Merge pull request #156 from broadinstitute/ct-landscape-improvements

Ct landscape improvements

.pylintrc

@@ -0,0 +1,5 @@
+# the default of five is a bit too restrictive
+max-args = 12
+
+# some sci computations require many local variables
+max-locals = 20

assembly.py

@@ -372,7 +372,7 @@ def parser_impute_from_reference(parser=argparse.ArgumentParser()):
 
 def refine_assembly(inFasta, inBam, outFasta,
                     outVcf=None, outBam=None, novo_params='', min_coverage=2,
-                    chr_names=[], keep_all_reads=False, JVMmemory=None, threads=1):
+                    chr_names=None, keep_all_reads=False, JVMmemory=None, threads=1):
     ''' This a refinement step where we take a crude assembly, align
         all reads back to it, and modify the assembly to the majority
         allele at each position based on read pileups.
@@ -383,6 +383,8 @@ def refine_assembly(inFasta, inBam, outFasta,
         and realigned with GATK's IndelRealigner (in order to call indels).
         Output FASTA file is indexed for Picard, Samtools, and Novoalign.
     '''
+    chr_names = chr_names or []
+
     # Get tools
     picard_index = tools.picard.CreateSequenceDictionaryTool()
     picard_mkdup = tools.picard.MarkDuplicatesTool()
@@ -840,7 +842,7 @@ def vcfrow_parse_and_call_snps(vcfrow, samples, min_dp=0, major_cutoff=0.5, min_
                 geno = [allele_depths[0][1]]
             else:
                 # call multiple alleles at this position if there is no clear winner
-                geno = [a for n, a in allele_depths]
+                geno = [a for _, a in allele_depths]
         if geno:
             yield (c, start, stop, sample, geno)