Hi, I am a beginner myself so I do not feel confident enough to comment on your code however I think you should look into Bioperl. There is a very simple way to read a GenBank file into perl (basically a parser). Once you have read it into perl there should be another function in Bioperl to easily extract the coding sequences; I do not know about 5 and 3 UTR though. I had to sort a similar issue last week where I had to extract the intergenic regions of a bacteria (but with a different twist, could not use already available methods for a variety of reasons); if I had to extract the CDS I could have easily used Bioperl, but after numerous attempt I just wrote my own Genbank parser and I extracted what I needed to extract them from a Fasta myself. Perl regex are very powerful; but again, for your case, there should be something on Bioperl to extract transcripts. Check this out : http://bioperl.org/howtos/Features_and_Annotations_HOWTO.html