another pull request after making the changes@Awasuggested. I am now getting another error saying the save-always has been deprecated. Does anyone know what I am doing wrong, and how I can solve it? - File (PNG): image.png

2024-10-14

Samted Uche (teddyCodex) (06:10:24) (in thread): > HI@Cate OumaDid you check that your package passed R CMD check before pushing to the remote repo? > You can do this in your R console or your terminal with the following commands:library(devtools)to load devtools (assuming you’ve already installed it)devtools::load_all() ``devtools::document()this should update certain files based on your changes. Please try this and try your PR again. > > you can find more detailed steps here:https://github.com/JRaviLab/MolEvolvR/blob/main/.github/CONTRIBUTING.md

Samuel Ogunleye (13:48:49): > Since themolevolvr 2.0(frontend)doesn’t have any other issues anymore, can one also contribute tomolevolvr1.0(frontend)? It’s a private repo atm.@Janani Ravi

UAPRC38U8 (13:53:52) (in thread): > Samuel, what kinds of issues are you looking for? there are a few open issues with dataviz tag inmolevolvr– did you look at those?

Samuel Ogunleye (13:57:20) (in thread): > I’m not familiar with theRlanguage, that was why I opted in for 2.0 since it’s solely React and Typescript. Maybe more issues around it would be nice

UAPRC38U8 (14:00:05) (in thread): > Are you familiar with R Shiny w/ css? MolEvolvR1.0 is built primarily with R/Shiny.<@U05RD2U4H29><@U05RD2TTT53>any recommendations?

UAPRC38U8 (14:00:51) (in thread): > If you are unfamiliar with R, you can take a look at the documentation issues inmolevolvrin the meantime or ones with non-package tags.

Samuel Ogunleye (14:01:35) (in thread): > Awesome, thanks.

Cate Ouma (14:34:33) (in thread): > Thank you@Samted Uche (teddyCodex)

UAPRC38U8 (16:03:23): > Are Slack/GitHub working just fine for folks to get their Qs clarified? Or would you like an additional office hour this week, too?:one:thanks, but Slack + GitHub work just fine!:two:one office hour this week might help. [<@U06S82GCD35>can't make it this week.:beach_with_umbrella:]

2024-10-15

Matthew Idungafa (03:54:02) (in thread): > @Janani Ravi, please can i get an invite to the frontend 1.0 repo? can’t seem to find it . My username iskemsguy7.

Mildred Anashie (06:19:02) (in thread): > Yea@Matthew Idungafait redirects to a 404 page from my end. Have you been able to get access to this?

Matthew Idungafa (09:30:41) (in thread): > @Mildred Anashienot yet

U05RD2TTT53 (09:32:37) (in thread): > i’lltry to find more 2.0 frontend issues today, but keep in mind there just reallyaren’tthat many at this point in time. even ifyou’remore familiar with react, you can still probably find regular javascript stuff in 1.0 to fix/work on

Samuel Ogunleye (10:00:40) (in thread): > Can you invite me to the 1.0 repo please? Username is Sproff

Mildred Anashie (10:03:53) (in thread): > Okay<@U05RD2TTT53>Can I get an invite to the 1.0 repo please:pray:My username is MyleeeA > Thank you:pray:

U05RD2TTT53 (10:14:32) (in thread): > we’ll getback to you on that in a few hours. we should be able to provide access so that you can at least fork it

UAPRC38U8 (12:32:01) (in thread): > Office hours tomorrow: > Wednesday,Oct 16, at 9a MT:zoom:https://bit.ly/jravilab-collab_zoom

U05RD2TTT53 (13:35:49): > a general note for those contributing to the 2.0 frontend (or really any repo in any project): look at the rest of the codebase before starting to write your own code, and try to match the style of what’s there already. this includes things like: file names, variable names, data structures, programming patterns, etc. this helps things be more predictable and thus maintainable in the future.

U05RD2TTT53 (15:23:29): > if anyone else would like invites to the 1.0 repo, please let me know by replying to this message in a thread

U05RD2TTT53 (15:23:45) (in thread): > granted read access to the people in this thread

2024-10-16

Daud Ahmed (00:14:42) (in thread): > :hand:

Awa (00:39:03) (in thread): > I would like an invite

Samted Uche (teddyCodex) (03:39:27) (in thread): > I’d like an invite<@U05RD2TTT53>

U05RD2TTT53 (09:27:31) (in thread): > i forgot to mention, please provide your github user name

Valentina Buoro (10:49:52) (in thread): > valentina-buoro is my username

Daud Ahmed (10:52:07) (in thread): > daud99

UAPRC38U8 (11:11:52) (in thread): > If you have any Qs on any of these issues/PRs, we are now on Zoom: > * <@U05RD2TTT53>: 2.0 > * <@U05RD2U4H29>: 1.0 (+ supporting R-pkg, 2.0) > * <@U03FSJFJF63>: bioinfo/bio

UAPRC38U8 (11:37:06) (in thread): > For those of you working on case study generation, esp. starting with the CARD AMR SNPs –> genes –> proteins –> MolEvolvR, take a look at the new comment on the older open issue:https://github.com/JRaviLab/MolEvolvR/issues/27Also, is anyone interested in taking on the new issue of creating anRmd/Qmdmarkdown report generator?https://github.com/JRaviLab/MolEvolvR/issues/102 - Attachment: #27 Automate case study creation | starting with rmd report - Attachment: #102 :new: [FEAT] Generate case study report Rmd/Qmd

Leenaa Al-Amin (15:07:56): > hello@Janani RaviI’m currently working on the network visualization issue, please would it possible for me to get access to the prot dataframe used in these scripts,https://github.com/JRaviLab/MolEvolvR/blob/main/R/networks_domarch.Rhttps://github.com/JRaviLab/MolEvolvR/blob/main/R/networks_gencontext.R

UAPRC38U8 (15:09:38) (in thread): > any downloaded dataframe fromjravilab.org/molevolvrshould serve as a reasonable input df. > do we have readymade examples,<@U05RD2U4H29>?

U05RD2U4H29 (15:30:19) (in thread): > hey<@UAPRC38U8>, not entirely sure if this is the right file, but here’s the domain architecture CSV for the sample job5uNQ9l: - File (CSV): MolEvolvR_domarch.csv

UAPRC38U8 (15:33:00) (in thread): > I think the homolog data from the same code would work for this. It should have the columns you are looking for (if not, leave a note for us in the:github:issue) - File (CSV): molevolvr-5uNQ9l-homolog.csv

UAPRC38U8 (15:54:07) (in thread): > @Leenaa Al-Aminthis is from thehomologtab in the web-app.There are other example codes you can test with, too.

Leenaa Al-Amin (16:01:31) (in thread): > Alright! Thank you so much@Janani Ravi<@U05RD2U4H29>

2024-10-17

Samuel Ogunleye (05:22:24): > Hello@Janani Ravi<@U05RD2U4H29>How can I spin up my local environment on 1.0. > * I ran the build and launch script > * prompted me to build my docker image or pull from registry > * pulling from registry failed due to authentication so I switched to building the image. It’s extremely slow (5/10) since. > Did I miss something please:grimacing:?

Valentina Buoro (07:55:29) (in thread): > Hi Samuel, have you been able to run molevolvr1.0 locally now? I’m also getting an errorj Warning in file(filename, "r", encoding = encoding) :``cannot open file '/data/research/jravilab/molevol_scripts/R/assign_job_queue.R': No such file or directory``Error in file(filename, "r", encoding = encoding) : cannot open the connection

Samuel Ogunleye (08:12:14) (in thread): > Not yet. > > Let’s figure it out together then. Walk me through your process

Valentina Buoro (08:17:55) (in thread): > Okay. > I’m using RStudio for staters. I cloned the repo and created my branch. > > Then I installed shiny package and ran the library. After which I installed the other packages that were required,Rstudio gave me a prompt for thatThen I ran the runApp command which gave me that error > > I basically followed the steps on theShiny documentation.

Andres Wokaty (10:07:23): > Outreachy contributors, this is a gentle reminder to please record your contributions athttps://www.outreachy.org/outreachy-dec-2024-internship-cohort/communities/bioconductor/developing-molevolvr-for-protein-analysis-extendin/contributions/as part of your final application. Thanks for your enthusiasm and hard work! - Attachment (outreachy.org): Outreachy | Internships Supporting Diversity in Tech > An internship program that supports diversity in free and open source software. Learn more by visiting our website!

Valentina Buoro (10:56:01): > Pls are we having office hour today<@UAPRC38U8>?

Valentina Buoro (11:02:43): > No office hours today guys:pray:

U05RD2U4H29 (12:26:15) (in thread): > hey@Samuel Ogunleye,@Valentina Buoro; i realized after we added you to 1.0 that you actually need to be added to a few more repos for things to work. i’ll be posting updated instructions for running the 1.0 app locally ASAP; i’m still working on getting it all set up

U05RD2U4H29 (12:44:41): > hey, for anyone who’s working on 1.0, we need to add you to a few more repos for the stack to run locally. i’ve invited everyone who asked for access to 1.0 to those additional repos. i’m still working on getting everything set up for you to use the new repos; stay tuned for instructions on how to get it set up

U05RD2U4H29 (12:57:56): > ok, so once you’ve accepted all the pending invites, you all can do the following to run the shiny app locally (note that you need docker installed for this to work – seehttps://docs.docker.com/get-started/get-docker/to get it): > 1. open a terminal and cd to a folder where you normally put projects; we’ll be copying the repo to that folder > 2. rungit clone --recursivehttps://github.com/JRaviLab/molevolvr_coordinator.git > 3. runcd molevolvr_coordinator > 4. rungit submodule foreach git checkout main, which checks outmainin all the submodules > 5. copy.env.TEMPLATEto a new file named.env > 6. run./run_stack.sh app, which will do some initial setup, pull the shiny app image, and launch it > 7. once you seeListening onhttp://127.0.0.1:37047in the console, openhttp://localhost:3838and you should see molevolvr 1.0 running locally > 8. you won’t be able to run new jobs, but you can retrieve the job code5uNQ9lif you want to see an example job’s results > more notes: > * the window where you ran./run_stack.sh appwill tail the logs of the containers forever; pressing ctrl-c in that window will shut down the stack > * if you make changes to the shiny app and want to see them without restarting the stack, runtouch molevolvr_app/restart.txtfrom themolevolvr_coordinatorfolder to cause the app to restart > if you want to submit a PR, feel free to fork themolevolvr_appormolevol_scriptsrepos. you can then switch your local repo to track your fork by cding into, say,molevolvr_appand running the following: > 1. OLD_ORIGIN=$( git remote get-url origin ) > 2. git remote set-url origin <URL of your fork> > 3. git remote add upstream ${OLD_ORIGIN} > please let me know if you run into any issues, and thanks for your patience

Samuel Ogunleye (14:15:45) (in thread): > Awesome, thank you.

Valentina Buoro (14:38:20) (in thread): > Thanks<@U05RD2U4H29>

Valentina Buoro (19:08:08) (in thread): > Thank you:pray:

2024-10-18

Samuel Ogunleye (00:44:13) (in thread): > Up and running, thanks!

Daud Ahmed (02:28:14) (in thread): > <@U05RD2U4H29>Thanks

Cate Ouma (04:48:27): > Hi guys, I am having issues with my devtools::check() for my code, it has been running for hours then finally, it returned this error. Any advise on what to do? - File (PNG): image.png

Samuel Ogunleye (05:40:28) (in thread): > This occurs most times if you are calling the same function that was defined inside the body. Just rename your function to something else and it should solve the issue.

Samuel Ogunleye (05:53:19) (in thread): > Hello<@U05RD2U4H29>, I want to make a PR but it seems you didn’t invite me to themolevolvr_apprepo so I’m unable to fork the repo. > > I made changes to themolevolvr_appthroughmolevolvr_coordinatorcloned repo.

Valentina Buoro (05:56:35) (in thread): > Hi<@U05RD2U4H29>I’m getting this errorError response from daemon: image with referenceus-central1-docker.pkg.dev/cuhealthai-foundations/jravilab-public/molevolvr_app:slurm-latestwas found but does not match the specified platform: wanted linux/arm64/v8, actual: linux/amd64when I run./run_stack.sh app

Samuel Ogunleye (05:57:07) (in thread): > What directory were you on@Valentina Buoro?

Valentina Buoro (05:57:46) (in thread): > themolevolvr_coordinatordirectory

Samuel Ogunleye (05:58:11) (in thread): > Before you run that command, make sure your docker is up and running, then you run the./run_stack.sh appcommand.

Valentina Buoro (05:59:10) (in thread): > yeah my docker is running. It was pulling the builds for a while before it exited with that error

Cate Ouma (06:01:41) (in thread): > Thank you Samuel. Is this where the function might be, because that’s the part that has been running for hours - File (PNG): image.png

Samuel Ogunleye (06:05:24) (in thread): > So this issue is with your OS compatibility. You will need to find or build your own image. Try this commanddocker build --platform linux/arm64/v8 -tus-central1-docker.pkg.dev/cuhealthai-foundations/jravilab-public/molevolvr_app:slurm-latest

Valentina Buoro (06:07:30) (in thread): > Thanks! Yeah it’s an OS issue.

Samuel Ogunleye (06:07:47) (in thread): > Not at all, these are just logs.

Samuel Ogunleye (06:08:13) (in thread): > Trace your file changes instead

Valentina Buoro (06:10:39) (in thread): > docker build --platform linux/arm64/v8 -tus-central1-docker.pkg.dev/cuhealthai-foundations/jravilab-public/molevolvr_app:slurm-latestis giving this errorERROR: "docker buildx build" requires exactly 1 argument.

Samuel Ogunleye (06:21:50) (in thread): > Wait you missed the.at the end of that command

Samuel Ogunleye (06:21:50) (in thread): > docker build --platform linux/arm64/v8 -tus-central1-docker.pkg.dev/cuhealthai-foundations/jravilab-public/molevolvr_app:slurm-latest.

Valentina Buoro (06:24:19) (in thread): > oh, it wasn’t included in your initial command. I got this error after adding the periodERROR: failed to solve: failed to read dockerfile: open /var/lib/docker/tmp/buildkit-mount1110348973/Dockerfile: no such file or directory

Cate Ouma (06:45:42) (in thread): > Okay, how can that be done and how will it help?

Valentina Buoro (06:51:08) (in thread): > you can trace your changes in theGitpanel in RStudio.

Valentina Buoro (06:52:10) (in thread): > this articlehttps://stackoverflow.com/questions/49368446/error-evaluation-nested-too-deeply-infinite-recursion-optionsexpressionsmight help@Cate Ouma - Attachment (Stack Overflow): Error: evaluation nested too deeply: infinite recursion / options(expressions=)? in R3.3.2 > I’m trying to use a function in order to read in different models. Using the code below without a function works. When I use the function and call it, I will get the error > > Error: evaluation neste…

Cate Ouma (06:56:04) (in thread): > Thank you for sharing this@Valentina Buoro

Valentina Buoro (06:56:46) (in thread): > You’rewelcome:innocent:

Cate Ouma (06:57:53) (in thread): > I still don’t understand how tracing my file changes will help though:slightly_smiling_face:

Valentina Buoro (06:59:20) (in thread): > You need to rename your functions. Tracing your file changes is just to help you find the particular places you worked on

Cate Ouma (07:03:09) (in thread): > Okay….understood. Let me try it out now:+1:

Valentina Buoro (07:04:14) (in thread): > Awesome! I sent you a picture with some explanation if you need more clarification

U05RD2U4H29 (11:27:44) (in thread): > @Samuel Ogunleyethemolevolvr_appsubmodule inmolevolvr_coordinatoris themolevolvr1.0repo (i.e.,https://github.com/JRaviLab/molevolvr1.0) to which you have read access

U05RD2U4H29 (11:30:45) (in thread): > @Valentina Buorointeresting; i can’t guarantee the 1.0 stack will work on windows if that’s what you’re using, but it’s been run successfully on linux and OS X. i’m not sure why it’d be requesting the arm64 image (which doesn’t exist), but it could be because windows doesn’t support amd64 emulation? i’m not sure. in any case, i’d probably go with trying to build the image yourself, as@Samuel Ogunleyesuggested, but i can’t guarantee it’ll work

Valentina Buoro (11:33:59) (in thread): > My MacBook uses arm64. I also tried building the image myself as@Samuel Ogunleyesuggested,but I gotthis errorERROR: failed to solve: failed to read dockerfile: open /var/lib/docker/tmp/buildkit-mount1110348973/Dockerfile: no such file or directory

U05RD2U4H29 (11:37:05) (in thread): > i’m assuming this is because you’re running that command from the project root, and there aren’t any dockerfiles there. try running./build_images.shfrom the project root instead; it’ll take a long time (something like 1-2 hours on a macbook), but it should work

U05RD2U4H29 (11:37:40) (in thread): > fwiw, even though your macbook is an arm64 architecture, docker is able to run amd64 images using an emulator

Valentina Buoro (11:37:43) (in thread): > Thank you.I’lltry that

Valentina Buoro (11:43:07) (in thread): > ERROR: could not find molevolvr_app/docker: stat molevolvr_app/docker: no such file or directory``* Failed to build molevol_base, abortingI got this error running./build_images.sh

U05RD2U4H29 (11:46:29) (in thread): > @Valentina Buoroi think you might be missing some of the submodules; try running the following from the coordinator folder: > > git submodule update --init --recursive > git submodule foreach git checkout main >

Valentina Buoro (11:49:06) (in thread): > I got this error fatal: Could not read from remote repository. > > Please make sure you have the correct access rights > and the repository exists. > fatal: clone of 'git@github.com:JRaviLab/molevol_scripts.git' into submodule path '/Users/valentinabuoro/projects-react-web/molevolvr_coordinator/molevol_scripts' failed > Failed to clone 'molevol_scripts'. Retry scheduled > Cloning into '/Users/valentinabuoro/projects-react-web/molevolvr_coordinator/molevolvr_app'... > git@github.com: Permission denied (publickey). > fatal: Could not read from remote repository. > > Please make sure you have the correct access rights > and the repository exists. > fatal: clone of 'git@github.com:JRaviLab/molevolvr_app.git' into submodule path '/Users/valentinabuoro/projects-react-web/molevolvr_coordinator/molevolvr_app' failed > Failed to clone 'molevolvr_app'. Retry scheduled > Cloning into '/Users/valentinabuoro/projects-react-web/molevolvr_coordinator/molevolvr_node'... > git@github.com: Permission denied (publickey).

U05RD2U4H29 (11:52:53) (in thread): > @Valentina Buorook, that could be because you don’t have an ssh key associated with github; you can replace the contents of your.gitmodulesfile in the coordinator root with the following: > > [submodule "molevolvr_app"] > path = molevolvr_app > url =[https://github.com/JRaviLab/molevolvr1.0.git](https://github.com/JRaviLab/molevolvr1.0.git)branch = main > [submodule "molevol_scripts"] > path = molevol_scripts > url =[https://github.com/JRaviLab/molevol_scripts.git](https://github.com/JRaviLab/molevol_scripts.git)branch = main > [submodule "molevolvr_node"] > path = molevolvr_node > url =[https://github.com/JRaviLab/molevolvr_node.git](https://github.com/JRaviLab/molevolvr_node.git)branch = main > > then re-run > > git submodule update --init --recursive > git submodule foreach git checkout main >

Valentina Buoro (11:58:40) (in thread): > I got the same error:pensive:

Valentina Buoro (12:01:08) (in thread): > should I try an generate an ssh key for my github account

U05RD2U4H29 (12:04:40) (in thread): > @Valentina Buorosure, that could work; it’s a good thing to have set up anyway. if you do go that route, you can restore your.gitmodulesfile to its previous contents, i.e.git checkout -- .gitmodules

U05RD2U4H29 (12:04:56) (in thread): > fwiw i verified that you have read access to all the repositories you need, assuming your github handle isvalentina-buoro

Valentina Buoro (12:12:47) (in thread): > Yes, it is. Thanks

Valentina Buoro (12:14:08) (in thread): > Will let you know if if I’m able to successfully run the application. Thank you very much for your help

U05RD2U4H29 (12:14:17) (in thread): > @Valentina Buorojust in case you haven’t figured it out already, here’s a guide for generating and registering an ssh key in github and on your local machine,https://docs.github.com/en/authentication/connecting-to-github-with-ssh/generating-a-new-ssh-key-and-adding-it-to-the-ssh-agent

Valentina Buoro (12:14:39) (in thread): > Thank you

2024-10-19

Samted Uche (teddyCodex) (13:31:38): > Hi<@U05RD2U4H29>molevolvr_apphas forking disabled. How can we contribute? - File (PNG): image.png

Matthew Idungafa (13:44:36): > @Samted Uche (teddyCodex)i guess you should just clone and checkout to your branch

Samted Uche (teddyCodex) (13:58:19): > write access error@Matthew Idungafa - File (PNG): image.png

Matthew Idungafa (14:02:30) (in thread): > Oh!. Our mentor will help you then

U05RD2U4H29 (14:08:47): > hi@Samted Uche (teddyCodex), you’re right; forking of private repos appears to be disabled at the org level. i’ll look into it and let you know how to proceed

2024-10-20

Matthew Idungafa (08:45:24): > <@U05RD2U4H29>, just to be clearmolevolvr_appandmolevolvr1.0are both the same repo:https://github.com/JRaviLab/molevolvr1.0right?

Matthew Idungafa (08:46:22) (in thread): > @Samted Uche (teddyCodex), so both repos cannot be forked at the moment, let’s see what happens tomorrow

UAPRC38U8 (09:36:01) (in thread): > Yes, the same.

Matthew Idungafa (09:55:10) (in thread): > @Janani Ravi, alright thanks

Valentina Buoro (16:54:00): > Hi<@U05RD2U4H29>:smile:, after hours of research and trying different things, I’ve figured it out and I’m able to run the application locally. > Things that helped included setting up my ssh key and runningsudo ./run_stack.sh.app.

Valentina Buoro (16:57:18) (in thread): > I also installed Rosetta 2 and enabled docker to use it in emulation mode.

U05RD2U4H29 (20:09:37) (in thread): > if you don’t want to have to run that script via sudo, you can add yourself to the docker group, i.e.sudo usermod -aG docker $USER

U05RD2U4H29 (20:10:04) (in thread): > but nice job figuring it out in any case; running the script via sudo will work, since you’ll have access to run docker containers (as well as everything else)

U05RD2U4H29 (20:12:32): > ok, for those of you working on 1.0 (akamolevolvr_app) and any other subrepo ofmolevolvr_coordinator: i’ve given you write permission tomolevolvr1.0andmolevol_scripts, so you’ll be able to push branches for either repo and submit PRs. you won’t be able to fork them, since forking private repos is disabled at the org level, but being able to submit branches should be sufficient; let me know if you run into issues and i can help you debug them

2024-10-21

Valentina Buoro (03:07:17) (in thread): > Okay. Thank you

2024-10-22

Samted Uche (teddyCodex) (06:51:05): > Hi<!channel>How’s your week going? > Just saying hi:wave:

Valentina Buoro (08:02:16) (in thread): > Intense. Rounding up my contributions:smile:

Matthew Idungafa (09:07:05) (in thread): > @Samted Uche (teddyCodex)We’re doing great and you?

Seyi Kuforiji (13:01:32) (in thread): > Doing great! Hope you’re well

Awa (16:07:07): > Hi , I took over the task of generating case study reports(https://github.com/JRaviLab/MolEvolvR/issues/27). There are many species. How many reports should be generated before moving onto the related task on doing same thing locally with R package(https://github.com/JRaviLab/MolEvolvR/issues/102)? > Species such asStaphylococcus aureusfailed the webapp job for drug:Aminocoumarinwith error (see attachment) Also there is a potential issue with downloading report for successful job. (see attachment)<@U06D7706Y69>@Cate Ouma - Attachment: #27 Automate case study creation | starting with rmd report - Attachment: #102 :new: [FEAT] Generate case study report Rmd/Qmd - File (PNG): Screenshot from 2024-10-22 22-23-40.png - File (PNG): Screenshot from 2024-10-22 22-22-58.png

U06D7706Y69 (16:42:39) (in thread): > @AwaCan you be more specific with the inputs for web app. You have to upload a list of fasta sequences.<@U03FSJFJF63>Do you have any experience with this error?

Awa (16:48:14) (in thread): > I provided fasta sequence for speciesStaphylococcus aureusand drugAminocoumarinThe sequence: > > >CAG39033.1 DNA gyrase subunit B [Staphylococcus aureus subsp. aureus MRSA252] > MTALSDVNNTDNYGAGQIQVLEGLEAVRKRPGMYIGSTSERGLHHLVWEIVDNSIDEALAGYANQIEVVI > EKDNWIKVTDNGRGIPVDIQEKMGRPAVEVILTVLHAGGKFGGGGYKVSGGLHGVGSSVVNALSQDLEVY > VHRNETIYHQAYKKGVPQFDLKEVGTTDKTGTVIRFKADGEIFTETTVYNYETLQQRIRELAFLNKGIQI > TLRDERDEENVREDSYHYEGGIKSYVELLNENKEPIHDEPIYIHQSKDDIEVEIAIQYNSGYATNLLTYA > NNIHTYEGGTHEDGFKRALTRVLNSYGLSSKIMKEDKDRLSGEDTREGMTAIISIKHGDPQFEGQTKTKL > GNSEVRQVVDKLFSEHFERFLYENPQVARTVVEKGIMAARARVAAKKAREVTRRKSALDVASLPGKLADC > SSKSPEECEIFLVEGDSAGGSTKSGRDSRTQAILPLRGKILNVEKARLDRILNNNEIRQMITAFGTGIGG > DFDLAKARYHKIVIMTDADVDGAHIRTLLLTFFYRFMRPLIEAGYVYIAQPPLYKLTQGKQKYYVYNDRE > LDKLKSELNPTPKWSIARYKGLGEMNADQLWETTMNPEHRALLQVKLEDAIEADQTFEMLMGDVVENRRQ > FIEDNAVYANLDF > > > >BAA11086.1 DNA topoisomerase IV GrlB subunit [Staphylococcus aureus subsp. aureus RN4220] > MNKQNNYSDDSIQVLEGLEAVRKRPGMYIGSTDKRGLHHLVYEIVDNSVDEVLNGYGNEIDVTINKDGSI > SIEDNGRGMPTGIHKSGKPTVEVIFTVLHAGGKFGQGGYKTSGGLHGVGASVVNALSEWLEVEIHRDGNI > YHQSFKNGGSPSSGLVKKGKTKKTGTKVTFKPDDTIFKASTSFNFDVLSERLQESAFLLKNLKITLNDLR > SGKERQEHYHYEEGIKEFVSYVNEGKEVLHDVATFSGEANGIEVDVAFQYNDQYSESILSFVNNVRTKDG > GTHEVGFKTAMTRVFNDYARRINELKTKDKNLDGNDIREGLTAVVSVRIPEELLQFEGQTKSKLGTSEAR > SAVDSVVADKLPFYLEEKGQLSKSLVKKAIKAQQAREAARKAREDARSGKKNKRKDTLLSGKLTPAQSKN > TEKNELYLVEGDSAGGSAKLGRDRKFQAILPLRGKVINTEKARLEDIFKNEEINTIIHTIGAGVGTDFKI > EDSNYNRVIIMTDADTDGAHIQVLLLTFFFKYMKPLVQAGRVFIALPPLYKLEKGKGKTKRVEYAWTDEE > LNKLQKELGKGFTLQRYKGLGEMNPEQLWETTMNPETRTLIRVQVEDEVRSSKRVTTLMGDKVQPRREWI > EKHVEFGMQEDQSILDNSEVQVLENDQFDEEEI >

Awa (16:57:49) (in thread): > Additionally you can check the bug job status with id:i31mAuhere:https://jravilab.cuanschutz.edu/molevolvr/?r=&p=upload

Awa (16:58:25) (in thread): > the subsequent job with download issues is here:https://jravilab.cuanschutz.edu/molevolvr/?r=C6kr9s&p=resultSummary

U03FSJFJF63 (17:34:35): > Hello all! Thanks for all the hard work you’ve been putting in. As a heads-up, we’re hosting another round of office hours this Friday at 10AM MT. If you have pressing questions or blockers, obviously feel free to notify us on Slack or Github as normal and don’t feel like you have to wait until the end of the week. > > WHAT: MolEvolvR Office Hours > > WHEN: Friday, October 25th, 10AM MDT/16:00 UTC > > WHERE:This Zoom link

U06D7706Y69 (18:42:34) (in thread): > > error generating report: invalid width argument > is popping up.

2024-10-23

Joy (00:24:41): > Thanks<@UAPRC38U8><@U03FSJFJF63>for the office hour notice

UAPRC38U8 (12:02:01) (in thread): > Is this a PR-related error or a website one, Abhirupa? Did you get a chance to dig in to seewhat’sthrowing this error?

U06D7706Y69 (12:06:34) (in thread): > Its a website error.

UAPRC38U8 (12:09:21) (in thread): > <@U05RD2U4H29><@U06S82GCD35>:point_up_2:

U05RD2U4H29 (12:10:03) (in thread): > commuting, but I’ll look into it when I’m in

U05RD2U4H29 (13:05:51) (in thread): > so, i added some debugging statements to the app to help me track down the error, but after restarting the stack to incorporate the new debug statements, i didn’t run into the issue when generating a report forC6kr9s. i’m a little uncomfortable that it just randomly started working again, but i suppose if the error occurs later on there’ll at least be more in the logs to debug it

Awa (18:11:06): > Hi, I have a question. > In the webapp, do we have a host blast database used for running the blast query or this is done against a remote database? > I am currently working on automating case study reports and I am not sure because the the database on ncbi is large.

Matthew Idungafa (19:02:40) (in thread): > @AwaI have asked a similar question earlier today on this issue:https://github.com/JRaviLab/molevolvr1.0/issues/207Hopefully we will get a reply from@Janani Raviwhen she’s available

UAPRC38U8 (19:22:10) (in thread): > So we have a local server copy of the blast and ipr DBs.<@U05RD2U4H29><@U06S82GCD35>were working on an api to interact with them instead of directly pinging the NCBI/IPR server each time.Then again, we have been looking into IPR lookup service as well as direct use of NCBI api.If you two want to take a lead on those options,<@U05RD2U4H29>can you give a quick update and we can continue from there.Should I create new issues for those, Faisal & David? API + our server vs.directly pinging theirs through our MolEvolvR package?

Matthew Idungafa (21:10:13) (in thread): > @Janani Ravi, we would appreciate that update

U05RD2U4H29 (21:27:48) (in thread): > hey@Matthew Idungafa, frankly it’s not feasible for us to modify the 1.0 stack in a week to allow you all to run it locally. if you need to run jobs, you should do so on the production molevolvr instance,https://jravilab.cuanschutz.edu/molevolvr/. you’ll only be able to run the shiny app locally (e.g., via./run_stack.sh appfrom the root of the coordinator repo), not the entire job-processing stack

U05RD2U4H29 (21:38:02) (in thread): > if you need to run blast/iprscan and can’t use the production molevolvr server (e.g., because you’re writing code for the package), i think the fastest path at the moment would be using the public APIs. for blast, the API docs are here,https://blast.ncbi.nlm.nih.gov/doc/blast-help/developerinfo.html; for interproscan a colleague of mine wrote a little R library that submits jobs to the public API and polls for results,https://github.com/JRaviLab/iprscanr.

2024-10-24

Matthew Idungafa (01:14:12) (in thread): > Thanks<@U05RD2U4H29>, will work with what’s available

Awa (11:09:47): > Hi<!here>. the week is an interesting one. While working on case study report automation with the r package, I have experimented with many interesting approaches. I am sure to round up with the workflow pretty soon. > However, if you would like to get started on running workflows on thewebapp, feel free to borrow my script. > It is perhaps easy to use if you know the pathogen or drug you are interested in. It makes use of the latest CARD > data. example usage:get_card_data(pathogen = "Acinetobacter baumannii", drug = "Beta-lactams")it writes the protein sequence tofiltered_proteins.fastain your WD. submit the file or sequence to thewebappto start a cluster jobs. The rest is straightforward on the app:slightly_smiling_face: > > library(dplyr) > library(stringr) > library(rentrez) > > get_card_data <- function(pathogen = NULL, drug = NULL) { > destination_dir <- "CARD_data" > # Check if CARD data exists > if (!dir.exists(destination_dir)) { > dir.create(destination_dir) > } > if (!file.exists("CARD_data/aro_index.tsv")) { > # Step 1: Download CARD Data > download.file("[https://card.mcmaster.ca/latest/data](https://card.mcmaster.ca/latest/data)", "card_data.tar.bz2") > #unzip("card_data.tar.bz2", exdir = "CARD_data") > system(paste("tar -xjf card_data.tar.bz2 -C", destination_dir)) > } > > # Step 2: Open ARO_index.tsv > aro_data <- read.delim("CARD_data/aro_index.tsv", sep = "\t", header = TRUE) > > # Step 3: Map CARD Short Name > antibiotics <- read.delim("CARD_data/shortname_antibiotics.tsv", sep = "\t", header = TRUE) > pathogens <- read.delim("CARD_data/shortname_pathogens.tsv", sep = "\t", header = TRUE) > > aro_data <- aro_data %>% > mutate( > pathogen = str_extract(CARD.Short.Name, "^[^_]+"), > gene = str_extract(CARD.Short.Name, "(?<=*)[^*]+"), > drug = str_extract(CARD.Short.Name, "(?<=*)[^*]+$") > ) %>% > left_join(pathogens, by = c("pathogen" = "Abbreviation")) %>% > left_join(antibiotics, by = c("drug" = "AAC.Abbreviation")) > > # Sort names > aro_data <- aro_data %>% > arrange(Pathogen, Molecule) %>% > group_by(Pathogen, Molecule) > > # Filter data based on user input > filtered_data <- aro_data > > if (!is.null(pathogen)) { > filtered_data <- filtered_data %>% filter(Pathogen == !!pathogen) > } > if (!is.null(drug)) { > filtered_data <- filtered_data %>% filter(Molecule == !!drug) > } > # Check if filtered data is empty > if (nrow(filtered_data) == 0) { > stop("No data found for the specified pathogen and drug.") > } > # Extract protein accessions > accessions <- filtered_data$Protein.Accession > # Function to fetch FASTA sequence from NCBI > get_fasta <- function(accession) { > entrez_fetch(db = "protein", id = accession, rettype = "fasta") > } > > # Download sequences > fasta_sequences <- lapply(accessions, get_fasta) > writeLines(unlist(fasta_sequences), "filtered_proteins.fasta") > > return("FASTA sequences downloaded to 'filtered_proteins.fasta'.") > } >

Awa (11:18:48) (in thread): > Awesome. thanks for these directives. I am able to run deltablast against ncbi but it doesn’t return taxon ids. > I will definitely tryiprscanr. the standalone interproscan is about ~34GB. I launched the wget before noticing.~in the case study report automation, i am at a the level of cleaning blast results. I noticed i am suppose to do some Assembly summaries and lineage lookups . Can you direct me on how to get these files?~BR. > Awa@Janani Ravi<@U06D7706Y69><@U05RD2U4H29>

Awa (12:01:11) (in thread): > I have been able to manage assembly summaries withdownloadAssemblySummary(). Now looking to get lineage lookups.

Awa (12:31:09) (in thread): > I got the lineage look up. Leaving the question in case another user faces the same issue. > To get the lineage lookup, get the tax dump files fromhttps://ftp.ncbi.nlm.nih.gov/pub/taxonomy/new_taxdump/run the following in console setting the output_file as needed.createLineageLookup("~~/new_taxdump/rankedlineage.dmp", "~~/any_location/lineage_lookup.tsv")

Samted Uche (teddyCodex) (14:17:55): > Hi@Janani RaviIs there any template or guide for creating a final Outreachy application? > Apologies if I have missed any prior communication on this.

UAPRC38U8 (14:20:23): > I’ll be posting more on this later today or tomorrow,@Samted Uche (teddyCodex)& others!

Awa (17:00:38): > I am currently facing some difficulties getting data files i need to complete my scripts for case study automation workflow with the r package. > I was able to get some of the data files such as Assembly summaries and lineage lookup for cleaning blasts results. Now i need the clean lookup table datacln_lookup_tbl.tsvI have tried to construct it to no avail as i cannot find it’s dependent files namely :****TMHMM_lookup.tsv,********SignalP_lookup.tsv,********Phobius_lookup.tsv,********pnthr_db_cln.tsv,********ln-cath.tsv****Is there an alternative way to get it? > > Best.<@U06D7706Y69><@U05RD2U4H29>

U05RD2U4H29 (17:02:10) (in thread): > hey@Awa, i think i can provide that file to you; here it is: - File (TSV): cln_lookup_tbl.tsv

U05RD2U4H29 (17:03:07) (in thread): > for future reference, this file is from thecommon_datafolder that’s created and populated fromhttps://storage.googleapis.com/jravilab-molevolvr/common_data.tar.gzwhen you run the stack via molevolvr_coordinator

UAPRC38U8 (17:03:44) (in thread): > Thanks, Faisal! > > Also,@Awa, if you want to regenerate these tables from create lookup (and similar scripts), go for it! That way, the lookup (lineage and iprscan) tables will also be more current & up-to-date.

2024-10-25

Joy (10:35:05): > This is a reminder of our office hour today - Attachment: Attachment > Hello all! Thanks for all the hard work you’ve been putting in. As a heads-up, we’re hosting another round of office hours this Friday at 10AM MT. If you have pressing questions or blockers, obviously feel free to notify us on Slack or Github as normal and don’t feel like you have to wait until the end of the week. > > > WHAT: MolEvolvR Office Hours > > WHEN: Friday, October 25th, 10AM MDT/16:00 UTC > > WHERE: This Zoom link

UAPRC38U8 (12:51:10): > @Joyet al., I only have these questions based on some of your responses. > 1. Past experience with this free software community > 2. Past experience with other free software communities > 3. Relevant personal or class projects > 4. Community-specific questions (asked twice?) > 5. Internship project timeline > As a mentor, I do not see the prompts (helper text below each Q) they give you to fill these out – e.g., the screenshot pasted by Joy on Zoom. Could you post the text prompts for each Q in this thread? That way, if you have questions about any specifics about the final application, I can clarify them. Thanks!

Cate Ouma (15:41:48) (in thread): > 1. Past experience with this free software community: Please describe your experience before this Outreachy application period with this free software community. You can describe your prior experiences as both a user and a contributor. > 2. Past experience with other free software communities: Please describe your experience before this Outreachy application period with any other free software communities. You can describe your prior experiences as both a user and a contributor. > 3. Relevant personal or class projects: Please describe any relevant projects (either personal, work, or school projects) that helped you gain skills you will use in this project. Talk about what knowledge you gained from working on them. Include links where possible. > 4. Community-specific questions: Some communities or projects may want you to answer additional questions. Please check with your mentor and community coordinator to see if you need to provide any additional information after you save your final application. > 5. Internship project timeline: Please work with your mentor to provide a timeline of the work you plan to accomplish on the project and what tasks you will finish at each step. Make sure take into account any time commitments you have during the Outreachy internship round. If you are still working on your contributions and need more time, you can leave this blank and edit your application later.

Joy (17:12:53) (in thread): > Thank you<@UAPRC38U8>

Joy (17:15:39) (in thread): > Thank you@Cate Ouma

Joy (17:22:04): > Thank you@Janani Ravifor the wonderful office hour, it was insightful,<@U06S82GCD35>

Joy (17:23:33): > Thank you<@U05RD2U4H29>and<@U03FSJFJF63>

UAPRC38U8 (20:43:21): > Dear<!channel>, as we enter the final week of contributions, we’d like you to reflect on your contribution phase in a manner that helps you (and us)! We hope you have had a good time thus far working with us and contributing to our open source project(s)! > 1. :star:If you haven’t already done so, pleaserecord your contributions on Outreachywith as much detail/specifics as possible. > 2. :writing_hand:Ifyou are interested in being considered as an Outreachy intern with MolEvolvR, fill out yourFinal Application— the questions will look like this (see below specific promptsin italicsto help answer these questions:point_down:): > > * Past experience with this free software community > * Past experience with other free software communities > * Relevant personal or class projects > > * Include which languages and platforms you are comfortable with > > * Community-specific questions > > * What are you hoping to learn through the MolEvolvR internship? State some specific technical skills you hope to get out of it. > * There are multiple tags on MolEvolvR – which ones do you best associate with? Which ones do you want to learn to get comfortable? Which MolEvolvR issues and discussions (from which repo) would you want to contribute to during the longer-term internship? > * How does a MolEvolvR internship fit into your career and professional trajectory? Are you interested in academia (grad school, postdoc, research/teaching) or industry (data science, software development)? Is your interest in data science domain-specific (e.g., computational biology, healthcare, finance, etc.)? > > * Internship project timeline > > * How would your MolEvolvR internship timeline look, given your understanding of the project and current state of affairs? Specifically, which issues/discussions do you want to focus on – what would you start with (Month 1), work on for a longer duration (Months 2–3), and wrap up & publish (Month 4)? > > 1. In addition to the application form, if there are ties close to the top, we will use this additionalevaluation rubricto help pick interns. > > * Quality of code > > * Follows a logical flow, adopts modular code design > * Comments and documents code appropriately and consistently > * Demonstrates understanding of code/functions implemented > > * Effective communication > > * Describes issues, attempts workarounds > * Asks questions to troubleshoot or implement changes, engages with code review constructively > > * Adaptability > > * This a complicated, interdisciplinary project – how well did the contributors balance the big picture with small details? Were their suggestions able to advance parts of the project as a whole? > * Were they able to engage with (not necessarily thrive with, but handle!) portions of the project that used field-specific jargon or the minutia of parameters we use? > > * Interest and engagement > > * Did the participant seem interested in specific problems, motivated to resolve them, and comfortable and patient while working in a diverseasynchronous team? > Please let us know if you have any additional questions. Enjoy the final contribution week!:raised_hands:

Samuel Ogunleye (22:14:52) (in thread): > Thank you for the breakdown. > > How about this last question - “Outreachy internship project timeline”? > > How are we going to determine that?@Janani Ravi

UAPRC38U8 (22:22:04) (in thread): > I’dsay start with tags and issues that seem most interesting to you.See if there are certain tags that are appropriate for certain phases of the project. Then it will be worth looking into whether conversations thus far (on issues & discussions) already give you time estimates.If not, asking us to clarify these issues better or working with us on breaking them down into logical next steps will help you get there.

Samuel Ogunleye (22:22:59) (in thread): > Awesome, thank you!

2024-10-26

Awa (05:12:54): > Hey! I’ve pushed some excitingWIPupdates for thecase study report automation!You can now test afull* analysisfor pathogen and/or drug sequence data, complete with BLAST+(ncbi api) and local InterProScan database support!:tada:I’m currently past the analysis stage and am now focused on post-processing withreport_template.Rmd. It’s not merged yet, but you can check out thePRfor more details on dependencies( lookup table data files , ncbi-blast+ tool, InterProScan) . > > To test, install dependencies(adjust paths: currently hardcoded:disappointed_relieved:), load thecase_study branch, and runrun_analysis(file_paths = list(fasta = "~/path to/filtered_proteins.fasta")). > All feedback and review is welcome as I make progress!:raised_hands:@Janani Ravi<@U06D7706Y69><@U05RD2U4H29><@U06S82GCD35>Happy testing!:dna:EDIT: the script functions are not exported to limit potential conflicts from other commits. For now, source the script files to load the functions. (Will export case study functions once commit merge ready) * - Attachment: #111 WIP: Case study report Automation

UAPRC38U8 (09:31:30) (in thread): > cc:@Cate Ouma@Matthew Idungafa—related toyesterday’szoom conversation?

Matthew Idungafa (09:42:44) (in thread): > Yes<@UAPRC38U8>, I’m in contact with@Awa, will reach out if I need anything. Thanks

Cate Ouma (10:50:45) (in thread): > Hello@Janani RaviI had a question about the timeline. What will successful applicants be doing/working on until the end of the internship period. I understand that there are different contributions for the MolEvolvR project e.g the web-app repo, R package development, documentation, case, study automation etc. When answering this part of the Outreachy application, should we just provide the timeline of the Issue we’re currently working on and those related to it or will we progress to other phases contributions as time goes by?

Joy (17:11:57) (in thread): > Thank you<@UAPRC38U8>This sums it up

2024-10-27

UAPRC38U8 (23:15:04) (in thread): > There are now milestones ina fewrepositoriesto also help withthe timeline. Thispartisstillwork in progress, but I hope it helpsfor now.

2024-10-28

Joy (00:21:46) (in thread): > Thank You<@UAPRC38U8>It does help. I would like to find out if there will be more issues that have not been assigned that can be worked on.

Janani Ravi (08:38:45) (in thread): > Maybebut not before the contribution period ends.I’ll be out Tu–Th for a meeting.So, I’ll have limited availability until the end of Oct.

Joy (09:08:00) (in thread): > Duly noted<@UAPRC38U8>Thanks for the reply. Wishing you a lovely meeting ahead!

2024-10-29

Samted Uche (teddyCodex) (04:50:06): > Last day to make your final application guys!! Don’t forget to finalize and submit:eyes:<!channel>

Joy (08:02:54) (in thread): > Thank you@Samted Uche (teddyCodex)

Joy (08:05:20): > Guys, this is a reminder, let’s send in our final application today. Today is the last day!!!<!channel> - Attachment: Attachment > Last day to make your final application guys!! Don’t forget to finalize and submit :eyes: <!channel>

Valentina Buoro (09:17:07) (in thread): > Thanks@Samted Uche (teddyCodex):pray:

Valentina Buoro (09:17:32) (in thread): > Thank you@Joy

Matthew Idungafa (10:03:47): > Congratulations to us all. You guys were amazing!. It’s been real fun

UAPRC38U8 (10:45:23): > Congratulations, indeed,<!channel>! > Excellent:github-octocat:contributions &:slack:energy through & through! Fantastic to see such camaraderie as well; I hope you all enjoyed this phase, the project, and what you learned interacting with each other & us!:raised_hands:It’s been great fun working (and catching up) with you all! > > [Please don’t worry if we have not responded to your PR yet. We are getting through them as quickly as we can. Continue to fill out yourcontributionsas best as possible while waiting.] > > Also, REMEMBER to fill out yourfinal applicationshould you be interested in being considered for an internship.

U03FSJFJF63 (11:58:27): > Thanks again to everybody for all the time, effort, and thoughtful contributions that have really helped improve MolEvolvR! We’ve obviously been thrilled and even a bit surprised by the outpouring of work and care that’s been put in by everyone. > > While the deadline may be imminent, and many members of our team are out on other business this week, if there are outstanding questions remaining, we could host one more office hour tomorrow on Wednesday the 30th.:round_pushpin:If you are interested, please reply to this message and we can arrange it. I’ll post the time and Zoom link if there’s interest. Thanks all!

Valentina Buoro (13:41:27) (in thread): > Thank you<@U03FSJFJF63>.I’minterested

U03FSJFJF63 (15:50:30) (in thread): > All right, one more time! We’ll have another hour tomorrow, so bring questions, thoughts, or feedback. > > WHAT: MolEvolvR Office Hours > > WHEN: Wednesday, October 30th, 9AM MDT/15:00 UTC > > WHERE:This Zoom link

Joy (18:57:24) (in thread): > Thank you<@UAPRC38U8>

Joy (19:01:58) (in thread): > This is awesome<@U03FSJFJF63>, I am highly interested. Thanks

2024-10-30

Valentina Buoro (11:57:57): > Thank you<@U03FSJFJF63><@U05RD2U4H29>@David Mayerfor the office hours today:pray:

Cate Ouma (12:04:22) (in thread): > The meeting has already happened:grimacing:. I messed up on the calculating the time zones and thought it would be starting now:ok_woman:

Valentina Buoro (12:05:46) (in thread): > Yes it has. So sorry@Cate Ouma

Valentina Buoro (12:06:12) (in thread): > But I believe you can still reach out/ask questions here on the slack channel

Cate Ouma (12:07:07) (in thread): > Yes I’ll do that > Thank you

U03FSJFJF63 (12:10:07) (in thread): > Sorry to hear that, Cate! I think we’ve all had time zones foil our plans at least a couple times in the past. > > Like Valentina suggested, please do feel free to drop a topic here in the Slack channel though. We’re still around and can pop in to offer feedback and thoughts between tasks.

Cate Ouma (12:11:23) (in thread): > Noted. I will post my questions here whenever I’m stuck > Thank you:pray:

Matthew Idungafa (12:28:53) (in thread): > @Cate Ouma, same here

Joy (17:02:44): > Thank you<@U03FSJFJF63><@U06S82GCD35><@U05RD2U4H29>for the office hour today, itwas insightful.

2024-10-31

Cate Ouma (09:22:21): > Hi team,<@U05RD2U4H29>,<@U03FSJFJF63>. I have a question concerning my R script that I recorded as a contribution. I have been troubleshooting it and working on making it more generalizable as suggested by the reviewers. However, I am not sure if I should make another R Script(in a new pull request) or replace the Bug-Drug.R Script with the new generalizable code(in the same pull request). Kindly advise

U05RD2U4H29 (09:28:32) (in thread): > hey@Cate Ouma, if it’s the same feature/bugfix please try to get it into your existing PR. if it’s not possible that’s alright, but it makes it easier to compare your changes if it’s in the same PR

Cate Ouma (09:29:22) (in thread): > Okay. Thank you:+1:

2024-11-01

Awa (05:00:08): > Update on case study report automation!:rocket:Hey! Just completed running a “full” MolEvolvR case study pipeline to the point of visualization, and the output(pathogen isAcinetobacter baumanniiand drug isBeta-lactams) files are ready to share(see attachment). For anyone interested in exploring or visualizing the results, I’ve added some sample code in the WIP branch that can help. Since these functions are not yet exported, you’ll need to source the file directly frominst/report/viz_utils.R. There’s still some refinement needed, but it’s a solid starting point. Feel free to try it out and share any ideas or alternatives! > > If you want to run the full pipeline yourself, here’s what you’ll need to set up: > > software: Install NCBI BLAST(just the tool as we make use of ncbi’s remote databases), InterProScan(~30gb) , and cd-hit(blastclust is deprecated) and make sure to set the paths. Install packages as well(readr, rentrez, data.table, …) > complimentary data files: download all additional data(lookup and clean table) here to support the run:Google Drive LinkWould love any feedback or suggestions!:blush: - File (Zip): reportL41Se0_full.zip

Samuel Ogunleye (13:21:36) (in thread): > Awesome, well done!

2024-11-05

Valentina Buoro (06:06:55) (in thread): > Nice work@Awa:rocket:

2024-11-06

Awa (10:46:41) (in thread): > Has anyone tried visualizing the analysis?

2024-11-13

Andres Wokaty (11:40:47): > Hi Outreachy applicants,@Svetlana Ugarcina Perovicwho mentors the Microbiome Study Curation project (aka#bugsigdb) is surveying all Outreachy applicants to learn about you and how we can improve the experience for future contributors. - Attachment: Attachment > SURVEY > > THANK YOU ALL for YOUR amazing contribution, keep up the good work! > > We would appreciate if you can share your experience with us by filling in this short survey https://forms.gle/brqH5M2uTYMKy23x7 > and help us improve experience for future contributors. > > Thank you, > @U1LCB8WEA @UBNSEMS3S @Chioma Onyido @Peace Sandy @Esther Afuape @Svetlana Ugarcina Perovic > > P.S. if you already filled in this survey in some previous rounds, please provide your feedback again, maybe something changed.

UAPRC38U8 (11:54:08) (in thread): > @Awa, we are slowlyyet steadilyclearing our PR & issue backlogs.<@U06S82GCD35><@U05RD2U4H29><@U03FSJFJF63><@U06D7706Y69>– we have these past few PRs on our radar, and we will be testing these out in full soon! Thanks again for these contributions!

2024-11-15

Joy (02:01:51) (in thread): > Thank you@Andres Wokaty

2024-11-18

Awa (04:19:53): > Breakthrough update!Hey , > I’m excited to share a major breakthrough in automating the case study report generation entirely within themolEvolvRpackage for the first case I’ve been working on((pathogen isAcinetobacter baumanniiand drug isBeta-lactams))! > Initially, I faced challenges after transitioning fromblastclust(discontinued) toCD-HITfor clustering. It turns out thatCD-HITproduces a completely different output format, which threw off the visualizations and bugged the pipeline. After some digging and retracing the steps, I identified and fixed the issue! > While at it, I also fine-tuned some functions to improve the overall workflow. Attached, you’ll find the newly generated case study report and the new pipeline output.Important Notes: > * You’ll need some additional R packages likeape,ggtree, andmsa. > * For generating the msa pdf figures, make sure a local TeX tool (likeTeXLive) is installed. > To compare themolEvolvR webapp report with local case study report****, look at ****https://jravilab.cuanschutz.edu/molevolvr/?r=TcDiPe&p=resultSummaryand the attachedreport12.htmlLet me know if you have any feedback or questions! - File (Zip): MolEvolvRsHzFyk_full.zip - File (HTML): report12.html

UAPRC38U8 (08:42:03) (in thread): > Great to hear this, Awa.We’ll look through this PR as well.:muscle:

UAPRC38U8 (19:38:02): > Dear MolEvolvR contributors (<!channel>), while the final Outreachy decisions are being made, please share your quick feedback on this contribution phase. > 1. :writing_hand:Please fill outthis formto share your experience with MolEvolvR & Outreachy. > 2. :slack:Based on your initial experience, if you want to continue contributing remotely, please let us know*. In the longer term, beyond this remote work over the next few months, if you are interested in applying for an internship with us next summer/fall, do send me a note via Slack over the next few weeks/months.[*Please note that Outreachy/Bioconductor can only officially support 1–2 interns, so we’re interested in finding additional mechanisms.] > 3. :mortar_board:Finally, a few of you noted that you want to apply to grad school at CU to work on compbio/bioinformatics. If applying this Fall 2025 and starting grad school in Fall 2026 sounds reasonable, check out this program for more methods development (Computational Biosciences, CPBS) or computational work focusing on (Human medical genetics and genomics, HMGGP). > :star2:Finally, if you learned something from the project and/or enjoyed contributing to MolEvolvR, please star the repo for your/our reference. Thanks again for your fantastic contributions! We’re looking forward to your feedback & suggestions.:raised_hands:

2024-11-19

Joy (02:45:51) (in thread): > Thank you<@UAPRC38U8>.We will sure do that.:hugging_face:

2024-11-21

Valentina Buoro (04:18:31) (in thread): > Thank you<@UAPRC38U8>

Joy (05:50:42): > :rocket:Hello MolEvolvR Team!:star2::tada:I’m thrilled to share my journey with our amazing project and hear more about yours! > Over the past weeks, I’ve been diving into the world of molecular evolution with MolEvolvR, tackling::white_check_mark:Fixing minor errors (those sneaky little bugs:bug:).:white_check_mark:Filing issues to pave the way for significant improvements.:white_check_mark:Setting up dependencies like a pro:computer:.:white_check_mark:RunningR CMD checkand celebrating each “0 ERROR” like it’s a championship win.:trophy::bulb:It’s fascinating to see how each contribution (big or small) fuels the engine of innovation we’re building. Working together, we’re shaping a tool that scientists can rely on for groundbreaking research.:microscope::sparkles:I’d love to know::hammer_and_wrench:What challenges have you faced recently?:dart:What’s the most rewarding task you’ve completed?:rainbow:Any creative ideas for enhancing MolEvolvR? > Let’s keep the momentum going and make molecular evolution accessible and powerful for all researchers out there.:muscle:****#MolEvolvRMagic #TeamworkInEvolution**

Awa (16:07:58) (in thread): > Hi Joy,:grinning::speech_balloon:Challenges:Reading through the existing codebase and debugging has been tough, especially understanding dependencies and ensuring seamless integration.:dart:Rewarding Task:Adding function man pages and working on case study report automation. It’s been great to see tasks directly enhance usability!:star2:Enhancement Idea:Simplifying out-of-package dependencies. For example, integratingiprscanrcould make InterProScan usage smoother, avoiding the need for users to install 30GB+ of data. This ties into my WIP on automating case studies!

2024-11-23

Cate Ouma (06:45:37) (in thread): > Hi@Joy, > It has truly been an incredible journey working with#molevolvrthis #HactoberFest. Here are my short highlights of my journey::hammer_and_wrench:Challenges: Working with real-world datasets like those from the CARD Database has been no small feat. It’s a world apart from clean, structured practice datasets like mtcars.:trophy:Rewarding Task: Successfully retrieving the FASTA Sequences for the bug-drug case study and having 0 ERRORS in my R CMD check! That moment when everything works as expected is unmatched.:tada:

Awa (18:31:54): > Hey team!:wave:As promised, I’m excited to share the instructions for running a case study automation!:tada:Over the last few days, I’ve been working on replacingInterProScan’slocal dependency withiprscanr, and it finally works seamlessly!:raised_hands:This significantly simplifies the setup by reducing the heavy dependency load and diverting it somewhere else( i don’t know how it works yet:sweat_smile:). > The only non-R dependencies remaining areBLASTandCD-HIT, which are simple command-line tools. Below are the complete instructions for setting everything up onWindowsandUbuntu, including how to install and use thegetCaseStudyReportfunction for automating the case study:Setup InstructionsWindows > 1. Install BLAST: > > * Download fromNCBI BLAST+and follow the instructions to install and add it to your PATH. > > > Verify: > > > blastn -version > > > > 2. Install CD-HIT: > > * Download binaries fromCD-HIT GitHub Releasesand extract the files. > * Add the folder containingcd-hit.exeto your PATH. > > > Verify(in cmd): > > > cd-hit -h > > > > 3. install iprscanr in R/Rstudio****devtools::install_github("jravilab/iprscanr")******** or******remotes::install_github("jravilab/iprscanr")******** *******4. Install MolEvolvR( in R/RStudio): > > > install.packages("devtools") > devtools::install_github("awasyn/MolEvolvR", ref = "case_study", dependencies = TRUE) > > > > > Ubuntu > 1. Install BLAST( in terminal): > > > sudo apt-get update > sudo apt-get install ncbi-blast+ -y > > > > 2. Install CD-HIT: > > > sudo apt-get install cd-hit -y > > > > 3. install iprscanr in R/Rstudiodevtools::install_github("jravilab/iprscanr")orremotes::install_github("jravilab/iprscanr")* *******4. Install MolEvolvR( in R/RStudio): > > > install.packages("devtools") > devtools::install_github("awasyn/MolEvolvR", ref = "case_study", dependencies = TRUE) > > Next add this support data. > create common_data folder inside the MolEvolvR library > > # console > common_data_path <- file.path(.libPaths()[1], "MolEvolvR", "common_data") > dir.create(common_data_path, showWarnings = FALSE) > > download data > 1. > > > download.file("[https://drive.google.com/uc?id=1qN3NGUahVZmRniedy_2ijMKaY_TBrwOz](https://drive.google.com/uc?id=1qN3NGUahVZmRniedy_2ijMKaY_TBrwOz)", > destfile = file.path(common_data_path, "cln_lookup_tbl.tsv")) > > 2. Go tohttps://drive.google.com/uc?id=1h_CjURK5laxT7Prhm_U2IqRYlCf6WwKJand download the data. > and copy the file to this path: common_data_path > > # in rstudio terminal > cp /path/to/your/downloaded/lineage_lookup.txt <common_data_path> > > Try out the case study***Once everything is set up, you can test the automation by running the following in R/RStudio: > > library(MolEvolvR) > getCaseStudyReport(pathogen = "Acinetobacter baumannii", drug = "Beta-lactams") > > This will generate and open a detailed case study report.:rocket:The output files are in the parent directory of the current directory where the function is called. > > Let me know how it works for you or if you run into any issues! Looking forward to hearing your feedback.:blush:

Awa (18:48:25) (in thread): > I will try to check the Card data for more single fasta we can test on this function. If many sequences are passed to the function, blast might reject the requests.(I don’t know but it rejected for me)

Awa (18:51:51) (in thread): > This is an example of the output from the function if everything works as it should. - File (HTML): report.html

UAPRC38U8 (19:22:24) (in thread): > Awesome, Awa! Looking forward to testing this soon!!:ok_hand:

Awa (20:12:46) (in thread): > The following can be tried. xn is the number of sequences it will run. so x3 means there are three protein sequences to be reported. > > # Acinetobacter baumannii x8 > getCaseStudyReport(pathogen = "Acinetobacter baumannii", drug = "Beta-lactams") # x1 > getCaseStudyReport(pathogen = "Acinetobacter baumannii", drug = "Fluoroquinolones") # x2 > getCaseStudyReport(pathogen = "Acinetobacter baumannii", drug = "Imipenem") # x1 > #getCaseStudyReport(pathogen = "Acinetobacter baumannii", drug = "NA") # x4 > > # Staphylococcus aureus x27 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Aminocoumarin") # x2 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Fluoroquinolones") # x3 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Fosfomycin") # x3 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Fusidic acid") # x2 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Linezolid") # x1 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Lysocin (E)") # x1 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Mupirocin") # x3 > getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Rifampicin") # x1 > # getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "Daptomycin") # x7 > # getCaseStudyReport(pathogen = "Staphylococcus aureus", drug = "NA") # x4 > > # Klebsiella pneumoniae x13 > getCaseStudyReport(pathogen = "Klebsiella pneumoniae", drug = "Beta-lactams") # x1 > getCaseStudyReport(pathogen = "Klebsiella pneumoniae", drug = "Ceftazidime-Avibactam") # x1 > getCaseStudyReport(pathogen = "Klebsiella pneumoniae", drug = "Colistin") # x1 > getCaseStudyReport(pathogen = "Klebsiella pneumoniae", drug = "Multiple antibiotics") # x1 > # getCaseStudyReport(pathogen = "Klebsiella pneumoniae", drug = "NA") # x9 > > # Pseudomonas aeruginosa x12 > getCaseStudyReport(pathogen = "Pseudomonas aeruginosa", drug = "Aztreonam") # x1 > getCaseStudyReport(pathogen = "Pseudomonas aeruginosa", drug = "Colistin") # x2 > getCaseStudyReport(pathogen = "Pseudomonas aeruginosa", drug = "Fluoroquinolones") # x3 > getCaseStudyReport(pathogen = "Pseudomonas aeruginosa", drug = "Imipenem") # x1 > # getCaseStudyReport(pathogen = "Pseudomonas aeruginosa", drug = "NA") # x5 > > # Enterococcus faecalis x10 > getCaseStudyReport(pathogen = "Enterococcus faecalis", drug = "Chloramphenicol") # x1 > # getCaseStudyReport(pathogen = "Enterococcus faecalis", drug = "Daptomycin") # x9 >

2024-11-25

Joy (02:28:47) (in thread): > This is good to know@Awa, I’ll try it out and revert.

Valentina Buoro (05:16:28) (in thread): > Awesome work@Awa. Looking forward to trying this out and giving feedback.

Valentina Buoro (05:30:00) (in thread): > You ladies are awesome:face_holding_back_tears:@Awa@Cate Ouma@JoyLike Awa said, reading through the existing codebase and understanding how things work was also challenging for me at first. > > But the awesomeness of fixing bugs, figuring out errors and how things worked were extremely rewarding for me.I’mlooking forward to making more significant improvements to MolEvolvr

Awa (07:26:05) (in thread): > Thanks. I am available to assist if you face any errors setting it up. To facilitate response, a screenshot of the error encountered and the command/activity performed will be appreciated.

2024-11-28

Awa (11:13:17): > Congratulations to all successful Outreachy applicantshttps://www.outreachy.org/alums/2024-12/ - Attachment (outreachy.org): Outreachy | Internships Supporting Diversity in Tech > An internship program that supports diversity in free and open source software. Learn more by visiting our website!

Cate Ouma (11:25:50) (in thread): > Congratulations to everyone who has been selected:raised_hands::tada:. The Outreachy experience has really been an eye-opener for me. Thank you to the MolevolvR team for giving us the chance to challenge ourselves and get out of our comfort zone:gratitude-thank-you:. This is definitely not the end of our open source journey. Looking forward to contributing more to @MolevolvR and other open source communities:face_with_cowboy_hat:

UAPRC38U8 (11:29:00) (in thread): > Congratulations, indeed!:tada:We have at least two in our midst,@Seyi Kuforiji&@Aleru Divine!We are stillhopingto work out thefundingsituationwithOutreachy because we’re a new project. Please stay tuned. In the meantime, if you’re interested in continuing to contribute remotely, please fill out the formhere.Irrespective of whether you become an official intern or not, please know that your contributions will be appropriately acknowledged on GitHub and any resulting products like publications, presentations.

UAPRC38U8 (11:41:28) (in thread): > Thank you, Cate!:100:totally agree, this is just a start!:muscle:

Aleru Divine (11:41:39) (in thread): > Thank you so much<@UAPRC38U8>:relaxed:

Aleru Divine (11:42:40) (in thread): > Thank you so much@Awa@Cate Ouma:blush:

Seyi Kuforiji (11:44:39) (in thread): > Thank you@Janani Raviand thank you all!

Samted Uche (teddyCodex) (12:00:56) (in thread): > Thanks a lot@Janani Raviand team<@U05RD2U4H29><@U05RD2TTT53><@U06S82GCD35>And to the other fantastic contributors > I look forward to continuouslearning and working with all of you.

Joy (15:38:49) (in thread): > Congratulations to the successful applicants!

Matthew Idungafa (16:11:01): > Congratulations to the selected interns