Closed rrwick closed 4 years ago
Thanks. As I remember, in the miniasm gfa, we only see
L utg000030c + utg000030c + 0M
but not
L utg000030c - utg000030c - 0M
I could be wrong. Haven't looked at the miniasm source code for a long time...
For edges between linear contigs in the miniasm graph, both directions are shown, e.g.:
L utg000013l + utg000036l + 60794M SD:i:46302
L utg000036l - utg000013l - 60576M SD:i:62899
Which is why I included both directions for the circularising links as well. But yes, it is redundant, so only including one direction would be fine too!
I see. Then you change is consistent. Thanks.
This is a small change to the unitig GFA output. Instead of just relying on the last character of the unitig name to indicate circularity (
l
for linear andc
for circular), this change gives circular unitigs an explicit overlap-free link in the GFA file.Before:
After:
This change is mainly to make miniasm assemblies for bacterial genomes (in which circular unitigs are common) display as circular when loaded in Bandage.
Thanks so much for this great tool (and all your others), and let me know if you have any questions!
Ryan