search

simroux / VirSorter

Source code of the VirSorter tool, also available as an App on CyVerse/iVirus (https://de.iplantcollaborative.org/de/)
GNU General Public License v2.0
104 stars 30 forks source link

Detection of eukaryotic viruses? #69

Open rprops opened 4 years ago

rprops commented 4 years ago

Is there any information on the performance/ability of VirSorter to detect signatures of eukaryotic viruses in environmental datasets? I've been playing a bit with the software but the output appears to be restricted to phage detection. In case not, any recommendations for a software that would be enable this? Thanks, Ruben

simroux commented 4 years ago

Hi Ruben, Eukaryotic viruses are not well handled by VirSorter, they will typically show up in Categories 3 or 6 because of low proportion of genes with a hit in PFAM (that's what happens for instance for a lot of giant viruses and phycodnaviridae), or will not be detected at all. For automatic detection of eukaryotic viruses in environmental datasets, this kind of depends on the type of eukaryotic viruses you are after, but typically I'd recommend working with marker genes of the viral group you are interested in. For instance, RdRP is a great marker gene for eukaryotic RNA viruses (and RNA phages as well). If you are interested in more of the larger viruses (Giant viruses, Phycodnaviridae), you can look into some recent paper including ours (https://www.nature.com/articles/s41586-020-1957-x) and other (https://www.nature.com/articles/s41467-020-15507-2), but it's typically (i) detecting candidate contigs / bins based on HMM of marker genes, then (ii) clean up these candidates by removing contigs / bins that are microbial. Unfortunately, I don't know of a software that would do this automatically.

Hope that helps ! Best, Simon

rprops commented 4 years ago

Hi Simon, This helps a lot - thanks for providing context and advice! Much appreciated! Ruben