search

PATRIC3 / patric3_website

Legacy PATRIC Website (JBoss Portal Version)
MIT License
5 stars 2 forks source link

Problem with Full Spades #2173

Open ARWattam opened 5 years ago

ARWattam commented 5 years ago

Job number: 51c1d012-c8ef-4976-ae46-7ab1286c06d1

I submitted a "full spades" assembly with the following P3 command:

p3-submit-genome-assembly --recipe full_spades --srr-id SRR1635127 /jdassembly@patricbrc.org/home/test test_run

In the text output it says #contigs >= 0bp equals 164 In the quast output it says # contigs = 83. The fasta file that is returned had 93 contigs.

Note that only one assembler should have been used. Is one skipping short contigs? I'm surprised nobody has complained about this yet.

AllanDickerman commented 5 years ago

There are issues with cut-off length for reporting contigs. New assembly system should: a) allow user to set this value (but have reasonable default, eg 500bp) b) pass the specified value to programs when needed (eg quast, unicycler) c) perhaps, if found necessary, edit the contigs.fasta file to remove shorter ones d) in short, make the counts of contigs line up and make sense