- get the user-specified sequence file, library file and cut-off percentage
- run fasta34.exe with the sequence file and library file
- open the file that was output by fasta34
- set INPUT_RECORD_SEPARATOR ($/) to ">>"
- reading each record of the fasta34 output file:
 -- the first read will contain just ">>", so skip it
 -- match /(\d+\.\d+)% identity/ and save the captured string as $per
 -- push the record onto an "org" array
 -- if $per > cut-off value, push the record onto an "input" array

# You now have all the original fasta output in @org, and
# all the "above-threshold" records in @input.  At this point,
# you want to run fasta again, but I can't figure out what
# input you want to give it, how many times you really need
# to run it, or how you should use the subsequent output.