two short(1000bp) contigs are generated by YaHS

[spoonbender76]

Hi,

Thank you for creating and maintaining YaHS. I installed YaHS 1.2a.1 by mamba install yahs -c bioconda.

~100x Hi-C data was mapped to assembly after purge_dups by Chromap following the pipeline described here. https://github.com/WarrenLab/hic-scaffolding-nf/blob/main/main.nf

I tried YaHs with or without -e GATC and left other options default. YaHS generated two short 1000 bp contigs, which are much shorter than minimum contig length from hifiasm p_ctg (15508 bp) and after purge_dups (31739 bp).

With --no-contig-ec YaHS did not generate 1000 bp contigs, but this version contains more errors and is not ideal compared to the contig error correction version.

These two 1000 bp contigs cannot be visualized or manually curated in juicebox. Can you give some advice on how to handle them or prevent YaHS from doing this?

[c-zhou]

Hello @spoonbender76,

These small pieces are debris generated during correcting the contigs. It is unavoidable if you want to do assembly error correction. They are very likely junk sequences. You can either leave them there in your assembly or simply remove them.

Best,
Chenxi