Tool Duel: DeepVariant (Google) vs Clair3 (Ruibang Luo's lab, HKU). One reads every candidate site as a pileup image; the other runs a fast Bi-LSTM on read counts and re-calls only the lowest-quality 30% with a small ResNet. Details at toolome.bio #Genomics #Bioinformatics
November 8, 2026 at 4:14 PM
Tool Duel: DeepVariant (Google) vs Clair3 (Ruibang Luo's lab, HKU). One reads every candidate site as a pileup image; the other runs a fast Bi-LSTM on read counts and re-calls only the lowest-quality 30% with a small ResNet. Details at toolome.bio #Genomics #Bioinformatics
4/9 Time is the ultimate enemy in critical genetic cases.
Using DeepVariant and DeepConsensus, Stanford achieved a genetic diagnosis via whole genome sequencing in under eight hours.
This unprecedented speed enabled life-saving interventions and secured a Guinness World Record.
Using DeepVariant and DeepConsensus, Stanford achieved a genetic diagnosis via whole genome sequencing in under eight hours.
This unprecedented speed enabled life-saving interventions and secured a Guinness World Record.
October 17, 2026 at 12:10 AM
4/9 Time is the ultimate enemy in critical genetic cases.
Using DeepVariant and DeepConsensus, Stanford achieved a genetic diagnosis via whole genome sequencing in under eight hours.
This unprecedented speed enabled life-saving interventions and secured a Guinness World Record.
Using DeepVariant and DeepConsensus, Stanford achieved a genetic diagnosis via whole genome sequencing in under eight hours.
This unprecedented speed enabled life-saving interventions and secured a Guinness World Record.
🧬 Support genetic diagnoses with ultrarapid genome sequencing in collaboration with @StanfordMed using DeepVariant and DeepConsensus in a clinical trial that set a Guinness World Record.
October 12, 2026 at 9:21 PM
🧬 Support genetic diagnoses with ultrarapid genome sequencing in collaboration with @StanfordMed using DeepVariant and DeepConsensus in a clinical trial that set a Guinness World Record.
Ultima Genomics Partners with Nvidia to Unlock Pangenome-Aware WGS
Ultima Genomics Partners with Nvidia to Unlock Pangenome-Aware WGS
The work will leverage Nvidia's computational power and Google's DeepVariant analysis pipeline.
ebx.sh
September 9, 2026 at 9:29 PM
Ultima Genomics Partners with Nvidia to Unlock Pangenome-Aware WGS
Ultima Genomics and NVIDIA Collaborate to Transform Whole Genome Sequencing with Enhanced Pangenome Capabilities#United_States#Fremont#NVIDIA#Ultima_Genomics#DeepVariant
Ultima Genomics and NVIDIA Collaborate to Transform Whole Genome Sequencing with Enhanced Pangenome Capabilities
Ultima Genomics partners with NVIDIA to enhance whole genome sequencing through pangenome-aware analysis, using Google's DeepVariant. This collaboration aims at large-scale applications.
third-news.com
September 9, 2026 at 12:43 PM
Ultima Genomics and NVIDIA Collaborate to Transform Whole Genome Sequencing with Enhanced Pangenome Capabilities#United_States#Fremont#NVIDIA#Ultima_Genomics#DeepVariant
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
Pangenome-aware DeepVariant
Population-scale genomics information provides valuable prior knowledge for various genomic analyses, especially variant calling. A notable example of such application is the human pangenome reference released by the Human Pangenome Reference Consortium, which has been shown to improve read mapping and structural variant genotyping. In this work, we introduce pangenome-aware DeepVariant, a variant caller that uses a pangenome reference alongside sample-specific read alignments. It generates pileup images of both reads and pangenome haplotypes near potential variants and uses a Convolutional Neural Network to infer genotypes. This approach allows directly using a pangenome for distinguishing true variant signals from sequencing or alignment noise. We assessed its performance on various short-read sequencing platforms and read mappers. Across all settings, pangenome-aware DeepVariant outperformed the linear-reference-based DeepVariant, reducing errors by up to 25.5%. We also show that El
www.biorxiv.org
August 26, 2026 at 1:15 PM
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
There's a lot of debate around whether the Moderna cancer vaccine uses "AI" or not.
AI/ML likely helps at two stages: finding the tumor's mutations (some variant callers use machine learning, like Google's DeepVariant; others use statistical models), and ranking those mutation candidates by
AI/ML likely helps at two stages: finding the tumor's mutations (some variant callers use machine learning, like Google's DeepVariant; others use statistical models), and ranking those mutation candidates by
August 24, 2026 at 5:45 PM
There's a lot of debate around whether the Moderna cancer vaccine uses "AI" or not.
AI/ML likely helps at two stages: finding the tumor's mutations (some variant callers use machine learning, like Google's DeepVariant; others use statistical models), and ranking those mutation candidates by
AI/ML likely helps at two stages: finding the tumor's mutations (some variant callers use machine learning, like Google's DeepVariant; others use statistical models), and ranking those mutation candidates by
Not precision/recall: they convert the calls back to sequence (DeepVariant on HiFi + HapCut2 applied to the reference, low-GQ bases masked), then score that and the assemblies identically with GQC against the T2T-HG002 diploid benchmark. QV 37-40 vs QV 51 for the best assembly.
August 16, 2026 at 6:30 AM
Not precision/recall: they convert the calls back to sequence (DeepVariant on HiFi + HapCut2 applied to the reference, low-GQ bases masked), then score that and the assemblies identically with GQC against the T2T-HG002 diploid benchmark. QV 37-40 vs QV 51 for the best assembly.
DeepTrio calls variants in parent-child trios using joint sequence data, learning inheritance patterns without explicit priors. At 20x coverage it matches DeepVariant at 30x. Models available for Illumina WGS, exome, and PacBio HiFi.
https://www.biorxiv.org/content/10.1101/2021.04.05.438434
https://www.biorxiv.org/content/10.1101/2021.04.05.438434
April 7, 2026 at 3:25 PM
DeepTrio calls variants in parent-child trios using joint sequence data, learning inheritance patterns without explicit priors. At 20x coverage it matches DeepVariant at 30x. Models available for Illumina WGS, exome, and PacBio HiFi.
https://www.biorxiv.org/content/10.1101/2021.04.05.438434
https://www.biorxiv.org/content/10.1101/2021.04.05.438434
DeepTrio employs deep learning to enhance variant calling accuracy in child-mother-father trios, outperforming DeepVariant across various sequencing data types and coverages.
🔗 Source: https://doi.org/10.1101/2021.04.05.438434
🔗 Source: https://doi.org/10.1101/2021.04.05.438434
July 25, 2026 at 6:57 AM
DeepTrio employs deep learning to enhance variant calling accuracy in child-mother-father trios, outperforming DeepVariant across various sequencing data types and coverages.
🔗 Source: https://doi.org/10.1101/2021.04.05.438434
🔗 Source: https://doi.org/10.1101/2021.04.05.438434
Why does GATK exist when bcftools does a perfectly good job, especially given that they both end up feeding DeepVariant or DeepSomatic? Why isn’t everything shoved into Parquet via GA4GH APIs?
You get the idea
You get the idea
November 7, 2025 at 12:45 AM
Why does GATK exist when bcftools does a perfectly good job, especially given that they both end up feeding DeepVariant or DeepSomatic? Why isn’t everything shoved into Parquet via GA4GH APIs?
You get the idea
You get the idea
Pipeline release! nf-core/raredisease v3.1.1 - 3.1.1 - Princess Peach (patch)!
Please see the changelog: https://github.com/nf-core/raredisease/releases/tag/3.1.1
Please see the changelog: https://github.com/nf-core/raredisease/releases/tag/3.1.1
Release 3.1.1 - Princess Peach (patch) · nf-core/raredisease
What's Changed
Release 3.1.1 (Patch deepvariant/rundeepvariant to detect silent failures) by @ramprasadn in #889
Full Changelog: 3.1.0...3.1.1
github.com
June 24, 2026 at 10:37 AM
Pipeline release! nf-core/raredisease v3.1.1 - 3.1.1 - Princess Peach (patch)!
Please see the changelog: https://github.com/nf-core/raredisease/releases/tag/3.1.1
Please see the changelog: https://github.com/nf-core/raredisease/releases/tag/3.1.1
Ok so this #deepvariant caller update from #google is BIG.
github.com/google/deepv...
- 1.6x time reduction (40% increase ??)
- Pangenome integration
- New model and improvements over previously model
Excited to try this new update 🧬💻
#bioinformatics #wgs #ngs #variantcalling
github.com/google/deepv...
- 1.6x time reduction (40% increase ??)
- Pangenome integration
- New model and improvements over previously model
Excited to try this new update 🧬💻
#bioinformatics #wgs #ngs #variantcalling
Release DeepVariant 1.8.0 · google/deepvariant
In this release:
Small model integration: Speed increased by ~1.7x (40% runtime reduction) for WGS, PacBio, and ONT by introduction of additional small model. The small model identifies easy-to-...
github.com
December 9, 2024 at 6:39 AM
Ok so this #deepvariant caller update from #google is BIG.
github.com/google/deepv...
- 1.6x time reduction (40% increase ??)
- Pangenome integration
- New model and improvements over previously model
Excited to try this new update 🧬💻
#bioinformatics #wgs #ngs #variantcalling
github.com/google/deepv...
- 1.6x time reduction (40% increase ??)
- Pangenome integration
- New model and improvements over previously model
Excited to try this new update 🧬💻
#bioinformatics #wgs #ngs #variantcalling
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
May 11, 2026 at 1:15 PM
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
8/ DeepVariant (Google): AI-powered variant calling from NGS data. Uses deep learning for high accuracy.
Open source and integrates with Google Cloud for scalability. github.com/google/deep...
Open source and integrates with Google Cloud for scalability. github.com/google/deep...
GitHub - google/deepvariant: DeepVariant is an analysis pipeline that uses a deep neural network to call genetic variants from next-generation DNA sequencing data.
DeepVariant is an analysis pipeline that uses a deep neural network to call genetic variants from next-generation DNA sequencing data. - google/deepvariant
github.com
February 18, 2026 at 2:15 PM
8/ DeepVariant (Google): AI-powered variant calling from NGS data. Uses deep learning for high accuracy.
Open source and integrates with Google Cloud for scalability. github.com/google/deep...
Open source and integrates with Google Cloud for scalability. github.com/google/deep...
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
October 26, 2025 at 1:45 PM
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
June 10, 2025 at 1:15 PM
1/
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
What reference genome should you use?
Sounds easy. It’s not.
GRCh37? GRCh38? hs37d5?
Have you heard of T2T or the new pan-genome-aware DeepVariant?
This matters more than you think.
www.biorxiv.org/content/10....
The Earth BioGenome Project, launched in 2018, aims to sequence 1.85 million eukaryotic species by 2028 at an estimated US$5 billion, has contributed genomes for 4,386 species, and benefits from AI tools like DeepVariant that improve accuracy.
How AI is sequencing the genomes of all known living species on Earth
An ambitious plan to sequence genomes for 1.85 million eukaryotic species on our planet is underway. It's a massive undertaking that will dramatically enhance our understanding of biology, and inform ...
newatlas.com
November 12, 2025 at 3:47 PM
The Earth BioGenome Project, launched in 2018, aims to sequence 1.85 million eukaryotic species by 2028 at an estimated US$5 billion, has contributed genomes for 4,386 species, and benefits from AI tools like DeepVariant that improve accuracy.
Pangenome-aware DeepVariant www.biorxiv.org/content/10.1... 🧬🖥️🧪 github.com/google/deepv...
Pangenome-aware DeepVariant
Population-scale genomics information provides valuable prior knowledge for various genomic analyses, especially variant calling. A notable example of such application is the human pangenome…
www.biorxiv.org
June 9, 2025 at 6:30 PM
Pangenome-aware DeepVariant www.biorxiv.org/content/10.1... 🧬🖥️🧪 github.com/google/deepv...
Yes, but it turned out to be a really silly mistake. Deepvariant kept saying my reference genome didn't match the alignments. Turns out I pulled the linear reference from the "raw" gfa instead of the gbz that I used with giraffe 🤦
December 25, 2024 at 3:12 PM
Yes, but it turned out to be a really silly mistake. Deepvariant kept saying my reference genome didn't match the alignments. Turns out I pulled the linear reference from the "raw" gfa instead of the gbz that I used with giraffe 🤦
Anyone in my Sky-ome have experience with pangenome-based population genomics? Calling SNPs with Deepvariant is giving me a headache 😭
December 13, 2024 at 11:43 AM
Anyone in my Sky-ome have experience with pangenome-based population genomics? Calling SNPs with Deepvariant is giving me a headache 😭
Pangenome-aware DeepVariant https://www.biorxiv.org/content/10.1101/2025.06.05.657102v1
June 6, 2025 at 10:48 PM
Pangenome-aware DeepVariant https://www.biorxiv.org/content/10.1101/2025.06.05.657102v1
The models from DeepVariant (and friends) have shown how not everything needs to be Illumina reads out there to squeeze all the signal in the data. Another example here for Roche SBX Duplex data, the technology that will become commercially available in 2026.
I'll be speaking in this webinar (go.roche.com/sbx-d) on September 10, where I'll share our benchmarks and observations for Roche's SBX sequencing instrument, as well as models developed by our team for SBX data.
Germline Small Variant Calling Workflow for SBX Duplex Data
Wednesday, September 10, 2025 at 12:00 PM Eastern Daylight Time.
go.roche.com
August 21, 2025 at 8:37 AM
The models from DeepVariant (and friends) have shown how not everything needs to be Illumina reads out there to squeeze all the signal in the data. Another example here for Roche SBX Duplex data, the technology that will become commercially available in 2026.
One hour run of 7-plex genome in a bottle genomes. Duplex efficiency of about 70-80% percent. Variant calling with GATK and Roche ML. DeepVariant slightly better.
February 20, 2025 at 5:28 PM
One hour run of 7-plex genome in a bottle genomes. Duplex efficiency of about 70-80% percent. Variant calling with GATK and Roche ML. DeepVariant slightly better.
Deepsomatic from the creators of Deepvariant github.com/google/deeps...
GitHub - google/deepsomatic
Contribute to google/deepsomatic development by creating an account on GitHub.
github.com
October 24, 2023 at 5:16 PM
Deepsomatic from the creators of Deepvariant github.com/google/deeps...