3.8 Article

From function to translation: Decoding genetic susceptibility to human diseases via artificial intelligence

Related references

Note: Only part of the references are listed.
Article Genetics & Heredity

Ethical layering in AI-driven polygenic risk scores-New complexities, new challenges

Marie-Christine Fritzsche et al.

Summary: Researchers aim to develop polygenic risk scores as a tool to prevent and more effectively treat serious diseases, disorders and conditions using machine learning techniques. However, the use of artificial intelligence for polygenic risk scores presents complex ethical challenges that have not been sufficiently considered.

FRONTIERS IN GENETICS (2023)

Review Genetics & Heredity

Obtaining genetics insights from deep learning via explainable artificial intelligence

Gherman Novakovsky et al.

Summary: This Review describes the advances in explainable artificial intelligence (xAI) in genomics, where researchers are using deep learning approaches to gain biological insights behind the models, moving beyond the traditional 'black box' nature.

NATURE REVIEWS GENETICS (2023)

Article Computer Science, Artificial Intelligence

AI under great uncertainty: implications and decision strategies for public policy

Maria Nordstrom

Summary: This paper discusses decisions made under great uncertainty, specifically focusing on the uncertainties surrounding public policy decisions related to machine learning and AI. It suggests that decision-makers can mitigate this uncertainty by adopting strategies from decision theory, such as framing and temporal strategies.

AI & SOCIETY (2022)

Article Multidisciplinary Sciences

The case for increasing diversity in tissue-based functional genomics datasets to understand human disease susceptibility

Erping Long et al.

Summary: The lack of diversity in ancestry and tissue types in tissue-based functional genomics resources limits comprehensive investigation of gene regulation. Global efforts to increase tissue diversity will promote more equitable medical care.

NATURE COMMUNICATIONS (2022)

Article Biochemical Research Methods

tmVar 3.0: an improved variant concept recognition and normalization tool

Chih-Hsuan Wei et al.

Summary: In this study, we propose tmVar 3.0, an improved variant recognition and normalization system. Compared to its predecessors, tmVar 3.0 is able to recognize a wider range of variant-related entities and improve accuracy by grouping variant mentions. It also provides advanced variant normalization options. Evaluation results show that tmVar 3.0 achieves state-of-the-art performance with over 90% F-measure. The tool and annotations for the entire PubMed and PMC datasets are freely available for download.

BIOINFORMATICS (2022)

Article Biochemical Research Methods

Empowering the discovery of novel target-disease associations via machine learning approaches in the open targets platform

Yingnan Han et al.

Summary: The study introduces a novel approach to uncover druggable disease to target indications through integrating additional target-disease features with machine learning models.

BMC BIOINFORMATICS (2022)

Article Genetics & Heredity

Combining SNP-to-gene linking strategies to identify disease genes and assess disease omnigenicity

Steven Gazal et al.

Summary: A heritability-based framework was developed to evaluate SNP-to-gene linking methods and a combined approach was constructed. This approach achieved high precision and recall in predicting causal SNP-gene-disease triplets. The study also provided insights into the architecture of gene-level variations and disease omnigenicity.

NATURE GENETICS (2022)

Article Genetics & Heredity

Epigenomic and transcriptomic analyses define core cell types, genes and targetable mechanisms for kidney disease

Hongbo Liu et al.

Summary: In this study, the genetic association with kidney function was analyzed in over 1.5 million individuals, leading to the identification of 878 loci and target genes. The role of methylation variation in heritability of kidney function was highlighted. The findings emphasize the importance of bulk and single-cell epigenomic information in understanding the mechanisms of complex traits.

NATURE GENETICS (2022)

Article Genetics & Heredity

A sequence-based global map of regulatory activity for deciphering human genetics

Kathleen M. Chen et al.

Summary: Sei is a new framework for integrating human genetics data with a sequence-based mapping of predicted regulatory activities to elucidate mechanisms contributing to complex traits and diseases.

NATURE GENETICS (2022)

Article Geriatrics & Gerontology

Identification of Therapeutic Targets for Amyotrophic Lateral Sclerosis Using PandaOmics - An AI-Enabled Biological Target Discovery Platform

Frank W. Pun et al.

Summary: In this study, an AI-driven target discovery platform called PandaOmics was used to analyze the expression profiles of CNS samples and direct iPSC-derived motor neurons from public datasets and Answer ALS. The study identified 17 high-confidence and 11 novel therapeutic targets, and verified their therapeutic effects in a fruit fly model. The research provides new insights into ALS pathophysiology and demonstrates the ability of AI to speed up the target discovery process.

FRONTIERS IN AGING NEUROSCIENCE (2022)

Review Biochemistry & Molecular Biology

Multimodal biomedical AI

Julian N. Acosta et al.

Summary: This Review discusses the potential applications, technical pitfalls, and challenges of multimodal artificial intelligence models in the field of health and medicine. By utilizing data from large biobanks, electronic health records, medical imaging, and genomics, multimodal AI solutions have the ability to capture the complexity of human health and disease. Key areas such as personalized medicine, digital clinical trials, remote monitoring and care, pandemic surveillance, and virtual health assistants offer opportunities, but data, modeling, and privacy challenges need to be addressed for the full potential of multimodal AI in health to be realized.

NATURE MEDICINE (2022)

Article Biotechnology & Applied Microbiology

Polygenic transcriptome risk scores (PTRS) can improve portability of polygenic risk scores across ancestries

Yanyu Liang et al.

Summary: This study introduces a polygenic transcriptome risk score (PTRS) based on predicted transcript levels, and explores its portability across populations using UK Biobank data. The results show that PTRS has significantly higher portability in African-descent samples, where the loss of performance is most acute, and performs better than PRS when used in combination.

GENOME BIOLOGY (2022)

Article Biochemical Research Methods

Varmole: a biologically drop-connect deep neural network model for prioritizing disease risk variants

Nam D. Nguyen et al.

Summary: Varmole is an interpretable deep learning method that reveals genomic functions and mechanisms while predicting phenotype from genotype. It embeds multi-omic networks into a deep neural network architecture and prioritizes variants, genes, and regulatory linkages through biological drop-connect without the need for prior feature selection.

BIOINFORMATICS (2021)

Article Multidisciplinary Sciences

Disease variant prediction with deep generative models of evolutionary data

Jonathan Frazer et al.

Summary: Quantifying the pathogenicity of protein variants in human disease-related genes is important for clinical decisions, but most variants still have unknown consequences. Current computational methods rely on training machine learning models on known disease labels, resulting in unreliable models due to sparse and biased labels.

NATURE (2021)

Article Genetics & Heredity

An open approach to systematically prioritize causal variants and genes at all published human GWAS trait-associated loci

Edward Mountjoy et al.

Summary: In this study, an open resource was introduced to systematically fine map and prioritize genes across published GWAS loci, aiming at identifying causal genes associated with complex traits. By training a machine learning model using genetics and functional genomics data, the authors were able to distinguish causal genes from neighboring genes, outperforming distance-based models. The prioritized genes were found to be enriched for known approved drug targets, suggesting their potential in drug development.

NATURE GENETICS (2021)

Review Genetics & Heredity

From GWAS to Gene: Transcriptome-Wide Association Studies and Other Methods to Functionally Understand GWAS Discoveries

Binglan Li et al.

Summary: GWAS has identified numerous SNP loci associated with complex human diseases, most of which are located in non-coding regions with unknown functions. Integrative bioinformatics methods combining GWAS and functional genomics knowledge help identify genetically regulated genes, facilitating the study of complex disease mechanisms.

FRONTIERS IN GENETICS (2021)

Article Health Care Sciences & Services

Biological data annotation via a human-augmenting AI-based labeling system

Douwe van der Wal et al.

Summary: HALS is a human-in-the-loop data labeling AI system that learns annotations from humans in real-time, reducing the workload of annotators and increasing the quality of data annotations.

NPJ DIGITAL MEDICINE (2021)

Review Genetics & Heredity

Advancing the use of genome-wide association studies for drug repurposing

William R. Reay et al.

Summary: Genome-wide association studies have provided important biological insights into complex diseases, offering opportunities for drug repurposing. Leveraging common variant genetics through approaches such as single-loci mapping to drug targets, transcriptome-wide association studies, gene-set association, causal inference by Mendelian randomization, and polygenic scoring, researchers aim to identify new treatment options more efficiently.

NATURE REVIEWS GENETICS (2021)

Article Genetics & Heredity

scGRNom: a computational pipeline of integrative multi-omics analyses for predicting cell-type disease genes and regulatory networks

Ting Jin et al.

Summary: The study developed a computational pipeline to predict cell-type disease genes and regulatory networks, including transcription factors and regulatory elements. Applications to schizophrenia and Alzheimer's disease revealed cross-disease and disease-specific functions and pathways at the cell-type level.

GENOME MEDICINE (2021)

Article Computer Science, Information Systems

Deep Survival Machines: Fully Parametric Survival Regression and Representation Learning for Censored Data With Competing Risks

Chirag Nagpal et al.

Summary: This study introduces a new approach for estimating relative risks in time-to-event prediction problems with censored data in a fully parametric manner, without assuming constant proportional hazards. Through experiments on real world datasets with different levels of censoring, the benefits of this approach in estimating survival risks and competing risks are demonstrated. This work represents the first study involving fully parametric estimation of survival times with competing risks in the presence of censoring.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2021)

Review Nanoscience & Nanotechnology

Machine learning methods to model multicellular complexity and tissue specificity

Rachel S. G. Sealfon et al.

Summary: Experimental methods to study tissue specificity reveal the nature and organization of cell types and tissues in complex multicellular organisms. Machine learning approaches provide a powerful tool for analyzing and interpreting tissue-specific experimental data. Deep learning and network-based methods can predict tissue-dependent effects on gene expression, disease phenotypes, and capture relationships between biomolecules.

NATURE REVIEWS MATERIALS (2021)

Letter Dermatology

Artificial intelligence, machine learning and deep learning: definitions and differences

D. Jakhar et al.

CLINICAL AND EXPERIMENTAL DERMATOLOGY (2020)

Article Biochemical Research Methods

Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data

Aditya Pratapa et al.

NATURE METHODS (2020)

Review Clinical Neurology

CRISPR-based functional genomics for neurological disease

Martin Kampmann

NATURE REVIEWS NEUROLOGY (2020)

Review Computer Science, Artificial Intelligence

Secure, privacy-preserving and federated machine learning in medical imaging

Georgios A. Kaissis et al.

NATURE MACHINE INTELLIGENCE (2020)

Review Genetics & Heredity

Deep learning: new computational modelling techniques for genomics

Gokcen Eraslan et al.

NATURE REVIEWS GENETICS (2019)

Article Genetics & Heredity

A primer on deep learning in genomics

James Zou et al.

NATURE GENETICS (2019)

Article Multidisciplinary Sciences

Deep learning for inferring gene relationships from single-cell expression data

Ye Yuan et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2019)

Article Biochemical Research Methods

Using deep learning to model the hierarchical structure and function of a cell

Jianzhu Ma et al.

NATURE METHODS (2018)

Review Genetics & Heredity

From genome-wide associations to candidate causal variants by statistical fine-mapping

Daniel J. Schaid et al.

NATURE REVIEWS GENETICS (2018)

Article Biochemical Research Methods

Scaling up data curation using deep learning: An application to literature triage in genomic variation resources

Kyubum Lee et al.

PLOS COMPUTATIONAL BIOLOGY (2018)

Article Multidisciplinary Sciences

Comprehensive functional genomic resource and integrative model for the human brain

Daifeng Wang et al.

SCIENCE (2018)

Review Genetics & Heredity

Dissecting the genetics of complex traits using summary association statistics

Bogdan Pasaniuc et al.

NATURE REVIEWS GENETICS (2017)

Article Multidisciplinary Sciences

Genetic effects on gene expression across human tissues

Francois Aguet et al.

NATURE (2017)

Article Endocrinology & Metabolism

Enhanced Identification of Potential Pleiotropic Genetic Variants for Bone Mineral Density and Breast Cancer

Cheng Peng et al.

CALCIFIED TISSUE INTERNATIONAL (2017)

Article Genetics & Heredity

A Dementia-Associated Risk Variant near TMEM106B Alters Chromatin Architecture and Gene Expression

Michael D. Gallagher et al.

AMERICAN JOURNAL OF HUMAN GENETICS (2017)

Review Genetics & Heredity

10 Years of GWAS Discovery: Biology, Function, and Translation

Peter M. Visscher et al.

AMERICAN JOURNAL OF HUMAN GENETICS (2017)

Review Biochemistry & Molecular Biology

An Expanded View of Complex Traits: From Polygenic to Omnigenic

Evan A. Boyle et al.

Article Biochemical Research Methods

Interpreting functional effects of coding variants: challenges in proteome-scale prediction, annotation and assessment

Khader Shameer et al.

BRIEFINGS IN BIOINFORMATICS (2016)

Review Biochemistry & Molecular Biology

Deep learning for computational biology

Christof Angermueller et al.

MOLECULAR SYSTEMS BIOLOGY (2016)

Article Biotechnology & Applied Microbiology

Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning

Babak Alipanahi et al.

NATURE BIOTECHNOLOGY (2015)

Article Genetics & Heredity

The support of human genetic evidence for approved drug indications

Matthew R. Nelson et al.

NATURE GENETICS (2015)

Article Pharmacology & Pharmacy

The Discovery and Development of Selective Estrogen Receptor Modulators (SERMs) for Clinical Practice

Philipp Y. Maximov et al.

CURRENT CLINICAL PHARMACOLOGY (2013)

Article Multidisciplinary Sciences

An integrated encyclopedia of DNA elements in the human genome

Ian Dunham et al.

NATURE (2012)

Article Multidisciplinary Sciences

Biological, clinical and population relevance of 95 loci for blood lipids

Tanya M. Teslovich et al.

NATURE (2010)

Article Genetics & Heredity

Genome-wide scan reveals association of psoriasis with IL-23 and NF-κB pathways

Rajan P. Nair et al.

NATURE GENETICS (2009)