Code assumes read specificity

[hbeale]

Per Colette Felton: SpliceDice assumes that the strand of the reads is important, which it often is not when the reads are from a cDNA library that wasn't prepared with strand-specificity. This is hardcoded into the code. This can cause some major issues down the line, since it splits each junction into two junctions (+ and - strand), then essentially randomly assigns reads to each.

[hbeale]

Here are two files containing Colette's modifications; she removed the strand from the junction ID, given that it’s really not important for the downstream analyses she works on.

compareSampleSets.py.txt
MESA.py.txt