Easier parsing of output

Author: petercombs

CluToGene.py

@@ -0,0 +1,22 @@
+def get_exon_pairs(gtf='Reference/mel_r5_good.gtf', key='gene_name'):
+    exon_pairs = {}
+    last_type = ''
+    last_stop = -1
+    for line in open(gtf):
+        data = line.split('\t')
+        chrom = data[0]
+        line_type = data[2]
+        start = int(data[3])
+        stop = int(data[4])
+        if last_type == 'exon' and line_type=='exon':
+            annot = {
+                d.strip().split()[0]: d.strip().split(maxsplit=1)[1].strip('";')
+                for d in data[-1].split(';')
+                if d.strip()
+            }
+            exon_pairs[(chrom, last_stop, start)] = annot.get(key)
+        last_stop = stop
+        last_type = line_type
+    return exon_pairs
+
+

VelvetAnt.py

@@ -169,6 +169,7 @@ def load_juncs(juncfile):
     args.samfile.close()
     out['pref_index'] = (out.ALT - out.REF)/(out.ALT + out.REF)
     out.ix[out.ALT + out.REF < args.min_reads, 'pref_index'] = pd.np.nan
+    out.index.name = 'clu'
     out.to_csv(args.outfile, sep='\t', na_rep='N/A')