Effective Gene Expression Prediction from Sequence by Integrating Long-range Interactions

Overview

Journal Nat Methods

Specialties Biomedical Engineering
Pathology

Date 2021 Oct 5

PMID 34608324

Citations 340

Authors

Ziga Avsec

Vikram Agarwal

Daniel Visentin

Joseph R Ledsam

Agnieszka Grabska-Barwinska

Kyle R Taylor

Yannis Assael

John Jumper

Pushmeet Kohli

David R Kelley

Affiliations

Soon will be listed here.

Abstract

How noncoding DNA determines gene expression in different cell types is a major unsolved problem, and critical downstream applications in human genetics depend on improved solutions. Here, we report substantially improved gene expression prediction accuracy from DNA sequences through the use of a deep learning architecture, called Enformer, that is able to integrate information from long-range interactions (up to 100 kb away) in the genome. This improvement yielded more accurate variant effect predictions on gene expression for both natural genetic variants and saturation mutagenesis measured by massively parallel reporter assays. Furthermore, Enformer learned to predict enhancer-promoter interactions directly from the DNA sequence competitively with methods that take direct experimental data as input. We expect that these advances will enable more effective fine-mapping of human disease associations and provide a framework to interpret cis-regulatory evolution.

Citing Articles

Foundation models in bioinformatics.

Guo F, Guan R, Li Y, Liu Q, Wang X, Yang C Natl Sci Rev. 2025; 12(4):nwaf028.

PMID: 40078374 PMC: 11900445. DOI: 10.1093/nsr/nwaf028.

Precise engineering of gene expression by editing plasticity.

Qiu Y, Liu L, Yan J, Xiang X, Wang S, Luo Y Genome Biol. 2025; 26(1):51.

PMID: 40065399 PMC: 11892124. DOI: 10.1186/s13059-025-03516-7.

Integration of proteomics profiling data to facilitate discovery of cancer neoantigens: a survey.

Luo S, Peng H, Shi Y, Cai J, Zhang S, Shao N Brief Bioinform. 2025; 26(2).

PMID: 40052441 PMC: 11886573. DOI: 10.1093/bib/bbaf087.

iModEst: disentangling -omic impacts on gene expression variation across genes and tissues.

Sokolowski D, Mai M, Verma A, Morgenshtern G, Subasri V, Naveed H NAR Genom Bioinform. 2025; 7(1):lqaf011.

PMID: 40041206 PMC: 11879402. DOI: 10.1093/nargab/lqaf011.

Enhancer reprogramming: critical roles in cancer and promising therapeutic strategies.

Yang J, Zhou F, Luo X, Fang Y, Wang X, Liu X Cell Death Discov. 2025; 11(1):84.

PMID: 40032852 PMC: 11876437. DOI: 10.1038/s41420-025-02366-3.

References

Agarwal V, Shendure J . Predicting mRNA Abundance Directly from Genomic Sequence Using Deep Convolutional Neural Networks. Cell Rep. 2020; 31(7):107663. DOI: 10.1016/j.celrep.2020.107663. View

Richter F, Morton S, Kim S, Kitaygorodsky A, Wasson L, Chen K . Genomic analyses implicate noncoding de novo variants in congenital heart disease. Nat Genet. 2020; 52(8):769-777. PMC: 7415662. DOI: 10.1038/s41588-020-0652-z. View

Alipanahi B, Delong A, Weirauch M, Frey B . Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat Biotechnol. 2015; 33(8):831-8. DOI: 10.1038/nbt.3300. View

Fudenberg G, Kelley D, Pollard K . Predicting 3D genome folding from DNA sequence with Akita. Nat Methods. 2020; 17(11):1111-1117. PMC: 8211359. DOI: 10.1038/s41592-020-0958-x. View

Gasperini M, Tome J, Shendure J . Towards a comprehensive catalogue of validated and target-linked human enhancers. Nat Rev Genet. 2020; 21(5):292-310. PMC: 7845138. DOI: 10.1038/s41576-019-0209-0. View

Ghandi M, Mohammad-Noori M, Ghareghani N, Lee D, Garraway L, Beer M . gkmSVM: an R package for gapped-kmer SVM. Bioinformatics. 2016; 32(14):2205-7. PMC: 4937197. DOI: 10.1093/bioinformatics/btw203. View

Moore J, Purcaro M, Pratt H, Epstein C, Shoresh N, Adrian J . Expanded encyclopaedias of DNA elements in the human and mouse genomes. Nature. 2020; 583(7818):699-710. PMC: 7410828. DOI: 10.1038/s41586-020-2493-4. View

Ji Y, Zhou Z, Liu H, Davuluri R . DNABERT: pre-trained Bidirectional Encoder Representations from Transformers model for DNA-language in genome. Bioinformatics. 2021; 37(15):2112-2120. PMC: 11025658. DOI: 10.1093/bioinformatics/btab083. View

Villar D, Berthelot C, Aldridge S, Rayner T, Lukk M, Pignatelli M . Enhancer evolution across 20 mammalian species. Cell. 2015; 160(3):554-66. PMC: 4313353. DOI: 10.1016/j.cell.2015.01.006. View

10.

Fulco C, Nasser J, Jones T, Munson G, Bergman D, Subramanian V . Activity-by-contact model of enhancer-promoter regulation from thousands of CRISPR perturbations. Nat Genet. 2019; 51(12):1664-1669. PMC: 6886585. DOI: 10.1038/s41588-019-0538-0. View

11.

Wang Q, Kelley D, Ulirsch J, Kanai M, Sadhuka S, Cui R . Leveraging supervised learning for functionally informed fine-mapping of cis-eQTLs identifies an additional 20,913 putative causal eQTLs. Nat Commun. 2021; 12(1):3394. PMC: 8184741. DOI: 10.1038/s41467-021-23134-8. View

12.

Kircher M, Xiong C, Martin B, Schubach M, Inoue F, Bell R . Saturation mutagenesis of twenty disease-associated regulatory elements at single base-pair resolution. Nat Commun. 2019; 10(1):3583. PMC: 6687891. DOI: 10.1038/s41467-019-11526-w. View

13.

Forrest A, Kawaji H, Rehli M, Baillie J, de Hoon M, Haberle V . A promoter-level mammalian expression atlas. Nature. 2014; 507(7493):462-70. PMC: 4529748. DOI: 10.1038/nature13182. View

14.

Nair S, Kim D, Perricone J, Kundaje A . Integrating regulatory DNA sequence and gene expression to predict genome-wide chromatin accessibility across cellular contexts. Bioinformatics. 2019; 35(14):i108-i116. PMC: 6612838. DOI: 10.1093/bioinformatics/btz352. View

15.

Schreiber J, Durham T, Bilmes J, Noble W . Avocado: a multi-scale deep tensor factorization method learns a latent representation of the human epigenome. Genome Biol. 2020; 21(1):81. PMC: 7104480. DOI: 10.1186/s13059-020-01977-6. View

16.

Schwessinger R, Gosden M, Downes D, Brown R, Oudelaar A, Telenius J . DeepC: predicting 3D genome folding using megabase-scale transfer learning. Nat Methods. 2020; 17(11):1118-1124. PMC: 7610627. DOI: 10.1038/s41592-020-0960-3. View

17.

Reshef Y, Finucane H, Kelley D, Gusev A, Kotliar D, Ulirsch J . Detecting genome-wide directional effects of transcription factor binding on polygenic disease risk. Nat Genet. 2018; 50(10):1483-1493. PMC: 6202062. DOI: 10.1038/s41588-018-0196-7. View

18.

Heinz S, Romanoski C, Benner C, Glass C . The selection and function of cell type-specific enhancers. Nat Rev Mol Cell Biol. 2015; 16(3):144-54. PMC: 4517609. DOI: 10.1038/nrm3949. View

19.

. The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science. 2020; 369(6509):1318-1330. PMC: 7737656. DOI: 10.1126/science.aaz1776. View

20.

Avsec Z, Weilert M, Shrikumar A, Krueger S, Alexandari A, Dalal K . Base-resolution models of transcription-factor binding reveal soft motif syntax. Nat Genet. 2021; 53(3):354-366. PMC: 8812996. DOI: 10.1038/s41588-021-00782-6. View