improve error message of computeGCbias when a DNA sequence containing NNNs is found

Author: fidelram

bin/computeGCBias

@@ -169,6 +169,7 @@ def getRequiredArgs():
 
     return parser
 
+
 def getPositionsToSample(chrom, start, end, stepSize):
     """
     check if the region submitted to the worker
@@ -257,6 +258,7 @@ def countReadsPerGC_worker(chromNameBam,
             gc = getGC_content(tbit[chromNameBit].get(i, i + regionSize))
         except Exception as detail:
             if verbose:
+                print "{}:{}-{}".format(chromNameBit, i, i + regionSize)
                 print detail
             continue
         numberReads = bam.count(chromNameBam, i, i + regionSize)

deeptools/utilities.py

@@ -8,7 +8,7 @@ def getGC_content(dnaString, as_fraction=True):
     if len(dnaString) == 0:
         return None
     if dnaString.count('N') > len(dnaString) * 0.05:
-        raise Exception("too many NNNs in assembly sequence")
+        raise Exception("WARNING: too many NNNs present in sequence of length {}".format(len(dnaString)))
         return None
 
     gc = 0