Merge pull request #107 from sanjaynagi/docker-user-config-01-10-24

add docker user option in config for parabricks LSTM use

.gitattributes

@@ -1,2 +1,5 @@
 *.smk linguist-language=Python
 Snakefile linguist-language=Python
+# GitHub syntax highlighting
+pixi.lock linguist-language=YAML
+

.gitignore

@@ -23,4 +23,6 @@ config/samples.tsv
 workflow/scripts/misc
 .test/workflow/scripts/
 .ipynb_checkpoints
-workflow/scripts/snpEff
+workflow/scripts/snpEff# pixi environments
+.pixi
+

.test/config/config_paired_end.yaml

@@ -1,5 +1,6 @@
 metadata: config/samples.tsv
 pipeline: cpu
+parabricks-docker-user: 1006:1006 # only if using parabricks pipeline at LSTM, otherwise ignore
 
 # Dataset name
 dataset: 'Test-GithubActionsCI'
@@ -47,7 +48,7 @@ DifferentialExpression:                       # Activate differential expression
       gene-level:
             activate: True
       isoform-level:
-            activate: False 
+            activate: True 
 
       progressiveGenes:
             activate: True

.test/config/config_single_end.yaml

@@ -1,5 +1,6 @@
 metadata: config/samples.tsv
 pipeline: cpu
+parabricks-docker-user: 1006:1006 # only if using parabricks pipeline at LSTM, otherwise ignore
 
 # Dataset name
 dataset: 'Test-GithubActionsCI'

config/exampleconfig.yaml

@@ -1,6 +1,7 @@
 # RNA-Seq-Pop
-pipeline: parabricks  #parabricks or cpu
 metadata: config/samples.tsv                    # samplesheet metadata file
+pipeline: cpu  # parabricks or cpu
+parabricks-docker-user: 1006:1006 # only if using parabricks pipeline at LSTM, otherwise ignore
 
 dataset: 'Ag_Bouake'                            # Dataset name: Can be anything, will be used to name some main output files
 

workflow/envs/sleuth.yaml

@@ -3,12 +3,11 @@ channels:
   - r
   - bioconda
 dependencies:
-  - "r-base>=4.0.0"
-  - "r-sleuth=0.30"
-  - "r-tidyverse=1.3.1"
-  - "r-data.table=1.14.0"
-  - "r-ggrepel=0.9.1"
-  - "r-openxlsx=4.2.3"
-  - "r-glue=1.4.2"
-  - "r-rcolorbrewer=1.1_2"
-  - "bioconductor-enhancedvolcano=1.8.0"
+  - r-base
+  - "r-sleuth=0.30.1"
+  - r-tidyverse
+  - r-data.table
+  - r-ggrepel
+  - r-openxlsx
+  - r-glue
+  - r-rcolorbrewer

workflow/rules/star-haplotypecaller.smk

@@ -44,10 +44,11 @@ rule fq2bam:
     resources:
         all_gpus = 1
     params:
-        wkdir=wkdir
+        wkdir=wkdir,
+        docker_user = config['parabricks-docker-user']
     shell:
         """
-        docker run --user 1006:1006 --rm --gpus all --volume {wkdir}:/workdir --workdir /workdir nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1 \
+        docker run --user {params.docker_user} --rm --gpus all --volume {wkdir}:/workdir --workdir /workdir nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1 \
             pbrun rna_fq2bam \
             --in-fq /workdir/{input.reads[0]} /workdir/{input.reads[1]} \
             --genome-lib-dir /workdir/resources/reference/star_index\
@@ -75,10 +76,11 @@ rule haplotype_caller:
         all_gpus = 1
     params:
         wkdir=wkdir,
+        docker_user = config['parabricks-docker-user'],
         ploidy=config['VariantAnalysis']['ploidy']
     shell:
         """
-        docker run --user 1006:1006 --rm --gpus all --volume {wkdir}:/workdir -w /workdir nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1 \
+        docker run --user {params.docker_user} --rm --gpus all --volume {wkdir}:/workdir -w /workdir nvcr.io/nvidia/clara/clara-parabricks:4.3.0-1 \
             pbrun haplotypecaller \
             --rna \
             --ref /workdir/{input.ref_fasta} \

workflow/scripts/SleuthIsoformsDE.R

@@ -15,7 +15,6 @@ library(ggrepel)
 library(openxlsx)
 library(glue)
 library(RColorBrewer)
-library(EnhancedVolcano)
 
 print("------------- Kallisto - Sleuth - RNASeq isoform Differential expression ---------")
 #### read data ####
@@ -87,13 +86,6 @@ for (cont in contrasts){
   labels = results %>% dplyr::mutate("Gene_name" = case_when(GeneName == "" ~ TranscriptID,
                                      is.na(GeneName) ~ TranscriptID,
                                      TRUE ~ GeneName)) %>% select(Gene_name) %>% deframe()
-  pdf(glue("results/isoformdiff/Volcano_plot_{cont}.pdf"))
-  print(EnhancedVolcano(results_list[[cont]],
-                  lab=labels,
-                  x='b',
-                  y='pval',
-                  title = cont))
-  garbage = dev.off()
   print(glue("{cont} complete!"))
 }