Flybase: FBgn <=> FBtr <=> FBpp IDs (fbgn_fbtr_fbpp_*.tsv)

input_file.txt:

FBpp0070037    
FBpp0070039;FBpp0070040    
FBpp0070041;FBpp0070042;FBpp0070043    
FBpp0070044;FBpp0110571

cat input_file.txt | while read LINE; do echo -en "$LINE\t" >> out_fbpp2fbgn.txt ; fbpp="$(echo $LINE | cut -d";" -f1)" ; grep "$fbpp" fbgn_fbtr_fbpp_fb_2011_10.tsv |  awk ' BEGIN {OFS = FS = "\t"}{print$1}' >>out_fbpp2fbgn.txt; done

out_fbpp2fbgn.txt:

FBpp0070037     FBgn0010215
FBpp0070039;FBpp0070040 FBgn0052230
FBpp0070041;FBpp0070042;FBpp0070043     FBgn0000258
FBpp0070044;FBpp0110571 FBgn0053217

ftp://ftp.flybase.net/releases/current/precomputed_files/genes/fbgn_fbtr_fbpp_fb_2011_10.tsv.gz

Hi,

I'd suggest (see below) to use a python script to do the parsing. The code works accordingly to you have said and it removes duplicates entries IDs in the same line.

Hope this can help you.

from string import strip

for line in open(inFile, "rU"):
  fields = map(strip, line.split())

  ids = map(strip, fields[1].split(";"))
  genes = [singleID for singleID in ids if singleID.startswith("Fbgn")]
  others = set(ids) - set(genes)

  print ("%s\t%s\t%s") % (fields[0], ";".join(sorted(others)), ";".join(sorted(genes)))

You can download the index file that RM suggested. It has three columns where first column is the gene ID and third column is the protein ID. Then use a python script to use the index file to transform your protein ID list. So something like this:

import sys

indexFile = open(sys.argv[1],'r')
index = {}
for line in indexFile:
    if line[0] != "#" and line != "":
        data = line.strip().split('\t')
        if len(data) > 2:
            index[data[2]] = data[0]

pidFile = open(sys.argv[2],'r')

for line in pidFile:
    data = line.strip().split(';')
    output = ''
    for item in data:
        output += index[data[0]] + ';'
    print line.strip() + "\t" + output[:-1]

Save as myScript.py. Use by: python myScript.py indexFile proteinIDFile