0 bins(0 bases in total) forms

Issue #79 resolved
lu kang created an issue

hi, I met some probelms when using runMetaBat.sh, I have assembly a genome,and using bwa to align it on its closest genome.

the assembly quanlity is good, so I’m very confused that my output is 0 bins(0 bases in total) forms.

My script and results:

the depth:

Thank you!

Comments (6)

  1. Feng Li

    looks like your depth file wasn’t right.

    only 2 contigs were recognized, one of which is 4e6 bps long.

    you should check your input files, remove the 'cdc5_depth.txt' file and try runMetaBat.sh again.

  2. Feng Li

    sorry i don’t get what you want to say.

    what i see is a contig that makes no sense of too long bps.

    it may be caused by a wrong input file.

  3. Rob Egan

    There are a few things going on.

    1) there are only 2 contigs in the depth file with depths of 0.9 and 0.0 respectively… insufficient to assist in binning.

    2) You are running an old version (2.12) that may contain a bug whereby a bin with a single contig is not reported.

    Please use the latest version 2.14

    And this looks like it is an assembly from long reads (PacBio or Nanopore), you will need to adjust the minimum %ID that metabat uses to filter out poorly mapping reads. You can set the environmental variable PCTID=85, and see if that gives you better results. However MetaBAT is NOT designed or tested for anything except short-read assemblies and data sets (i.e. Illumina), and furthermore we do not expect it work very well for long read assemblies because of the reliance on abundances to distinguish strains and species but the error rates of the underlying reads are far higher than strain/species variations in their genomes.

  4. Log in to comment