ctat-VirusIntegrationFinder (ctat-VIF) Release v1.3.0

[Brian Haas]

A new release of CTAT-VirusIntegrationFinder (ctat-VIF) is now available:

https://github.com/broadinstitute/CTAT-VirusIntegrationFinder/releases/tag/ctat-VIF-v1.3.0

Release notes as follows:

• improved sensitivity and overall accuracy for virus integrations, allowing for multimapping reads.
• using STAR single-pass mode
• uses nr virus db, no repeat masking within virus db
• masking K113 in human genome in 2nd phase search.
• maxhits = 50
• virus integrations organized according to virus breakpoint coordinates.
• human genome insertions organized according to flanking sequence similarity
• low complexity insertion flank sequences filtered to further exclude likely false positives

Continues to be compatible with ctat genome lib: https://data.broadinstitute.org/Trinity/CTAT_RESOURCE_LIB/__genome_libs_StarFv1.10/

which uses STAR v2.7.8a as per: https://github.com/STAR-Fusion/STAR-Fusion/wiki/STAR-Fusion-release-and-CTAT-Genome-Lib-Compatibility-Matrix

[Brian Haas]

For previous users of ctat-VIF, it's important to install the new virus database for use with the new version:

https://data.broadinstitute.org/Trinity/CTAT_RESOURCE_LIB/VIRUS_INSERTION_FINDING_LIB_SUPPLEMENT/virus_db.nr.fasta

Remove the earlier CTAT_GENOME_LIB/VIF subdirectory and reinstall according to:
https://github.com/broadinstitute/CTAT-VirusIntegrationFinder/wiki/Human-Virus-Database-Prep

[Robert Allaway]

Hi Brian,

Sorry for commenting on this old thread! I noticed that this release filters out K113. I am curious why this is?

We used a much older version of this tool a couple in mid-2021 and found that all of our samples (33 or so) were positive for this HERV. Recently, trying to reproduce/validate this in some other similar data with the latest release of CTAT-VIF, I was unable to reproduce this, which is how I ended up reading the changelog :)

Is HERV K113 something that you see in most or all samples/do you believe it is a spurious finding?

Cheers,

Robert

[Brian Haas]

Hi Robert,

The latest version should find K113 just fine.  Earlier versions didn't find it as well.  Just be sure to use the latest ctat genome lib and build the ctat-vif supplement / installation part from scratch.

Most samples do find K113 now - I figure many of them are likely normal structural variants.  I've been looking into them as part of a larger study.

If you found K113 before and aren't finding them now, that would be super strange - at least it's the opposite of what I've been experiencing. The latest code should be specifically tuned for improved K113 finding. :-)

Happy to continue this discussion and happy to look at some examples if you'd like me to. 

best,

~brian

--
You received this message because you are subscribed to the Google Groups "Trinity_CTAT_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinity_ctat_us...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/trinity_ctat_users/7f4c066e-2dea-4428-9fca-3aa3ea2b4fd9n%40googlegroups.com.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

[Robert Allaway]

Hi Brian,

Thanks for the info! I misinterpreted what "masking" meant, I think. :)

We were finding K113 before and aren't now, but I'm also just trying to reproduce someone else's code and updating it, so it's probable that I've done something wrong...

I appreciate the offer to look at some examples! I will take another stab at this and a close look at my code and let you know how it goes!

Cheers,

Robert

robert allaway (he/him)

principal scientist

sage bionetworks

seattle, wa

[Robert Allaway]

Hi Brian,

I think I figured out what was going on. In the output directory, I have a folder for the sample ID containing a bunch of empty files - this is what I was looking at previously when I thought I was getting no results. e.g.:

[ec2-user SL10629]$ ls -s SL10629.vif.html
0 SL10629.vif.html

However, if I print the full path, now see they are symlinks...

[ec2-user SL10629]$ ls -sld SL10629.vif.html
0 lrwxrwxrwx 1 root root 137 Oct 21 19:44 SL10629.vif.html ->

/data/VIF/results/SL10629/cromwell-executions/ctat_vif/e90dfdac-511e-4095-a348-96e24d3f28c7/call-SummaryReport/execution/SL10629.vif.html

I took a look at

/data/VIF/results/SL10629/cromwell-executions/ctat_vif/e90dfdac-511e-4095-a348-96e24d3f28c7/call-SummaryReport/execution/SL10629.vif.html and it indeed has some info in it and shows that K113 is detected!

Cheers,

Robert

robert allaway (he/him)

principal scientist

sage bionetworks

seattle, wa

[Brian Haas]

Terrific! Great to hear.

best,

~b

On Fri, Oct 21, 2022 at 3:52 PM Robert Allaway