Merge pull request #228 from TheJacksonLaboratory/fix-issue-200

Adds clipping parameter and bumps rMATS version
TheJacksonLaboratory · Jun 29, 2021 · 6ee54c9 · 6ee54c9
2 parents b490b06 + 9c03591
commit 6ee54c9
Show file tree

Hide file tree

Showing 6 changed files with 45 additions and 16 deletions.
diff --git a/containers/rmats4/Dockerfile b/containers/rmats4/Dockerfile
@@ -1,7 +1,15 @@
-FROM nfcore/base:1.9
+# continuumio/miniconda3:4.8.2
+FROM continuumio/miniconda3@sha256:456e3196bf3ffb13fee7c9216db4b18b5e6f4d37090b31df3e0309926e98cfe2
+
 LABEL authors="phil@lifebit.ai laura.urbanski@jax.org" \
-	description="Docker image containing rMATS v4.1.0"
+	description="Docker image containing rMATS v4.1.1"
 
 COPY environment.yml /
-RUN conda env create -f /environment.yml && conda clean -a
+RUN apt-get update && apt-get install -y procps && \
+	conda env create -f /environment.yml && conda clean -a
+
+# Make RUN commands use the new environment:
+RUN echo "conda activate rmats4" >> ~/.bashrc
+SHELL ["/bin/bash", "--login", "-c"]	
+
 ENV PATH /opt/conda/envs/rmats4/bin:$PATH
diff --git a/containers/rmats4/environment.yml b/containers/rmats4/environment.yml
@@ -4,6 +4,6 @@ channels:
   - bioconda
   - defaults
 dependencies:
-  - rmats=4.1.0
+  - rmats=4.1.1
   - bioconductor-bsgenome.hsapiens.ucsc.hg38=1.4.1
   - bioconductor-bsgenome.mmusculus.ucsc.mm10=1.4.0
diff --git a/docs/containers.md b/docs/containers.md
@@ -42,3 +42,10 @@ docker push <registry_user>/<image_name>:<tag>
 gcloud auth login
 docker push gcr.io/<project_id>/<image_name>:<tag>
 ```
+
+
+## Troubleshooting Singularity Images on Sumner
+On Sumner, you may need to remove old singularity images from the cache dir in order to implement updated/new image:
+- Having a cache dir is a Nextflow thing. The idea of this is to save the images to prevent needing to pull the images on each execution which would be really slow 
+- You should only need to clear the cache as when the containers are updated.
+- We ended up setting the cacheDir to cacheDir = "/projects/anczukow-lab/.singularity_cache/"
diff --git a/docs/usage.md b/docs/usage.md
@@ -112,15 +112,15 @@ Main arguments:
                                 (default: false)
   --stranded                    Specifies that the input is stranded ('first-strand', 'second-strand', false (aka unstranded))
                                 (default: 'first-strand')
-  -profile                      Configuration profile to use. Can use multiple (comma separated, string)
-                                Available: base, docker, sumner, test and more.
   --readlength                  Read length - Note that all reads will be cropped to this length(int)
                                 (default: no read length specified)
-                                
+  -profile                      Configuration profile to use. Can use multiple (comma separated, string)
+                                Available: base, docker, sumner, test and more.
+                        
 Trimmomatic: 
   --minlen                      Drop the read if it is below a specified length (int)
 				Default parameters turn on --variable-readlength
-				To crop all reads and turn off, set minlen = readlength                               
+				To crop all reads and turn off --variable-readlength, set minlen = readlength                               
                                 (default: 20)
   --slidingwindow               Perform a sliding window trimming approach (bool)
                                 (default: true)
@@ -138,6 +138,11 @@ Star:
                                 (default: 0.66)
   --sjdbOverhangMin             Controls --alignSJDBoverhangMin (int)
                                 (default: 3)
+  --soft_clipping               Enables soft clipping (bool)
+                                If true, the STAR parameter will be --alignEndsType Local and the rMATS parameter --allow-clipping will be added.
+				If false, the STAR parameter will be --alignEndsType 'EndToEnd' and no rMATS parameter is added.
+				NOTE: Soft Clipping will cause read lengths to be variable, so turn soft_clipping off if reads need to be same length. Variable read length   	                                 parameter is turned on in rMATS when minlen does not equal readlength.
+                                (default: true)
   --star_memory                 Max memory to be used by STAR to sort BAM files.
                                 (default: Available task memory)
 
@@ -169,11 +174,14 @@ Other:
                                 (default: false)
   --outdir                      The output directory where the results will be saved (string)
                                 (default: directory where you submit the job)
-  --gc_disk_size                Only specific to google-cloud executor. Adds disk-space for few aggregative processes.
-                                (default: "200 GB" based on 100 samples. Simply add 2 x Number of Samples)
   --mega_time                   Sets time limit for processes withLabel 'mega_memory' in the main.nf using the base.config (time unit)
                                 Make sure '#SBATCH -t' in 'main.pbs' is appropriately set if you are changing this parameter.
                                 (default: 20.h)
+  --gc_disk_size                Only specific to google-cloud executor. Adds disk-space for few aggregative processes.
+                                (default: "200 GB" based on 100 samples. Simply add 2 x Number of Samples)
+  --debug                       This option will enable echo of script execution into STDOUT with some additional 
+                                resource information (such as machine type, memory, cpu and disk space)
+                                (default: false)
 
 	    
 ```

diff --git a/main.nf b/main.nf
@@ -74,8 +74,10 @@ def helpMessage() {
                                     (default: readlength - 1)
       --filterScore                 Controls --outFilterScoreMinOverLread and outFilterMatchNminOverLread
                                     (default: 0.66)
-      --sjdbOverhangMin              Controls --alignSJDBoverhangMin (int)
+      --sjdbOverhangMin             Controls --alignSJDBoverhangMin (int)
                                     (default: 3)
+      --soft_clipping               Enables soft clipping (bool)
+                                    (default: true)
       --star_memory                 Max memory to be used by STAR to sort BAM files.
                                     (default: Available task memory)
 
@@ -169,6 +171,7 @@ log.info "Single-end                  : ${download_from('tcga') ? 'Will be check
 log.info "GTF                         : ${params.gtf}"
 log.info "STAR index                  : ${star_index}"
 log.info "Stranded                    : ${params.stranded}"
+log.info "Soft_clipping               : ${params.soft_clipping}"
 log.info "rMATS pairs file            : ${params.rmats_pairs ? params.rmats_pairs : 'Not provided'}"
 log.info "Adapter                     : ${download_from('tcga') ? 'Will be set for each sample based based on whether the sample is paired or single-end' : adapter_file}"
 log.info "Read Length                 : ${params.readlength}"
@@ -592,6 +595,7 @@ if (!params.bams){
     out_filter_intron_motifs = params.stranded ? '' : '--outFilterIntronMotifs RemoveNoncanonicalUnannotated'
     out_sam_strand_field = params.stranded ? '' : '--outSAMstrandField intronMotif'
     xs_tag_cmd = params.stranded ? "samtools view -h ${name}.Aligned.sortedByCoord.out.bam | gawk -v strType=2 -f /usr/local/bin/tagXSstrandedData.awk | samtools view -bS - > Aligned.XS.bam && mv Aligned.XS.bam ${name}.Aligned.sortedByCoord.out.bam" : ''
+    endsType = params.soft_clipping ? 'Local' : 'EndToEnd'
     // Set maximum available memory to be used by STAR to sort BAM files
     star_mem = params.star_memory ? params.star_memory : task.memory
     avail_mem_bam_sort = star_mem ? "--limitBAMsortRAM ${star_mem.toBytes() - 2000000000}" : ''
@@ -624,7 +628,7 @@ if (!params.bams){
       --outBAMsortingThreadN $task.cpus \
       --outFilterType BySJout \
       --twopassMode Basic \
-      --alignEndsType EndToEnd \
+      --alignEndsType $endsType \
       --alignIntronMax 1000000 \
       --outReadsUnmapped Fastx \
       --quantMode GeneCounts \
@@ -783,6 +787,7 @@ if (!params.test) {
       statoff = params.statoff ? '--statoff' : ''
       paired_stats = params.paired_stats ? '--paired-stats' : ''
       novelSS = params.novelSS ? '--novelSS' : ''   
+      allow_clipping = params.soft_clipping ? '--allow-clipping' : ''
       if (b1_only) {
         b1_bams = bams.join(",")
         b2_cmd = ''
@@ -808,7 +813,7 @@ if (!params.test) {
         --nthread $task.cpus \
         --readLength ${params.readlength} \
         --mil ${params.mil} \
-        --mel ${params.mel} $variable_read_length_flag $statoff $paired_stats $novelSS
+        --mel ${params.mel} $variable_read_length_flag $statoff $paired_stats $novelSS $allow_clipping
       rmats_config="config_for_rmats_and_postprocessing.txt"
       echo b1 b1.txt > \$rmats_config
       $b2_config_cmd

diff --git a/nextflow.config b/nextflow.config
@@ -33,6 +33,7 @@ params {
     filterScore    = 0.66
     sjdbOverhangMin = 3
     star_memory    = false
+    soft_clipping = true
 
     // rMATS
     statoff        = false
@@ -78,16 +79,16 @@ process {
     container = 'lifebitai/download_reads:latest'
   }
   withName: 'rmats' {
-    container = 'gcr.io/nextflow-250616/rmats:4.1.0'
+    container = 'lifebitai/splicing-rmats:4.1.1'
   }
   withName: 'paired_rmats' {
-    container = 'gcr.io/nextflow-250616/rmats:4.1.0'
+    container = 'lifebitai/splicing-rmats:4.1.1'
   }
   withName: 'collect_tool_versions_env1' {
     container = 'gcr.io/nextflow-250616/splicing-pipelines-nf:gawk'
   }
   withName: 'collect_tool_versions_env2' {
-    container = 'gcr.io/nextflow-250616/rmats:4.1.0'
+    container = 'lifebitai/splicing-rmats:4.1.1'
   }
 }