The current GATK version is 3.7-0
Examples: Monday, today, last week, Mar 26, 3/26/04

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Get notifications!

You can opt in to receive email notifications, for example when your questions get answered or when there are new announcements, by following the instructions given here.

Got a problem?

1. Search using the upper-right search box, e.g. using the error message.
2. Try the latest version of tools.
3. Include tool and Java versions.
4. Tell us whether you are following GATK Best Practices.
5. Include relevant details, e.g. platform, DNA- or RNA-Seq, WES (+capture kit) or WGS (PCR-free or PCR+), paired- or single-end, read length, expected average coverage, somatic data, etc.
6. For tool errors, include the error stacktrace as well as the exact command.
7. For format issues, include the result of running ValidateSamFile for BAMs or ValidateVariants for VCFs.
8. For weird results, include an illustrative example, e.g. attach IGV screenshots according to Article#5484.
9. For a seeming variant that is uncalled, include results of following Article#1235.

Did we ask for a bug report?

Then follow instructions in Article#1894.

Formatting tip!

Wrap blocks of code, error messages and BAM/VCF snippets--especially content with hashes (#)--with lines with three backticks ( ``` ) each to make a code block as demonstrated here.

Jump to another community
Picard 2.10.2 is now available at
GATK version 4.beta.2 (i.e. the second beta release) is out. See the GATK4 BETA page for download and details.

VariantRecalibration, numBadVariants, and size of the data set

I'm somewhat struggling with the new negative training model in 2.7. Specifically, this paragraph in the FAQ causes me trouble:

Finally, please be advised that while the default recommendation for --numBadVariants is 1000, this value is geared for smaller datasets. This is the number of the worst scoring variants to use when building the model of bad variants. If you have a dataset that's on the large side, you may need to increase this value considerably, especially for SNPs.

And so I keep thinking about how to scale it with my dataset, and I keep wanting to just make it a percentage of the total variants - which is of course the behavior that was removed! In the Version History for 2.7, you say

Because of how relative amounts of good and bad variants tend to scale differently with call set size, we also realized it was a bad idea to have the selection of bad variants be based on a percentage (as it has been until now) and instead switched it to a hard number

Can you comment a little further about how it scales? I'm assuming it's non-linear, and my intuition would be that smaller sets have proportionally more bad variants. Is that what you've seen? Do you have any other observations that could help guide selection of that parameter?

Best Answer


  • Ahh, that would be cool. I'm just now moving my production pipelines to 2.7, maybe I'll hold off for a bit longer. I should be able to scare up a dataset or two I could test with, please do let me know when it appears in the nightlies. Thanks!

Sign In or Register to comment.