#rcwl

2020-07-29

Qian Liu (14:21:03): > @Qian Liu has joined the channel

Qian Liu (14:21:04): > set the channel description: The Bioconductor toolset for usage and development of reproducible bioinformatics pipelines in CWL. This is intended to be a community efforts and we welcome collaborations and contributions from the community. Look forward to working with you!

Qiang Hu (15:42:05): > @Qiang Hu has joined the channel

Leonardo Collado Torres (18:00:52): > @Leonardo Collado Torres has joined the channel

Nick Eagles (18:00:59): > @Nick Eagles has joined the channel

Randall Johnson (18:01:18): > @Randall Johnson has joined the channel

Leonardo Collado Torres (18:04:34): > Hi! Nick@Nick Eagleshas been working at LIBD on anextflowbased pipeline, and while it’s probably too far now to pivot toRcwl, I’m just curious if you plan some “pipeline import/conversion” scripts or guides fromnextflow,snakemake, etc toRcwlor if there are other tools out there that can get you halfway there. > > Also Nick, maybe you can mention to them some of the complicated dependency graph issues you encountered. I think one is for example, you have a pipeline that is step a) align, b) quantify gene counts. But what if you want to have multiple options for the aligner. Then you either have to make multiple nextflowmain.nfscripts (right Nick?) or have people manually tweak things. > > Thanks again for your workshop presentation!

Qiang Hu (18:14:54): > Hi@Leonardo Collado Torres, great suggestion. Actually, the snakemake and WDL provide tools to convert its scripts to CWL. I don’t know whether there is a similar tool for nextflow. We have functionreadCWLto load CWL pipelines to Rcwl object.

Qiang Hu (18:20:30): > We have helper functions to customize theRcwlpipelines easily, so we don’t have to modify any configuration file manually. For example, we want to change a default argument for one tool in a heavy pipeline. We can change it likeargument(heavyPipeline, "workflow1/worflow2/tool") <- '--abc', then the heavyPipeline is ready to use.

Leonardo Collado Torres (18:32:23): > awesome, thanks for the info Qiang!

2020-07-30

Nick Eagles (07:40:34) (in thread): > In nextflow you would likely make one process inmain.nffor each aligner and use if statements or thewhendeclaration. Then a command-line option would specify the aligner to use. It’s fairly easy for the user to choose an aligner from there, but managing input and output channels with conditional execution can get somewhat complicated (and maybeRcwldoes it more elegantly)

Nick Eagles (07:46:54) (in thread): > In general I think nextflow’s channel management is pretty difficult at first (understanding operations liketoList(),flatten(), etc)

CristinaChe (08:51:11): > @CristinaChe has joined the channel

Hyun-Hwan Jeong (08:59:08): > @Hyun-Hwan Jeong has joined the channel

Nick Owen (09:12:36): > @Nick Owen has joined the channel

Qiang Hu (10:12:30) (in thread): > Thanks for sharing your experience with nf.Rcwlhas a simple way to manage inputs for parallel computing, e.g. a list for a list of samples per input parameter.

Yuri Kotliarov (11:59:17): > @Yuri Kotliarov has joined the channel

2020-07-31

bogdan tanasa (13:56:48): > @bogdan tanasa has joined the channel

2020-08-13

Qiang Hu (13:46:52): > A request from the workshop is implemented to therequirementsfunction. It is much easier to manage the container in any step. > For example, thebwaMRecalcombines multiple workflows (bwaAlign, markdup, BaseRecal). We can change thebwaversion under thebwaAlignworkflow. > > > cwlInstall("pl_bwaMRecal") > > requirements(bwaMRecal, "bwaAlign/bwa") <- list(requireDocker("biocontainers/bwa:v0.7.17_cv1")) > > requirements(bwaMRecal, "bwaAlign/bwa")[[1]] > $class > [1] "DockerRequirement" > > $dockerPull > [1] "biocontainers/bwa:v0.7.17_cv1" >

2020-10-11

Kozo Nishida (21:42:36): > @Kozo Nishida has joined the channel

2020-10-23

Daniela Cassol (13:29:18): > @Daniela Cassol has joined the channel

2020-12-12

Huipeng Li (00:38:55): > @Huipeng Li has joined the channel

2021-01-22

Annajiat Alim Rasel (15:45:14): > @Annajiat Alim Rasel has joined the channel

2021-04-12

Kozo Nishida (04:52:07): > Hi@Qian LiuI triedhttps://liubuntu.github.io/Bioc2020RCWL/articles/Bioc2020RCWL.htmlwith > > docker pull liubuntu/bioc2020rcwl:latest > docker run -e PASSWORD=bioconductor -p 8787:8787 -v $(pwd):/home/rstudio/outdir -d --privileged liubuntu/bioc2020rcwl > > and got the following error: > > > library(Rcwl) > > library(RcwlPipelines) > Loading required package: BiocFileCache > Loading required package: dbplyr > > atls <- cwlUpdate() ## sync the tools/pipelines > Update scripts... > trying URL '[https://github.com/hubentu/RcwlRecipes/archive/master.zip](https://github.com/hubentu/RcwlRecipes/archive/master.zip)' > downloaded 316 KB > > Error in cwlProcess(baseCommand = "table_annovar.pl", requirements = list(req1, : > could not find function "cwlProcess" > > Is the cwlProcess function gone now? - File (PNG): image.png

Kozo Nishida (11:58:50): > And the I got the “could not find function”cwlProcess” Error” with the Hello World example too. > If there is something that I should do, please let me know. > > > library(Rcwl) > > inputs <- InputParamList( > + InputParam(id = "sth") > + ) > > echo <- cwlProcess(baseCommand = "echo", inputs) > Error in cwlProcess(baseCommand = "echo", inputs) : > could not find function "cwlProcess" > > echo$sth <- "Hello World!" > Error in echo$sth <- "Hello World!" : object 'echo' not found > > res <- runCWL(echo) > Error in inputs(cwl) : object 'echo' not found >

Qian Liu (12:12:44): > Hi@Kozo Nishida, thank you for raising up this question. This docker image was for Bioc2020 workshop. We have made some significant update to the packages recently, especially due to the manuscript submission to Bioinformatics (which was already accepted and the advance version is available here:https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btab208/6194565).

Qian Liu (12:13:59): > ThecwlProcessis actually a new function which was renamed fromcwlParam.

Qian Liu (12:22:22): > > The error can be fixed by specifying the version 1.6 for now: atls <- cwlUpdate(branch="rcwl1.6") >

Qian Liu (12:22:46): > After the new release in the end of April, all the new features/functions/recipes will be available in the new release.

Kozo Nishida (12:24:55): > Thank you for the information! I understand.

Qian Liu (12:25:00): > If you are interested, you can also run the new workshop material we have prepared for the RLadies-Tunis in March, which includes a real example for scRNA-seq pre-processing.https://github.com/rworkflow/Rcwl_scRNAseq

2021-05-11

Megha Lal (16:45:30): > @Megha Lal has joined the channel

2021-06-04

Izaskun Mallona (08:57:02): > @Izaskun Mallona has joined the channel

2022-01-28

Megha Lal (11:14:19): > @Megha Lal has left the channel

2022-12-23

Nicholas Cooley (09:35:13): > @Nicholas Cooley has joined the channel

2023-08-03

Jill Lundell (10:51:51): > @Jill Lundell has joined the channel

Ritika Giri (15:59:14): > @Ritika Giri has joined the channel

2023-08-04

Ray Su (10:49:40): > @Ray Su has joined the channel

2024-05-14

Lori Shepherd (10:26:10): > archived the channel