Duplicated contigs in CONCOCT bins

[HansCastorp77]

Hi all,
I found something a bit strange. I was using MetaWrap (ver. 1.2.1 due to availability on institute HPC) for metagenome binning from a megahit assembly, maxbin2 and metabat2 produced bins as expected, however CONCOCT produced a set of bins where each contig appears twice in it's respective bin, these bin fastas contain a set of unique contigs followed by a repeat of the exact same contigs in the same order. I haven't seen this happen before and I am at a loss to explain how this could be happening. Has anyone else encountered this?

Note - I used sed -i '/^>/s/ .*//' *.fa to truncate the contig fasta sequence headers in the megahit assembly, as I've previously found Metawrap just returns a single bin using CONCOCT with a megahit assembly - I assumed this was due to the whitespaces in the megahit contigs file sequence headers, which do not occur in say a metaspades assembly file (which seem to work fine with the 1.2.1 MetaWrap implementation of CONCOCT).

Very best,
David

[ZhangMYioz]

I encountered the same situation as you.
When annotating species with GTDB, I was informed ‘FATAL: Sequence identifiers must be unique. Your fasta file contains two sequences with the same id (DG-3-DCD1100328_1)’.
I checked the bins output by dRep using grep -r ‘>DG-3-DCD1100328’ . and found that the problem originates from the concoct output. The entire bin was repeated twice. Only this one bin had such an issue.

[yqy6611]

We have to reset the arguments of CONCOCT manually in the binning script. For example, enlarge the arguments of -c to 1000 or larger.