Simple python script for generating a firehose-parseable text file from MS-Dos formatted TSV spreadsheets (of the type that we get from project management). Will be deprecated in a few weeks with the advent of direct BSP ID entries, but useful until then.

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@2820 348d0f76-0448-11de-a6fe-93d51630548a

python/getBamFilesFromSpreadsheet.py

@@ -0,0 +1,76 @@
+#!/usr/bin/env python
+
+import sys
+import os
+
+bamfile_base = "/seq/picard_aggregation/"
+fingerprint_base = "/seq/references/reference_genotypes/non-hapmap/"
+hg18_reference = "/seq/references/Homo_sapiens_assembly18/v0/Homo_sapiens_assembly18.fasta"
+hg18_dbsnp = "/humgen/gsa-hpprojects/GATK/data/dbsnp_130_hg18.rod"
+b36_dbsnp = "/humgen/gsa-hpprojects/GATK/data/dbsnp_130_b36.rod"
+b36_reference = "/broad/1KG/reference/human_b36_both.fasta"
+hg18_intervals = "/seq/references/HybSelOligos/whole_exome_agilent_1.1_refseq_plus_3_boosters/whole_exome_agilent_1.1_refseq_plus_3_boosters.targets.interval_list"
+b36_intervals = ""
+
+min_base_q = "10"
+min_map_q = "10"
+max_reads = "1000000"
+min_conf = "0"
+
+spreadsheetPath = sys.argv[3]
+projectName = sys.argv[2]
+groupName = sys.argv[1]
+reference = sys.argv[4]
+
+if ( reference != "hg18" and reference != "b36" ):
+    raise ValueError("Illegal reference type")
+elif ( reference == "hg18" ):
+    reference = hg18_reference
+    dbsnp = hg18_dbsnp
+    intervals = hg18_intervals
+    fpref = "Homo_sapiens_assembly18"
+else:
+    reference = b36_reference
+    dbsnp = b36_dbsnp
+    intervals = b36_intervals
+    fpref = "human_b36"
+
+outputFile = projectName+"_bam_files.txt"
+OUTPUT_HEADER = ["sample_id","recalbrated_bam_file","individual_id","fingerprint_file","reference_file","dbsnp_file","interval_list","max_reads_at_locus","min_confidence","min_mapping_quality","min_base_quality"]
+
+if ( spreadsheetPath.find("/") > -1 ):
+    newSpreadsheet = spreadsheetPath.rsplit("/",1)[1].rsplit(".",1)[0]+"_proper_format.tsv"
+else:
+    newSpreadsheet = spreadsheetPath.rsplit(".",1)[0]+"_proper_format.tsv"
+
+# convert to proper format
+os.system("sed 's/\\r/\\n/g' "+spreadsheetPath+" > "+newSpreadsheet)
+
+project_info = open(newSpreadsheet)
+
+header = project_info.readline().strip().split("\t")
+
+project_index = header.index("Project")
+sample_index = header.index("Sample")
+
+def versionCompare(version1,version2):
+    return -int(version1.split("v")[1])+int(version2.split("v")[1])
+
+def getNewestVersion(baseDir):
+    versions = os.listdir(baseDir)
+    versions.sort(versionCompare)
+    for version in versions:
+        if ( "finished.txt" in os.listdir(baseDir+version+"/") ):
+            return version
+
+outputFile = open(outputFile,'w')
+outputFile.write("\t".join(OUTPUT_HEADER)+"\n")
+
+for line in project_info.readlines():
+    if ( not line.startswith("\n") and not line.startswith(" ") and not line.startswith("\t") ):
+        spline = line.strip().split("\t")
+        versioningDirectory = bamfile_base+spline[project_index]+"/"+spline[sample_index]+"/"
+        version = getNewestVersion(versioningDirectory)
+        bamfile = versioningDirectory+version+"/"+spline[sample_index]+".bam"
+        fingerprint_file = fingerprint_base+spline[project_index]+"/"+fpref+"/"+spline[sample_index]+".fingerprint.geli"
+        outputFile.write(projectName+"_"+spline[sample_index]+"\t"+bamfile+"\t"+groupName+"\t"+fingerprint_file+"\t"+reference+"\t"+dbsnp+"\t"+intervals+"\t"+max_reads+"\t"+min_conf+"\t"+min_map_q+"\t"+min_base_q+"\n")