Natasha Latysheva

Natasha Latysheva

Auto-generated profile, based on publicly available information

Natasha Latysheva

Google DeepMind, London, UK

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 January 2026 in Nature
Verified
Journal Article
Open access
PDF file
ŽAŽiga Avsec
Žiga Avsec
NLNatasha Latysheva
Natasha Latysheva
JCJun Cheng
Jun Cheng
GNGuido Novati
Guido Novati
KTKyle R. Taylor
Kyle R. Taylor
PKPushmeet Kohli
Pushmeet Kohli
Deep learning models that predict functional genomic measurements from DNA sequences are powerful tools for deciphering the genetic regulatory code. Existing methods involve a trade-off between input sequence length and prediction resolution, thereby limiting their modality scope and performance 1–5 . We present AlphaGenome, a unified DNA sequence model, which takes as input 1 Mb of DNA sequence and predicts thousands of functional genomic tracks up to single-base-pair resolution across diverse modalities. The modalities include gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chromatin contact maps, splice site usage and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest available external models in 25 of 26 evaluations of variant effect prediction. The ability of AlphaGenome to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically relevant variants near the TAL1 oncogene 6 . To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.

27 June 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
ŽAŽiga Avsec
Žiga Avsec
NLNatasha Latysheva
Natasha Latysheva
JCJun Cheng
Jun Cheng
GNGuido Novati
Guido Novati
KTKyle R. Taylor
Kyle R. Taylor
PKPushmeet Kohli
Pushmeet Kohli
Deep learning models that predict functional genomic measurements from DNA sequence are powerful tools for deciphering the genetic regulatory code. Existing methods trade off between input sequence length and prediction resolution, thereby limiting their modality scope and performance. We present AlphaGenome, which takes as input 1 megabase of DNA sequence and predicts thousands of functional genomic tracks up to single base pair resolution across diverse modalities – including gene expression, transcription initiation, chromatin accessibility, histone modifications, transcription factor binding, chro- matin contact maps, splice site usage, and splice junction coordinates and strength. Trained on human and mouse genomes, AlphaGenome matches or exceeds the strongest respective available external models on 24 out of 26 evaluations on variant effect prediction. AlphaGenome’s ability to simultaneously score variant effects across all modalities accurately recapitulates the mechanisms of clinically-relevant variants near the TAL1 oncogene. To facilitate broader use, we provide tools for making genome track and variant effect predictions from sequence.

15 July 2026 in Nature
Verified
Journal Article
Open access
AGAndreas R. Gschwind
Andreas R. Gschwind
KMKristy S. Mualim
Kristy S. Mualim
AKAlireza Karbalayghareh
Alireza Karbalayghareh
MSMaya U. Sheth
Maya U. Sheth
KDKushal K. Dey
Kushal K. Dey
JEJesse M. Engreitz
Jesse M. Engreitz
Identifying transcriptional enhancers and their target genes is essential for understanding gene regulation and the effect of human genetic variation on disease 1–6 . Here we create and evaluate a resource of more than 92 million enhancer–gene regulatory interactions across 1,458 biosamples covering 369 cell types and tissues, by integrating predictive models, chromatin states, three-dimensional contacts and large-scale genetic perturbations generated by the ENCODE Consortium 7 . We first create a systematic benchmarking pipeline to compare predictive models, assembling a dataset of 10,356 element–gene pairs measured in CRISPR perturbation experiments, more than 30,000 fine-mapped expression quantitative trait loci and 569 fine-mapped genome-wide association study (GWAS) variants linked to a probable causal gene. Using this framework, we develop ENCODE-rE2G, a predictive model achieving state-of-the-art performance across several prediction tasks, demonstrating that iterative perturbations and supervised machine learning can build increasingly accurate predictive models of enhancer regulation. Using ENCODE-rE2G, we build an encyclopedia of enhancer–gene regulatory interactions in the human genome, revealing global properties of enhancer networks, identifying differences in regulatory complexity across genes and improving analyses linking noncoding variants to target genes and cell types for common complex diseases. By interpreting the model, we find that beyond enhancer activity and three-dimensional enhancer–promoter contacts, additional features that guide enhancer–promoter communication include promoter class and enhancer–enhancer synergy. These genome-wide maps of enhancer–gene regulatory interactions, benchmarking software, predictive models and insights about enhancer function provide a valuable resource for future studies of gene regulation and human genetics.

20 September 2026 in medRxiv
Verified
Preprint
Open access
PDF file
JCJun Cheng
Jun Cheng
KTKyle R. Taylor
Kyle R. Taylor
LNLauren Nicolaisen
Lauren Nicolaisen
JPJoshua Pan
Joshua Pan
CBClare Bycroft
Clare Bycroft
ŽAŽiga Avsec
Žiga Avsec
A major challenge in genomics is deciphering the functional consequences of non-coding genetic variation. Here we present AlphaGenome Atlas, a comprehensive resource that enables the joint interpretation and prioritization of variant effects across the entire human genome. Using AlphaGenome, we predicted the regulatory effects across thousands of molecular phenotypes for every possible human single nucleotide variant and many observed indels. These predictions were then used to derive a unified and interpretable AlphaGenome Variant Impact (AVI) score and to map cis-regulatory motifs across the genome. AVI achieved state-of-the-art performance across diverse benchmarks with improved prioritization of deleterious non-coding variants. Application of the combined Atlas resource helped solve an epileptic encephalopathy rare disease case, increased the statistical power to detect rare non-coding variants driving population-level phenotypes, and enhanced the mechanistic interpretation of these variants. Thus, AlphaGenome Atlas improves the prioritization and molecular interpretation of non-coding variants with genetic and clinical significance.

17 November 2026 in bioRxiv
Verified
Preprint
Open access
AGAdam Gayoso
Adam Gayoso
AMAnne Mottram
Anne Mottram
AKAlexander Karollus
Alexander Karollus
GNGuido Novati
Guido Novati
PDPavol Drotár
Pavol Drotár
ŽAŽiga Avsec
Žiga Avsec
Most genetic variants associated with human traits and diseases lie in non-coding DNA, where they can alter gene regulation in specific cell types and states. Single-cell atlases chart these states across tissues and development, but rarely profile enough donors to associate variants with expression directly. Deep learning models could bridge this gap by predicting variant effects from DNA sequence, yet they poorly distinguish related cell states and require fine-tuning for new datasets. Here we present AlphaGenome 2 (AG2), which substantially advances cell-type-specific prediction of regulatory activity and variant effects for gene expression and chromatin accessibility. Trained on organism-wide human and mouse single-cell atlases and bulk functional genomics data, AG2 takes as input a 1-Mb DNA sequence and a condition specifying a target tissue, cell type, or cell state. Conditioning enables AG2 to impute unmeasured epigenomic tracks and predict multimodal variant effects across cell types. Inferring this condition directly from gene expression extends AG2 to unseen datasets and cell states without fine-tuning. AG2 outperforms fine-tuned AlphaGenome 1 and specialized models trained on target datasets, doubling recall of causal single-cell eQTLs at 90% precision. AG2 also achieves state-of-the-art enhancer-gene linking and can pinpoint causal variants at complex trait loci. Across hundreds of human cell types, we generated an atlas of enhancer-gene links and scores for 7 million common variants. As single-cell profiling expands across development, tissues, and disease cohorts, AG2 provides a blueprint for turning observational atlases into variant effect maps.