To everyone who has responded to my query:
I have a lot to think about. Java, Perl, Python, and Tcl all seem to
have a critical mass of advocates within the bioinformatics community.
I am also finding class libraries on the Web in several of these
languages, which may help to carry the water for me on my first
I may be able to stick with my initial, less-than-fully-informed
language choice. The BioJava library has a GenBank file class,
although I can't yet tell how comprehensive it is. You can put a lot
of stuff into a GenBank file, however it's not the SAME stuff in each
file. I'm looking to strip exons out of eukaryotic genome sequence
files, so I'll need access to the information in the CDS field. I can
probably restrain my own inquiries to files that have only one CDS
contained within them -- but a truly robust computer program would be
able to extract multiple CDS's from a single file, if they were
present, and keep track of each gene separately. Whew! Maybe another
Perl looks useful. But its readability appears to be subject to the
whims of the programmer, just like in C. Shortcuts abound. You can
write a single line of code that does everything but the laundry.
Good luck trying to figure out what it means when you go back to read
it a month later!
Python looks very easy. Almost too easy. Is it underpowered? I'm
still trying to find out. There's a BioPython web page... it can't be
Tcl looks reasonable, from what little I've seen so far. Some of the
links I've tried to follow which would describe Tcl bioinformatics
tools are broken. Still searching.
John J. Ladasky Jr., Ph.D.
Department of Biology
Johns Hopkins University
Baltimore MD 21218