Gherman Novakovsky (слава Україні! 🇺🇦)

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/)

Models are typically specialized to new domains by finetuning on small, high-quality datasets. We find that repeating the same dataset 10–50× starting from pretraining leads to substantially better downstream performance, in some cases outperforming larger models. 🧵

Great to the see the flurry of single gene knockdown Perturb-seq like atlases from cell-lines, mouse brain etc over the last few days. These are undoubtedly very valuable datasets. I just want to re-iterate a few other very important expt. design considerations 1/

@SashaGusevPosts @javier_maravall Gaps *this* wide have been shown before, in Figure 2D, for splice variant effect prediction (SpliceAI is 700K parameters). The x-axis ranges from 0 to 1 here so it may not be immediately apparent, but its the same 0.6 to 0.9 gap.

Nature research paper: Functional dissection of complex trait variants at single-nucleotide resolution go.nature.com/4ck0zkd

Can we simulate realistic evolutionary trajectories and “replay the tape of life”? In this work, we propose a flexible, generalizable framework for modeling how the entire protein seq evolves over time while capturing complex interactions across sites. 1/n doi.org/10.64898/2026.…

Nature research paper: Regulatory grammar in human promoters uncovered by MPRA-based deep learning go.nature.com/3NUvNEh

AlphaGenome is out in @nature today along with model weights! 🧬 📄 Paper: nature.com/articles/s4158… 💻 Weights: github.com/google-deepmin… Getting here wasn’t a straight path. We sat down @googledeepmind to discuss the story behind the model, paper & API: youtu.be/V8lhUqKqzUc

Molecular Genetics @ University of Toronto 🇨🇦@UofT🦠 is recruiting an Assistant Professor in #Virology!! Come build your dream lab! 🔗jobs.utoronto.ca/job/Toronto-As…

A paper in Nature presents a detailed map of human chromosomes within the nucleus. This resource provides a foundation for an improved understanding of how the physical layout of human DNA is associated with biological expression. go.nature.com/44uMmft

I've been thinking about the "virtual cell" concept and wanted to write up a few thoughts. Specifically on how I think the prior experience in GWAS informs the most likely way these models will be useful. andrewcarroll.github.io/2025/12/23/the…

How does a stem cell "decide" its fate? Development requires both reliability (consistent cell types) AND flexibility (diverse outcomes from identical progenitors). Cells achieve this by dynamically tuning deterministic drift and stochastic diffusion. New in @NatMachIntell: scDiffEq models state-dependent drift AND diffusion, improving fate prediction by ~8% over SOTA. scDiffEq also enables genome-wide in silico perturbation screens and reveals temporal gene dynamics. 🧵nature.com/articles/s4225…

🚀 Our new Science paper is out (w/ B DeMeo, D Burkhardt, A Shalek, M Cortes): science.org/doi/10.1126/sc… We show that active learning + transcriptomic perturbations can guide which exps to run next, boosting phenotypic hit rates >13x. AI not just predicting bio, but designing it.🔁

Systema: a framework for evaluating genetic perturbation response prediction beyond systematic variation go.nature.com/4791kda

Our latest book on the mathematical principles of deep learning and intelligence has been released publicly at: ma-lab-berkeley.github.io/deep-represent… It also comes with a customized Chatbot that helps readers study and a Chinese version translated mainly by AI. This is an open-source project.

A reminder that not every problem benefits from large language models. Low-rank matrix completion or collaborative filtering will likely give you even stronger baselines. nature.com/articles/s4159…

We observed the same across six promoter variant effect benchmarks. Evo2 did worse than basic CNNs and had ~0.5 auROC on some tasks.

Excited to introduce LiftOn – an open-source tool for accurate liftover of genome annotations (GFF) across assemblies. 🚀 👉 Code &community: github.com/Kuanhao-Chao/L… It’s been incredibly rewarding building this for the genomics community. Thank you to all collaborators/friends!

Congratulations Gherman! 🖥️🧬🥳 A tour-de-force of AI/ML on predicting promoter variant effects in humans. 🔗: science.org/doi/10.1126/sc… #MachineLearning #Bioinformatics #Genomics

Huge thanks to the amazing Illumina team—this was an incredible learning experience! I'm excited to keep pushing forward as we develop models to tackle gene expression and non-coding variant interpretation. (16/)

A complementary thread from my colleague Kishore Jaganathan @kjaganatha x.com/kjaganatha/sta… (15/)

Want to learn more about PromoterAI? 📄 Read the paper: science.org/doi/10.1126/sc… 💻 Explore the code & precomputed scores: github.com/Illumina/Promo…. (14/)

In the Genomics England rare disease cohort, functional promoter variants predicted by PromoterAI were enriched in phenotype-matched Mendelian genes. These variants accounted for an estimated 6% of the rare disease genetic burden. (11/)