Sanger-Sized Pairwise Alignment To Variants To Translation

Question

[Jeremy Leipzig]

I'm looking for some implementations of the following pattern (in any language):

1. One Sanger-sized query sequence is pairwise aligned against a short (<100kbp) reference with some BED/GFF gene calls.
2. Alignment is parsed - sequence substitutions and indels in the non-genic areas are reported as is.
3. Mutations in the genic sections are examined for syn/non-syn substitutions, frameshifts.

I imagine this would be a fairly common script so I'd like to see some existing examples and approaches before reinventing the wheel. "Use Bioperl" is not an acceptable answer.