@NovakovskyG

PhD, Illumina AI lab; interested in Deep Learning and genome regulation; also drawing, martial arts, guitar, and death metal! (he/him)

Joined January 2018
  • Tweets799
  • Following404
  • Followers276
23Media
Excited to share my first contribution here at Illumina! We developed PromoterAI, a deep neural network that accurately identifies non-coding promoter variants that disrupt gene expression.🧵 (1/)
2
30
6
109
33,749
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Kyle Farh MD PhD - Illumina AI Lab
@KyleFarhOct 20
Excited to be introducing Illumina's dnaAI model for the opening talk at ASHG 2026! #ASHG26 #ASHG2026
1
3
25
1,104
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Anshul Kundaje@anshulkundajeOct 20
Cell-state conditioned sequence model of chromatin & expression trained jointly across diverse cell contexts by @zhou_jingtian & colleagues. Nice addition to the growing collection of joint cis-trans models of gene regulation. Still lots to do make the trans component causal.
Arc Institute
@arcinstituteOct 20
Our cells carry the same DNA, but how this DNA is interpreted differs wildly from cell to cell. In a new preprint, Science Fellow @zhou_jingtian and team introduce Bolero, a model conditioned on cell state that predicts chromatin accessibility and expression from DNA alone.
1
10
90
5,634
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Anshul Kundaje@anshulkundajeOct 15
This work is not getting enough attention. These are pretty massive leaps in modeling splicing. Very excited about this. Need to do a deep read but the performance on diverse splicing related tasks looks really good.
Kishore Jaganathan@kjaganathaOct 15
#SpliceAI2 splice variant effect benchmarks. Considerably improved performance in 9 evaluations spanning cryptic splicing, sQTLs, MPRAs, population genetics, plasma proteins, and rare disease. 🧬✂️🧵
2
19
134
9,097
Take a deeper dive into SpliceAI2 performance!
Kishore Jaganathan@kjaganathaOct 15
#SpliceAI2 splice variant effect benchmarks. Considerably improved performance in 9 evaluations spanning cryptic splicing, sQTLs, MPRAs, population genetics, plasma proteins, and rare disease. 🧬✂️🧵
4
473
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Xiangmeng(Shawn) Cai@XiangmengCaiOct 9
Have you ever looked at a gene in IGV or a genome browser and wondered “What are all those (ATAC-seq/H3K27ac/CTCF ChIP-seq/…) peaks doing?” 🤔 I’m thrilled to share that we developed a new technology called Swap-seq to answer that exact question. biorxiv.org/content/10.64898…

Systematic mapping of enhancers, silencers, and topological elements with prime editing deletion...

Noncoding regulatory elements in the human genome can control gene expression through distinct mechanisms, including acting as enhancers, silencers, or topological elements. However, we lack tools to...

biorxiv.org
2
65
4
196
23,755
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Jeff Vierstra@jeffvierstraOct 8
Just put the finishing touches on this beast. 17yrs to get here and the largest, most information dense dataset I had the privilege to generate and analyze. Look for it on bioRxiv next week!
1
4
82
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
David Kelley@drkllyOct 9
This week we're releasing a preprint exploring state space models for long-sequence genomic prediction, alongside Cerberus, a set of open PyTorch models built on this work. biorxiv.org/content/10.64898…

Cerberus: bidirectional state space blocks improve accuracy and efficiency of regulatory sequence...

Sequence-to-function models predict regulatory activity and variant effects from DNA sequence alone, and leading models such as Borzoi compute long-range interactions with self-attention. The...

biorxiv.org
4
47
5
161
25,494
Check out SpliceAI2, a SOTA deep learning model designed to understand and predict RNA splicing!
Kishore Jaganathan@kjaganathaOct 8
Incredibly, incredibly excited to introduce SpliceAI2, the next generation of splicing prediction! 🧬🚀🧵
2
10
825
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Kyle Farh MD PhD - Illumina AI Lab
@KyleFarhOct 9
Excited to announce SpliceAI2: illumina.com/science/genomic… Open model weights: huggingface.co/illumina-ai/S… Source code: github.com/illumina/spliceAI…

Introducing SpliceAI2: The Next Generation of Splicing and Transcript Isoform Prediction

illumina.com
9
19
609
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Illumina
@illuminaOct 8
Illumina's genomic AI models now enable researchers to identify 2x as many variants with predicted biological impact. Identifying these variants and understanding their downstream effects can unlock new biological insights to advance rare disease and drug discovery research. The newest generation of our leading-edge genomic AI models are designed to help researchers assess the functional impact of splice, promoter, and missense variants. With SpliceAI2, out today, we aim to help researchers advance rare disease research. In the words of the vice president of BioInsight AI Lab, Kyle Farh: “Illumina is advancing AI to systematically shrink the portion of the genome that remains uninterpretable.” 🔗 Learn more: assets.illumina.com/content/… For Research Use Only. Not for use in diagnostic procedures.
1
14
1
26
3,327
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Jianbin Chen@Bright_88Oct 8
Super excited to share SpliceAI2! ✂️🧬 It predicts splice-site and splice-junction usage, reconstructs complete transcripts, and models tissue- and disease-specific splicing. Proud to have contributed, and huge credit to the incredible team behind it! Check out the tweetorial!⬇️
Kishore Jaganathan@kjaganathaOct 8
Incredibly, incredibly excited to introduce SpliceAI2, the next generation of splicing prediction! 🧬🚀🧵
1
4
7
270
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Kishore Jaganathan@kjaganathaOct 8
Incredibly, incredibly excited to introduce SpliceAI2, the next generation of splicing prediction! 🧬🚀🧵
7
36
4
99
9,109
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Han Yuan@HY3952Oct 5
Saturation mutagenesis of long-context seq2fun models requires millions of evaluations per gene. We introduce Multi-ISM to make this mapping routine: biorxiv.org/content/10.64898…

Scalable saturation mutagenesis reveals gene regulatory architecture and rare variant effects

Long-context sequence-to-function models enable nucleotide-resolution prediction of regulatory variant effects, motivating comprehensive mutational interrogation across broad genomic contexts. Yet...

biorxiv.org
1
30
2
94
9,952
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
antisense.@razoralignSep 18
scBaseCount: An AI agent-curated, standardized, auto-updated single-cell data repository cell.com/cell/fulltext/S0092…
1
17
57
3,904
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Philippe Batut@philippe_batutSep 11
🧬🧪🔬Why are enhancers transcribed? How does it impact gene regulation? I’m excited to share our new paper in @ScienceMagazine showing that ncRNAs control the timing of gene activation in embryos. w/Mike Levine #ScienceResearch #RNA A few highlights 🧵👇 science.org/doi/10.1126/scie…

Noncoding transcription controls the developmental dynamics of long-range gene regulation

The genomic regions regulating gene expression are often themselves transcribed into a variety of noncoding RNAs (ncRNAs). However, the regulatory roles of this noncoding transcription remain largely...

science.org
4
65
6
257
20,437
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Shiron Drusinsky@ShirondrusinskySep 11
Excited to announce a new preprint that investigates the ability of sequence-to-function models (S2F; e.g., AlphaGenome) to identify causal expression modifying variants. (1\n) biorxiv.org/content/10.64898…

Causal variant underestimation is a major overlooked driver of sequence-to-function model underpe...

Deep learning sequence-to-function (S2F) models represent tremendous promise for functionally fine-mapping causal variants associated with traits and disease. Yet it remains unclear precisely how...

biorxiv.org
1
35
2
146
16,945
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Surag Nair@suragnairAug 5
I can not emphasize enough how much love and care has gone into this resource. Each of the 3865 models has been individually optimized and thoroughly vetted. It’s a truly foundational resource, and holds many many stories that are begging to be told.
Anshul Kundaje@anshulkundajeAug 5
Very excited to announce ENCODE GRAMMAR (Genomic Regulatory Atlas of sequence Models, Motifs, Annotations & Rules): 3,865 experiment-specific deep learning model sets and sequence annotations for decoding human regulatory DNA. 1/
1
24
130
13,857
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Peter Koo@pkoo562Aug 3
[Reviewer Request] I need a few emergency reviewers for MLCB 2026. Please let me know if you have capacity to help over the next day or two. DM me with your contact email! Please RT!
13
29
3
45
15,625
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
Jay Shendure@JShendureJul 30
New preprint from Shendure Lab on Locus-Scale aka Long-@$$ MPRAs led by the amazing Abby McGee & @CGBiar! Most MPRAs test ~300 bp fragments next to a promoter. But real enhancers are bigger, act combinatorially and from a distance. 1/n biorxiv.org/content/10.64898…
5
55
3
177
25,495
Gherman Novakovsky (слава Україні! 🇺🇦) retweeted
David Garfield (@[email protected] and post)@dagarfieldJul 21
We did a thing! A great collaboration from my @genentech days biorxiv.org/content/10.64898…

Deep learning design and in vivo validation of Müller glia-specific cis-regulatory elements

Effective recombinant adeno-associated virus gene therapies require promoters that are compact and cell-type-specific. Ideally, promoters should also exhibit functional conservation when tested in...

biorxiv.org
3
10
1
59
6,667
Load more