A production-ready suite of tools for single-sample variant filtration

After training on a massive amount of data, the neural network models at the heart of CNNvariant have learned enough to be released from beta.

This suite of tools for single-sample variant filtration uses machine learning to differentiate between good variants and artifacts of the sequencing process, a fairly new approach that is especially effective at correctly calling indels. To enable the models to accurately filter and score variants from VCF files, we trained on validated VCFs (from truth models including SynDip, Genomes in a bottle, and Platinum Genomes) with unvalidated VCFs aligned to different reference builds (HG19, HG38), sequenced on different machines, using different protocols. The result is a single-sample tool that complements, rather than replaces, VQSR (our "traditional" cohort based variant filtering algorithm).

What’s in the CNN toolbox? Pre-trained and train-your-own CNNScoreVariant is a pre-trained Convolutional Neural Network ready to score variants. After scoring, you can filter your VCF by applying a sensitivity threshold with the tool FilterVariantTranches. Or you can train your own models with the tools CNNVariantWriteTensors and CNNVariantTrain, as long as you have validated VCFs to use as training data.

While the current release is only recommended for single sample variant filtration, in the future we hope to explore ways a similar model could be applied to joint-called VCFs.

Under the hood For details on how our deep learning tool works and how it stacks up to other variant filtering tools (both Deep Learning and traditional), as well as a bit of history on Deep Learning in genomics, see our other blog post.

Weichi on 12 Feb 2019

It's great to hear that! Thanks! Just remind you there is an error on [CNNScoreVariants documentation ](https://software.broadinstitute.org/gatk/documentation/tooldocs/current/org_broadinstitute_hellbender_tools_walkers_vqsr_CNNScoreVariants.php "CNNScoreVariants documentation "). The value in -tensor-type should be _read_tensor_ not _read-tensor_. Because I got the below error: >! A USER ERROR has occurred: Argument tensor-type has a bad value: read-tensor. 'read-tensor' is not a valid value for TensorType. Possible values: { >! reference ( 1 Hot encoding of a reference sequence. ) >! read_tensor (Read tensor are 3D tensors spanning aligned reads, sites and channels. The maximum number of reads is a hyper-parameter typically set to 128. There are 15 channels in the read tensor. They correspond to the reference sequence data (4), read sequence data (4), insertions and deletions (2) read flags (4) and mapping quality (1).) >! } Regards Weichi

- Recent posts

- Upcoming events

See Events calendar for full list and dates

- Recent events

See Events calendar for full list and dates

- Follow us on Twitter

GATK Dev Team


We’re hiring! Looking for a software QA engineer to help us make our tools more robust and reliable than ever. Plea… https://t.co/cxelLMws6M
20 Sep 19
RT @curroortuno: Do you want to learn about sequencing data analysis in an amazing city? Register now at @gatk_dev workshop "From reads to…
3 Sep 19
Thank you @murilocervato for hosting our GATK workshop in Sao Paolo last week! Great group of participants, we’ll s… https://t.co/QbT7P5dw5k
3 Sep 19
@RealMattJM “Convoluted”, huh? We see what you did there... https://t.co/GbnpCFWPOX
29 Aug 19
#GATK workshop caption competition: what is deep learning developer Sam Friedman trying to say here? https://t.co/qaJwg3lF1W
28 Aug 19

- Our favorite tweets from others

Do you want to learn about sequencing data analysis in an amazing city? Register now at @gatk_dev workshop "From re… https://t.co/ISBVX2Xwr5
3 Sep 19
Another successful #GATK workshop in the books! @TerraBioApp @gatk_dev https://t.co/oFzPLei0f9
3 Sep 19
Day 2 of #GATK workshop this time in São Paulo, Brazil! Hands-on tutorials using @TerraBioApp #GATK Best Practices… https://t.co/GgtENDNfhk
28 Aug 19
In spite of their stated mission to support human health through genomics, many GATK pipelines are applicable to no… https://t.co/FKQTouZjbv
29 Jul 19
Me: driving myself insane over what data to keep and what to not bother with for thesis and also frantically trying… https://t.co/er2klIcw5i
18 Jul 19

See more of our favorite tweets...