Evaluating the Informativeness of Deep Learning Annotations for Human Complex Diseases

Overview

Journal Nat Commun

Specialty Biology

Date 2020 Sep 18

PMID 32943643

Citations 19

Authors

Kushal K Dey

Bryce van de Geijn

Samuel Sungil Kim

Farhad Hormozdiari

David R Kelley

Alkes L Price

Affiliations

Soon will be listed here.

Abstract

Deep learning models have shown great promise in predicting regulatory effects from DNA sequence, but their informativeness for human complex diseases is not fully understood. Here, we evaluate genome-wide SNP annotations from two previous deep learning models, DeepSEA and Basenji, by applying stratified LD score regression to 41 diseases and traits (average N = 320K), conditioning on a broad set of coding, conserved and regulatory annotations. We aggregated annotations across all (respectively blood or brain) tissues/cell-types in meta-analyses across all (respectively 11 blood or 8 brain) traits. The annotations were highly enriched for disease heritability, but produced only limited conditionally significant results: non-tissue-specific and brain-specific Basenji-H3K4me3 for all traits and brain traits respectively. We conclude that deep learning models have yet to achieve their full potential to provide considerable unique information for complex disease, and that their conditional informativeness for disease cannot be inferred from their accuracy in predicting regulatory annotations.

Citing Articles

Benchmarking DNA Sequence Models for Causal Regulatory Variant Prediction in Human Genetics.

Benegas G, Eraslan G, Song Y bioRxiv. 2025; .

PMID: 39990426 PMC: 11844472. DOI: 10.1101/2025.02.11.637758.

Iterative improvement of deep learning models using synthetic regulatory genomics.

Ribeiro-Dos-Santos A, Maurano M bioRxiv. 2025; .

PMID: 39974895 PMC: 11838587. DOI: 10.1101/2025.02.04.636130.

Deciphering the tissue-specific functional effect of Alzheimer risk SNPs with deep genome annotation.

Pugalenthi P, He B, Xie L, Nho K, Saykin A, Yan J BioData Min. 2024; 17(1):50.

PMID: 39538253 PMC: 11558841. DOI: 10.1186/s13040-024-00400-1.

Current genomic deep learning models display decreased performance in cell type-specific accessible regions.

Kathail P, Shuai R, Chung R, Ye C, Loeb G, Ioannidis N Genome Biol. 2024; 25(1):202.

PMID: 39090688 PMC: 11293111. DOI: 10.1186/s13059-024-03335-2.

Cross-Species Prediction of Transcription Factor Binding by Adversarial Training of a Novel Nucleotide-Level Deep Neural Network.

Zhang Q, Wang S, Li Z, Pan Y, Huang D Adv Sci (Weinh). 2024; 11(36):e2405685.

PMID: 39076052 PMC: 11423150. DOI: 10.1002/advs.202405685.

References

Maurano M, Humbert R, Rynes E, Thurman R, Haugen E, Wang H . Systematic localization of common disease-associated variation in regulatory DNA. Science. 2012; 337(6099):1190-5. PMC: 3771521. DOI: 10.1126/science.1222794. View

Trynka G, Sandor C, Han B, Xu H, Stranger B, Liu X . Chromatin marks identify critical cell types for fine mapping complex trait variants. Nat Genet. 2012; 45(2):124-30. PMC: 3826950. DOI: 10.1038/ng.2504. View

Pickrell J . Joint analysis of functional genomic data and genome-wide association studies of 18 human traits. Am J Hum Genet. 2014; 94(4):559-73. PMC: 3980523. DOI: 10.1016/j.ajhg.2014.03.004. View

Finucane H, Bulik-Sullivan B, Gusev A, Trynka G, Reshef Y, Loh P . Partitioning heritability by functional annotation using genome-wide association summary statistics. Nat Genet. 2015; 47(11):1228-35. PMC: 4626285. DOI: 10.1038/ng.3404. View

Visscher P, Wray N, Zhang Q, Sklar P, McCarthy M, Brown M . 10 Years of GWAS Discovery: Biology, Function, and Translation. Am J Hum Genet. 2017; 101(1):5-22. PMC: 5501872. DOI: 10.1016/j.ajhg.2017.06.005. View

Ernst J, Kheradpour P, Mikkelsen T, Shoresh N, Ward L, Epstein C . Mapping and analysis of chromatin state dynamics in nine human cell types. Nature. 2011; 473(7345):43-9. PMC: 3088773. DOI: 10.1038/nature09906. View

Andersson R, Gebhard C, Miguel-Escalada I, Hoof I, Bornholdt J, Boyd M . An atlas of active enhancers across human cell types and tissues. Nature. 2014; 507(7493):455-461. PMC: 5215096. DOI: 10.1038/nature12787. View

Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, Heravi-Moussavi A . Integrative analysis of 111 reference human epigenomes. Nature. 2015; 518(7539):317-30. PMC: 4530010. DOI: 10.1038/nature14248. 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

10.

Zhou J, Troyanskaya O . Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015; 12(10):931-4. PMC: 4768299. DOI: 10.1038/nmeth.3547. View

11.

Kelley D, Snoek J, Rinn J . Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 2016; 26(7):990-9. PMC: 4937568. DOI: 10.1101/gr.200535.115. View

12.

Zhou J, Theesfeld C, Yao K, Chen K, Wong A, Troyanskaya O . Deep learning sequence-based ab initio prediction of variant effects on expression and disease risk. Nat Genet. 2018; 50(8):1171-1179. PMC: 6094955. DOI: 10.1038/s41588-018-0160-6. View

13.

Kelley D, Reshef Y, Bileschi M, Belanger D, McLean C, Snoek J . Sequential regulatory activity prediction across chromosomes with convolutional neural networks. Genome Res. 2018; 28(5):739-750. PMC: 5932613. DOI: 10.1101/gr.227819.117. View

14.

Zou J, Huss M, Abid A, Mohammadi P, Torkamani A, Telenti A . A primer on deep learning in genomics. Nat Genet. 2018; 51(1):12-18. PMC: 11180539. DOI: 10.1038/s41588-018-0295-5. View

15.

Eraslan G, Avsec Z, Gagneur J, Theis F . Deep learning: new computational modelling techniques for genomics. Nat Rev Genet. 2019; 20(7):389-403. DOI: 10.1038/s41576-019-0122-6. View

16.

Gazal S, Finucane H, Furlotte N, Loh P, Palamara P, Liu X . Linkage disequilibrium-dependent architecture of human complex traits shows action of negative selection. Nat Genet. 2017; 49(10):1421-1427. PMC: 6133304. DOI: 10.1038/ng.3954. View

17.

Ernst J, Kellis M . Large-scale imputation of epigenomic datasets for systematic annotation of diverse human tissues. Nat Biotechnol. 2015; 33(4):364-76. PMC: 4512306. DOI: 10.1038/nbt.3157. View

18.

Ernst J, Kellis M . ChromHMM: automating chromatin-state discovery and characterization. Nat Methods. 2012; 9(3):215-6. PMC: 3577932. DOI: 10.1038/nmeth.1906. View

19.

Ernst J, Kellis M . Chromatin-state discovery and genome annotation with ChromHMM. Nat Protoc. 2017; 12(12):2478-2492. PMC: 5945550. DOI: 10.1038/nprot.2017.124. View

20.

MacArthur J, Bowler E, Cerezo M, Gil L, Hall P, Hastings E . The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res. 2016; 45(D1):D896-D901. PMC: 5210590. DOI: 10.1093/nar/gkw1133. View