Enhance RunVisualizePlots.wdl

.dockstore.yml

@@ -0,0 +1,11 @@
+version: 1.2
+workflows:
+  - subclass: WDL
+    name: RunVisualizePlots
+    primaryDescriptorPath: /wdl/RunVisualizePlots.wdl
+    filters:
+      branches:
+        - main
+        - kj_modifications
+      tags:
+        - /.*/

.gitignore

@@ -0,0 +1,6 @@
+.DS_Store
+*.pyc
+**.egg-info/
+**__py_cache/
+.idea/
+data/

scripts/MakeRDtest.py

@@ -123,19 +123,19 @@ def makeplot(self,pedir,rddir,outdir,flank,build="hg38"):
         #img_v_resize = vconcat_resize([resized_igv,rd]) # combine rd pe and sr together
         # img_v_resize = hconcat_resize([igv,rd])
         img_h_resize = hconcat_resize([ igv, rd ])
-        cv2.imwrite(outdir+self.varname+"_denovo.png", img_h_resize)
+        cv2.imwrite(outdir+self.varname+".png", img_h_resize)
     elif pesrplot!='Error' and rdplot=='Error':
         igv = cv2.imread(pesrplot)
         STR1=self.chr+":"+'{0:,}'.format(int(self.start))+'-'+'{0:,}'.format(int(self.end))+" (hg38)"
         outfile='info.jpg'
         words(STR1,STR2,outfile,50)
-        cv2.imwrite(outdir+self.varname+"_denovo.png", igv)
+        cv2.imwrite(outdir+self.varname+".png", igv)
     elif pesrplot=='Error' and rdplot!='Error':
         rd = cv2.imread(rdplot)
         STR1=self.chr+":"+'{0:,}'.format(int(self.start))+'-'+'{0:,}'.format(int(self.end))+" (hg38)"
         outfile='info.jpg'
         words(STR1,STR2,outfile,50)
-        cv2.imwrite(outdir+self.varname+"_denovo.png", rd)
+        cv2.imwrite(outdir+self.varname+".png", rd)
 
 class VariantInfo():
   def __init__(self,pedfile,prefix):
@@ -196,7 +196,7 @@ def __init__(self,inputfile,pedfile,prefix):
           dat=line.rstrip().split("\t")
           [chr,start,end,name,type,samples]=dat[0:6]
           sample=samples.split(',')[0]
-          varname=samples.split(',')[0]+'_'+name
+          varname=name+'_'+samples.split(',')[0]
           if "," in sample:
             raise Exception("should only have 1 sample per variant")
           prefix=self.variantinfo.getprefix(sample)

scripts/renameCrams.py

@@ -40,7 +40,7 @@ def get_sample_role(row):
         elif (row.IndividualID in mothers.values()):
             role = 'MOTHER'
         else:
-            role = 'UNKNOWN'
+            role = 'PROBAND'
     elif (row.MotherID in mothers.values()) & (row.FatherID in fathers.values()):
         role = 'PROBAND'
     else:

scripts/renameCramsLocalize.py

@@ -43,7 +43,7 @@ def get_sample_role(row):
         elif (row.IndividualID in mothers.values()):
             role = 'MOTHER'
         else:
-            role = 'UNKNOWN'
+            role = 'PROBAND'
     elif (row.MotherID in mothers.values()) & (row.FatherID in fathers.values()):
         role = 'PROBAND'
     else:

wdl/CreateIgvCramPlots.wdl

@@ -1,11 +1,5 @@
 version 1.0
 
-##########################################################################################
-
-## Github commit: talkowski-lab/gatk-sv-v1:<ENTER HASH HERE IN FIRECLOUD>
-
-##########################################################################################
-
 import "IgvCramPlots.wdl" as igv_plots
 import "Structs2.wdl"
 
@@ -30,6 +24,7 @@ workflow IGV_all_samples {
         RuntimeAttr? runtime_attr_run_igv
         RuntimeAttr? runtime_attr_igv
         RuntimeAttr? runtime_attr_cpx
+        RuntimeAttr? runtime_attr_localize_reads
     }
 
     if (defined(fam_ids)) {
@@ -105,7 +100,8 @@ workflow IGV_all_samples {
                     reference_index = reference_index,
                     igv_docker = igv_docker,
                     variant_interpretation_docker = variant_interpretation_docker,
-                    runtime_attr_igv = runtime_attr_igv
+                    runtime_attr_igv = runtime_attr_igv,
+                    runtime_attr_localize_reads = runtime_attr_localize_reads
             }
         }
 
@@ -259,7 +255,7 @@ task update_sample_crai_cram{
                                       mem_gb: base_mem_gb,
                                       disk_gb: ceil(10 + input_size),
                                       cpu: 1,
-                                      preemptible: 2,
+                                      preemptible: 3,
                                       max_retries: 1,
                                       boot_disk_gb: 8
                                   }

wdl/IgvCramPlots.wdl

@@ -1,11 +1,5 @@
 version 1.0
 
-##########################################################################################
-
-## Github commit: talkowski-lab/gatk-sv-v1:<ENTER HASH HERE IN FIRECLOUD>
-
-##########################################################################################
-
 import "Structs2.wdl"
 
 workflow IGV {

wdl/RdVisualization.wdl

@@ -173,6 +173,7 @@ task rdtest {
         cat ~{ped_file} | grep -w -f samples.txt | cut -f1 | sort -u > families.txt
         cat ~{ped_file} | grep -w -f families.txt | cut -f2 | sort -u > all_samples.txt
         fgrep -wf all_samples.txt ~{sample_batches} |awk '{print $2}' |sort -u >existing_batches.txt
+        tr -d '\r' < existing_batches.txt > existing_batches.unix && mv existing_batches.unix existing_batches.txt
         grep -w -f existing_batches.txt ~{batch_bincov} > bincovlist.txt
         #paste ~{sep=" " medianfile} > medianfile.txt