ruanjue / wtdbg2

Redbean: A fuzzy Bruijn graph approach to long noisy reads assembly
GNU General Public License v3.0
512 stars 94 forks source link

-edge-min 1: does it report all edges? or just best-overlap edges? #173

Closed chocotwig closed 4 years ago

chocotwig commented 4 years ago

Hi, if wtdbg2 is attempting to untangle a repeat as in the attached picture, where there is no single long read that spans the region from any node on the left to any node on the right, would wtdbg2 simply give up on assembly? (edge-min 1 settings) As the picture seeks to show, read 2 overlaps most with Read 1 and read 3, and less so with reads 4 and 5. Would wtdbg2 report all possible contigs, i.e. 1-2-3, 1-2-5, 4-2-3 and 5-2-5? Or would it only select the path with the maximum overlap (i.e. 1-2-3?) image

Thanks!

ruanjue commented 4 years ago

Hi, wtdbg2 is conservative in untangling repeats, it won't attempt to solve such case. When only there are enough reads spaning the repetitive region, wtdbg2 try to untangle it.