dDocent v2.04 overrides user-defined $simC in paired-end mode ?

[philip.d.anderson1]

Hello dDocent-ers,

On line 549, which concerns paired-end assembly, the cd-hit -c value is hard-coded: cd-hit-est -i uniq.F.fasta -o xxx -c 0.8 -T 0 -M 0 -g 1

Is that intentional, to override the user-defined $simC ?

Best, Philip A

[Jon Puritz]

Hi Philip,

It's not really an override. That line is for the first stage of read clustering. After that step, rainbow will divide clusters with too many variants (potential paralogs), so it makes sense to cluster at a low similarity threshold. The final cd-hit step still takes the simC parameter. I have been thinking of changing this though, so that the first clustering is at a similarity 10% less than the final clustering. 

Thanks for your question,

Jon