Tophat vs Tophat-fusion concordant alignment rate

Question

[IP]

Hi everyone!

I am working with RNA-seq paired-end data. When I run tophat-fusion on a dataset, I get mapping percentage of 70-80% for both left and right reads and the concordant pair alignment rate is about 72%. However, if I run Tophat without fusion search, I get the same mapping percentage for left and right reads, but the concodant alignment rate decreases to 30-40%. I am running both commands with the same arguments.

Does anyone know why the concordant pair alignment rate decreases with tophat-fusion? My guess would be that it should increase the concordant alignment rate as it is going to be able to map fusion events that would be unmapped by tophat.

Thank you very much!

Answer 1

[IP]

Tophat fusion removes reads that map to more than two places in the genome, thus, the tophat mapping percentage can be higher because it won't remove multimappers by default