Integrative approaches to improve the informativeness of deep learning models for human complex diseases

Dey, Kushal K.; Kim, Samuel S.; Gazal, Steven; Nasser, Joseph; Engreitz, Jesse M.; Price, Alkes L.

doi:10.1101/2020.09.08.288563

Cited by 4 publications

(2 citation statements)

References 104 publications

(256 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More fine-grained enhancer-gene linking strategies will likely prove beneficial, but the strategies that we used here provide a clear improvement over standard gene window-based approaches. We did not perform a comprehensive evaluation of enhancer-gene linking strategies and methods to combine them, which will be provided elsewhere( 155, 156 ) (S. Gazal, unpublished data). Second, we focus on genome-wide disease heritability (rather than a particular locus); however, our approach can be used to implicate specific genes and gene programs.…”

Section: Discussionmentioning

confidence: 99%

Identifying disease-critical cell types and cellular processes across the human body by integration of single-cell profiles and human genetics

Jagadeesh

Dey

Montoro

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Cellular dysfunction is a hallmark of disease. Genome-wide association studies (GWAS) have provided a powerful means to identify loci and genes contributing to disease risk, but in many cases the related cell types/states through which genes confer disease risk remain unknown. Deciphering such relationships is important both for our understanding of disease, and for developing therapeutic interventions. Here, we introduce a framework for integrating single-cell RNA-seq (scRNA-seq), epigenomic maps and GWAS summary statistics to infer the underlying cell types and processes by which genetic variants influence disease. We analyzed 1.6 million scRNA-seq profiles from 209 individuals spanning 11 tissue types and 6 disease conditions, and constructed gene programs capturing cell types, disease progression in cell types, and cellular processes both within and across cell types. We evaluated these gene programs for disease enrichment by transforming them to SNP annotations with tissue-specific epigenomic maps and computing enrichment scores across 60 diseases and complex traits (average N=297K). The inferred disease enrichments recapitulated known biology and highlighted novel relationships for different conditions, including GABAergic neurons in major depressive disorder (MDD), disease progression programs in M cells in ulcerative colitis, and a disease-specific complement cascade process in multiple sclerosis. Our framework provides a powerful approach for identifying the cell types and cellular processes by which genetic variants influence disease.

show abstract

Section: Discussionmentioning

confidence: 99%

Identifying disease-critical cell types and cellular processes across the human body by integration of single-cell profiles and human genetics

Jagadeesh

Dey

Montoro

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The Sei model simultaneously predicted 21,907 binary assay labels which were dimension-reduced to 40 features representing sequence classes. Assembling deep learning model predictions and training with comprehensive features reduced the variances of our final models and generalizes well to new cell lines and less-studied organs (35, 36).…”

Section: Resultsmentioning

confidence: 99%

Organ-specific prioritization and annotation of non-coding regulatory variants in the human genome

Zhao,

Dong,

Boyle

2023

Preprint

View full text Add to dashboard Cite

Identifying non-coding regulatory variants in the human genome remains a challenging task in genomics. Recently we advanced our leading regulatory variant database, RegulomeDB, to its second version. Building upon this comprehensive database, we developed a novel machine-learning architecture with stacked generalization, TLand, which utilizes RegulomeDB-derived features to predict regulatory variants at cell or organ-specific levels. In our holdout benchmarking, TLand consistently outperformed state-of-the-art models, demonstrating its ability to generalize to new cell lines or organs. We trained three types of organ-specific TLand models to overcome the common model bias toward high data availability cell lines or organs. These models accurately prioritize relevant organs for 2 million GWAS SNPs associated with GWAS traits. Moreover, our analysis of top-scoring variants in specific organ models showed a high enrichment of relevant GWAS traits. We expect that TLand and RegulomeDB will further advance our ability to understand human regulatory variants genome-wide.

show abstract

Artificial Intelligence for Risk Assessment on Primary Prevention of Coronary Artery Disease

Chen,

Loguercio,

Chen

et al. 2023

Curr Cardiovasc Risk Rep

View full text Add to dashboard Cite

Purpose of Review Coronary artery disease (CAD) is a common and etiologically complex disease worldwide. Current guidelines for primary prevention, or the prevention of a first acute event, include relatively simple risk assessment and leave substantial room for improvement both for risk ascertainment and selection of prevention strategies. Here, we review how advances in big data and predictive modeling foreshadow a promising future of improved risk assessment and precision medicine for CAD. Recent Findings Artificial intelligence (AI) has improved the utility of high dimensional data, providing an opportunity to better understand the interplay between numerous CAD risk factors. Beyond applications of AI in cardiac imaging, the vanguard application of AI in healthcare, recent translational research is also revealing a promising path for AI in multi-modal risk prediction using standard biomarkers, genetic and other omics technologies, a variety of biosensors, and unstructured data from electronic health records (EHRs). However, gaps remain in clinical validation of AI models, most notably in the actionability of complex risk prediction for more precise therapeutic interventions. Summary The recent availability of nation-scale biobank datasets has provided a tremendous opportunity to richly characterize longitudinal health trajectories using health data collected at home, at laboratories, and through clinic visits. The ever-growing availability of deep genotype-phenotype data is poised to drive a transition from simple risk prediction algorithms to complex, “data-hungry,” AI models in clinical decision-making. While AI models provide the means to incorporate essentially all risk factors into comprehensive risk prediction frameworks, there remains a need to wrap these predictions in interpretable frameworks that map to our understanding of underlying biological mechanisms and associated personalized intervention. This review explores recent advances in the role of machine learning and AI in CAD primary prevention and highlights current strengths as well as limitations mediating potential future applications.

show abstract

Integrative approaches to improve the informativeness of deep learning models for human complex diseases

Cited by 4 publications

References 104 publications

Identifying disease-critical cell types and cellular processes across the human body by integration of single-cell profiles and human genetics

Identifying disease-critical cell types and cellular processes across the human body by integration of single-cell profiles and human genetics

Organ-specific prioritization and annotation of non-coding regulatory variants in the human genome

Artificial Intelligence for Risk Assessment on Primary Prevention of Coronary Artery Disease

Contact Info

Product

Resources

About