Hi all,
Could anyone please help me how to install Busco using singularity on HPC? Here they have mentioned documentation with docker (https://busco.ezlab.org/busco_userguide.html#docker-image) and also here with singularity (https://wiki.unil.ch/ci/books/service-de-calcul-haute-performance-%28hpc%29/page/running-busco) but still not sure which command to follow for Busco installation.
Many thanks,
•
written
by
Bioinfonext
That unil.ch link looks a bit too complicated, these are the commands I ran to get BUSCO via Singularity:
Download the image:
singularity pull docker://ezlabgva/busco:v5.2.2_cv1
Run the image with example genome.fasta, assuming it's a genome and a plant:
singularity run -B $(pwd):/busco_wd/ busco_v5.2.2_cv1.sif busco -l viridiplantae -m genome -i genome.fasta --out results
This will store the temporary files in the current working directory, the -v flag in Docker is the -B flag in Singularity, which means to mount (~connect) your current working directory inside the container as /busco_wd/
You will have a new folder in your working directory called busco_downloads which contains the models from, in this case, the viridiplantae dataset.
Dear Philipp, thank you so much. I do have de novo assembled transcriptome data. so not sure if it needs any specific options as well.
with a transcriptome you'd need to replace the -m genome by -m transcriptome and you should be good to go (see https://busco.ezlab.org/busco_userguide.html#transcriptome-mode-assessing-assembled-transcripts )
On a HPC you will need to use singularity. Docker requires root privileges to run which I'll guess you don't have.
Once you have singularity working try the following command:
singularity pull busco.img docker://ezlabgva/busco:v5.2.2_cv1
I know absolutely nothing about BUSCO so I cannot verify the container, but the Dockerhub page looks very well maintained with new version being released regularly.
Log in to answer this question.
Thanks. it is sorted. Busco installed by using following command;
And now running it with this command, is it correct command with these options? I do have plant fruit transcriptome assembly as input fasta file.
./busco_5.1.3.sif busco -i Trinity.fasta -l eukaryota_odb10 --augustus -o busco_output -m transcriptome