CHANGELOG

# 0.3.0

This is the result of the 2023 hackathon. Current features:

1. The [Coordinate submodule](src/isocomp/Coordinates/) creates windows based 
on overlapping transcripts in the concatenated set of input sequences
2. The [Compare submodule](src/isocomp/Compare/) outputs unique transcripts 
based on first whether they are individual transcripts in a given overlap 
bin, next whether they have the exact same start/end points, and finally 
those transcripts which do have the same start/end are pair-wise compared
3. the command line tools function. On a 16 core machine on DNAnexus, runtime 
is ~15 minutes with less than 7GB on 16 CPU

## Caveats

1. The output should be considered an intermediate result. It is unparsed and
not immediately useful to anyone. However, there is good information there

2. We are not conducting exon level coordinate matching on the transcripts. We 
are therefore doing sequence comparison on transcripts which are not actually 
the same (eg, transcripts form the same individual with different exon usage), 
and we are not reporting the wealth of information that we could using the
interval data alone.

## Future directions

1. The Coordinate submodule should create an interval tree structure from the
input gtf files using exon coordinates. Exons should be labelled with the 
transcript and gene IDs
2. The interval tree can then be used to more finely compare intervals and
label different TSS/TTS, exon usage, intron retention, etc
3. The output format(s) must be refined