Closed jackhump closed 4 years ago
reduce the chance of (3) by reducing maximum intron length - will remove some long-range spurious junctions that connect two genes (probably due to cross-mapping).
hmm, now it's throwing out my junctions because of stranding mismatches. what to do?
does not take strand into account - this is obvious.
should be run downstream of junction filtering to avoid mis-classifying a cluster
if multiple genes overlap junctions in a cluster, which gene is picked as the name?