scriptlet to create sample meta table entries for an ensembl core database

random_sample_chooser.py

#!/usr/bin/env python
# random_sample_chooser.py <GFF3>
import re
import sys
import random

gff=open(sys.argv[1],"r")

mRNAids = {}

for line in gff:
  if not re.search('#',line):
                
    columns = line.split("\t")
 
    if columns[2]=='mRNA':
      ifmrna=re.search("Name=(.*?)(;|$)",line)
      mrna=ifmrna.group(1)

      findgene=re.search("Parent=(.*?);",line)
      geneid=findgene.group(1)

      #look for the seq and position of the transcript
      region=columns[0] + ':' + columns[3] + '-' + columns[4]

      mRNAids[mrna]=[geneid,region]


sample = random.choice(mRNAids.keys())

print "meta.sample.gene_param: " + mRNAids[sample][0]
print "meta.sample.gene_text: " + mRNAids[sample][0]
print "meta.sample.location_param: " + mRNAids[sample][1]
print "meta.sample.location_text: " + mRNAids[sample][1]
print "meta.sample.search_text: ribosomal"
print "meta.sample.transcript_param: " + sample
print "meta.sample.transcript_text: " + sample