search

cmmr / virmap

GNU Affero General Public License v3.0
25 stars 14 forks source link

unmatched PE reads counts after dereplicate and normalize #15

Closed jingzhejiang closed 4 years ago

jingzhejiang commented 4 years ago

Hi,

I tried to assamble 7 libraries using virmap. Only 3 of them have results after megahit assembly step:

-rw-rw-r-- 1 ubuntu ubuntu 0 Sep 28 04:31 S1/S1.contigs.fa -rw-rw-r-- 1 ubuntu ubuntu 204113 Sep 28 03:29 S2/S2.contigs.fa -rw-rw-r-- 1 ubuntu ubuntu 2733672 Sep 28 04:26 S3/S3.contigs.fa -rw-r--r-- 1 ubuntu ubuntu 341697 Sep 29 21:20 S4/S4.contigs.fa -rw-rw-r-- 1 ubuntu ubuntu 0 Sep 28 04:26 S5/S5.contigs.fa -rw-r--r-- 1 ubuntu ubuntu 0 Sep 29 00:22 S6/S6.contigs.fa -rw-r--r-- 1 ubuntu ubuntu 0 Sep 30 02:18 S7/S7.contigs.fa

The other four libraries reported with a same error (unmatched PE reads counts). I think this maybe because the dereplicate or normalize step filtrated unequal number of reads from raw PE reads. How should I solve this problem?

Thank you!

torptube commented 4 years ago

The normalization and dereplication steps shouldn't affect megahit's inputs. Are your read1 and read2 files balanced going into virmap? Can you post line counts for the input fastqs for S1?

Also can you post the contents of S1.assembly.err?

jingzhejiang commented 4 years ago

Thank you for your reply!! I think you are right! My read1 and read2 files were not balanced going into virmap. Thank you!