search

maickrau / GraphAligner

MIT License
256 stars 30 forks source link

CCS/HiFi reads #21

Open jdmontenegro opened 4 years ago

jdmontenegro commented 4 years ago

Hi, I have been using this tool for a few months and find it extremely useful. Thank you.

I wanted to know what would be your recommended parameters for mapping CCS reads to a assembly graph produced from raw PacBio reads. I am trying to separate haplotypes from my initial assembly, so it would be very useful to use someting like -asm10 or asm20 as minimap2 to increase alignment stringency.

Cheers,

Juan D. Montenegro

maickrau commented 4 years ago

Hi Juan,

Currently there's no recommended parameters for mapping CCS reads to a large graph. For a small graph (multiple megabases) you can try --seeds-first-full-rows 64 -b 35 which is more accurate than the default parameters, but for a whole genome graph it will be very slow. This is assuming that your assembly graph contigs have a consensus of the PacBio reads. If it has miniasm-style raw read sequences instead of consensus then -x vg will likely work better.

jelber2 commented 1 year ago

Is there an update for this since version 1.0.16?

jelber2 commented 1 year ago

ok, so it was renamed to --seedless-DP