Closed hoelzer closed 3 years ago
See VirSorter answer here: https://github.com/simroux/VirSorter/issues/68
So we can wait for a fix or simply adjust the length to the actual contig size in case the reported length is too long.
the problem is fixed already in the recent github version of VirSorter: https://github.com/simroux/VirSorter/issues/68
Pending update of container with latest release of virsorter (https://github.com/simroux/VirSorter/releases/tag/v1.0.6)
Maybe helps to build a Docker from the 1.0.6 VirSorter branch: https://github.com/replikation/What_the_Phage/blob/master/phage-tool-Dockerfiles/virsorter/Dockerfile
Or this should also work:
FROM continuumio/miniconda3
ENV VERSION 1.0.6
ENV TOOL virsorter
RUN apt update && apt install -y procps wget gzip pigz bc && apt-get clean && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
RUN conda config --add channels conda-forge && \
conda config --add channels bioconda && \
conda config --add channels default
RUN conda install $TOOL=$VERSION && conda clean -a
based on the bioconda recipe.
Or that container also works for me in the dev branch :)
docker pull mhoelzer/virsorter:1.0.6
Thanks! I got it to work now using 1.0.6.
Closing this ticket :+1:
Test assembly:
kleiner_2015.fasta.gz
Observation
It seems that VirSorter predicts a prophage in a range that is actually larger than the contig size. Example:
So contig
NODE_51
(seq51
after renaming) has a length of 63443 nt.Now VirSorter predicts a prophage for this contig from position 19922-63493:
So the predicted prophage's stop position is larger than the actual contig size. I have the feeling this is a VirSorter problem or maybe a wanted feature. Maybe we should restrict the predicted prophage stop to the length of the contig.
@mberacochea will add some checks and print a message to let us know.