long-reads

[katievigil]

Hi,

I was wondering if Virsorter2 can be used for nanopore long-read sequences that aren't de novo assembled? Some of my fastq files can be de novo assembled to create a fasta file, but some are not and are in fastq.gz file. Can I use VirSorter2 on my fastq.gz unassembled reads?

Thanks! Katie

[jiarong]

Hi, Virsorter2 depends on good gene prediction to assign scores, so it depends on quality of your reads. Raw nanopore reads have high insertion/deletion error that can impact gene prediction. Quality control and read error correction step can help.

[katievigil]

Hi @jiarong I use medaka polishing for read error correction. Is there an example of what the output file looks like for predicting viral families?

thanks

[jiarong]

Hi, here is the output format. Virsorter2 does not assign viral taxonomy tho. For that, I recommend vContact2. Here is tutorial on the whole workflow:
https://maveric-informatics.readthedocs.io/en/latest/Processing-a-Viral-Metagenome.html#identifying-viruses

[katievigil]

Hi @jiarong thank you for your information!