Multi-omics approaches for understanding gene-environment interactions in noncommunicable diseases: techniques, translation, and equity issues DOI Creative Commons

Robel Alemu,

Nigussie Tadesse Sharew,

Yodit Y. Arsano

и другие.

Human Genomics, Год журнала: 2025, Номер 19(1)

Опубликована: Янв. 31, 2025

Non-communicable diseases (NCDs) such as cardiovascular diseases, chronic respiratory cancers, diabetes, and mental health disorders pose a significant global challenge, accounting for the majority of fatalities disability-adjusted life years worldwide. These arise from complex interactions between genetic, behavioral, environmental factors, necessitating thorough understanding these dynamics to identify effective diagnostic strategies interventions. Although recent advances in multi-omics technologies have greatly enhanced our ability explore interactions, several challenges remain. include inherent complexity heterogeneity multi-omic datasets, limitations analytical approaches, severe underrepresentation non-European genetic ancestries most omics which restricts generalizability findings exacerbates disparities. This scoping review evaluates landscape data related NCDs 2000 2024, focusing on advancements integration, translational applications, equity considerations. We highlight need standardized protocols, harmonized data-sharing policies, advanced approaches artificial intelligence/machine learning integrate study gene-environment interactions. also opportunities translating insights (GxE) research into precision medicine strategies. underscore potential advancing enhancing patient outcomes across diverse underserved populations, emphasizing fairness-centered strategic investments build local capacities underrepresented populations regions.

Язык: Английский

Wearable chemical sensors for biomarker discovery in the omics era DOI Open Access
Juliane R. Sempionatto, José A. Lasalde‐Ramírez, Kuldeep Mahato

и другие.

Nature Reviews Chemistry, Год журнала: 2022, Номер 6(12), С. 899 - 915

Опубликована: Ноя. 15, 2022

Язык: Английский

Процитировано

327

Machine learning for multi-omics data integration in cancer DOI Creative Commons
Zhaoxiang Cai, Rebecca C. Poulos, Jia Liu

и другие.

iScience, Год журнала: 2022, Номер 25(2), С. 103798 - 103798

Опубликована: Янв. 22, 2022

Multi-omics data analysis is an important aspect of cancer molecular biology studies and has led to ground-breaking discoveries. Many efforts have been made develop machine learning methods that automatically integrate omics data. Here, we review tools categorized as either general-purpose or task-specific, covering both supervised unsupervised for integrative multi-omics We benchmark the performance five approaches using from Cancer Cell Line Encyclopedia, reporting accuracy on type classification mean absolute error drug response prediction, evaluating runtime efficiency. This provides recommendations researchers regarding suitable method selection their specific applications. It should also promote development novel methodologies integration, which will be essential discovery, clinical trial design, personalized treatments.

Язык: Английский

Процитировано

163

Smart breeding driven by big data, artificial intelligence, and integrated genomic-enviromic prediction DOI Creative Commons
Yunbi Xu, Xingping Zhang, Huihui Li

и другие.

Molecular Plant, Год журнала: 2022, Номер 15(11), С. 1664 - 1695

Опубликована: Сен. 7, 2022

The first paradigm of plant breeding involves direct selection-based phenotypic observation, followed by predictive using statistical models for quantitative traits constructed based on genetic experimental design and, more recently, incorporation molecular marker genotypes. However, performance or phenotype (P) is determined the combined effects genotype (G), envirotype (E), and environment interaction (GEI). Phenotypes can be predicted precisely training a model data collected from multiple sources, including spatiotemporal omics (genomics, phenomics, enviromics across time space). Integration 3D information profiles (G-P-E), each with multidimensionality, provides both tremendous opportunities great challenges. Here, we review innovative technologies breeding. We then evaluate multidimensional that integrated strategy, particularly envirotypic data, which have largely been neglected in collection are nearly untouched construction. propose smart scheme, genomic-enviromic prediction (iGEP), as an extension genomic prediction, multiomics information, big technology, artificial intelligence (mainly focused machine deep learning). discuss how to implement iGEP, models, environmental indices, factorial structure cross-species prediction. A strategy proposed prediction-based crop redesign at macro (individual, population, species) micro (gene, metabolism, network) scales. Finally, provide perspectives translating into gain through integrative platforms open-source initiatives. call coordinated efforts institutional partnerships, technological support.

Язык: Английский

Процитировано

159

Multimodal data fusion for cancer biomarker discovery with deep learning DOI
Sandra Steyaert,

Marija Pizurica,

Divya Nagaraj

и другие.

Nature Machine Intelligence, Год журнала: 2023, Номер 5(4), С. 351 - 362

Опубликована: Апрель 6, 2023

Язык: Английский

Процитировано

154

Artificial intelligence assists precision medicine in cancer treatment DOI Creative Commons
Jinzhuang Liao, M Kellis, Yu Gan

и другие.

Frontiers in Oncology, Год журнала: 2023, Номер 12

Опубликована: Янв. 4, 2023

Cancer is a major medical problem worldwide. Due to its high heterogeneity, the use of same drugs or surgical methods in patients with tumor may have different curative effects, leading need for more accurate treatment tumors and personalized treatments patients. The precise essential, which renders obtaining an in-depth understanding changes that undergo urgent, including their genes, proteins cancer cell phenotypes, order develop targeted strategies Artificial intelligence (AI) based on big data can extract hidden patterns, important information, corresponding knowledge behind enormous amount data. For example, ML deep learning subsets AI be used mine deep-level information genomics, transcriptomics, proteomics, radiomics, digital pathological images, other data, make clinicians synthetically comprehensively understand tumors. In addition, find new biomarkers from assist screening, detection, diagnosis, prognosis prediction, so as providing best individual improving clinical outcomes.

Язык: Английский

Процитировано

117

A guide to multi-omics data collection and integration for translational medicine DOI Creative Commons
Efi Athieniti, George M. Spyrou

Computational and Structural Biotechnology Journal, Год журнала: 2022, Номер 21, С. 134 - 149

Опубликована: Дек. 1, 2022

The emerging high-throughput technologies have led to the shift in design of translational medicine projects towards collecting multi-omics patient samples and, consequently, their integrated analysis. However, complexity integrating these datasets has triggered new questions regarding appropriateness available computational methods. Currently, there is no clear consensus on best combination omics include and data integration methodologies required for This article aims guide studies field types method choose. We review articles that perform multiple measurements from samples. identify five objectives applications: (i) detect disease-associated molecular patterns, (ii) subtype identification, (iii) diagnosis/prognosis, (iv) drug response prediction, (v) understand regulatory processes. describe common trends selection omic combined different diseases. To choice tools, we group them into scientific they aim address. main methods adopted achieve present examples tools. compare tools based how deal with challenges comment against predefined objective-specific evaluation criteria. Finally, discuss downstream analysis further extraction novel insights datasets.

Язык: Английский

Процитировано

87

Machine Learning: A New Prospect in Multi-Omics Data Analysis of Cancer DOI Creative Commons
Babak Arjmand, Shayesteh Kokabi Hamidpour, Akram Tayanloo-Beik

и другие.

Frontiers in Genetics, Год журнала: 2022, Номер 13

Опубликована: Янв. 27, 2022

Cancer is defined as a large group of diseases that associated with abnormal cell growth, uncontrollable division, and may tend to impinge on other tissues the body by different mechanisms through metastasis. What makes cancer so important incidence rate growing worldwide which can have major health, economic, even social impacts both patients governments. Thereby, early prognosis, diagnosis, treatment play crucial role at front line combating cancer. The onset progression occur under influence complicated some alterations in level genome, proteome, transcriptome, metabolome etc. Consequently, advent omics science its broad research branches (such genomics, proteomics, transcriptomics, metabolomics, forth) revolutionary biological approaches opened new doors comprehensive perception landscape. Due complexities formation development cancer, study underlying has gone beyond just one field arena. Therefore, making connection between resultant data from examining them multi-omics pave way for facilitating discovery novel prognostic, diagnostic, therapeutic approaches. As volume complexity studies are increasing dramatically, use leading-edge technologies such machine learning promising assessments data. Machine categorized subset artificial intelligence aims parsing, classification, pattern identification applying statistical methods algorithms. This acquired knowledge subsequently allows computers learn improve accurate predictions experiences processing. In this context, application learning, computational technology offers opportunities achieving in-depth analysis studies. it be concluded roles fight against

Язык: Английский

Процитировано

83

Web-based multi-omics integration using the Analyst software suite DOI
Jessica Ewald,

Guangyan Zhou,

Yao Lü

и другие.

Nature Protocols, Год журнала: 2024, Номер 19(5), С. 1467 - 1497

Опубликована: Фев. 14, 2024

Язык: Английский

Процитировано

71

Missing data in multi-omics integration: Recent advances through artificial intelligence DOI Creative Commons
Javier E. Flores, Daniel Claborne, Zachary D. Weller

и другие.

Frontiers in Artificial Intelligence, Год журнала: 2023, Номер 6

Опубликована: Фев. 9, 2023

Biological systems function through complex interactions between various 'omics (biomolecules), and a more complete understanding of these is only possible an integrated, multi-omic perspective. This has presented the need for development integration approaches that are able to capture complex, often non-linear, define biological adapted challenges combining heterogenous data across 'omic views. A principal challenge missing because all biomolecules not measured in samples. Due either cost, instrument sensitivity, or other experimental factors, sample may be one techologies. Recent methodological developments artificial intelligence statistical learning have greatly facilitated analyses multi-omics data, however many techniques assume access completely observed data. subset methods incorporate mechanisms handling partially samples, focus this review. We describe recently developed approaches, noting their primary use cases highlighting each method's approach additionally provide overview traditional workflows limitations; we discuss potential avenues further as well how issue its current solutions generalize beyond context.

Язык: Английский

Процитировано

67

Discovery of drug–omics associations in type 2 diabetes with generative deep-learning models DOI Creative Commons
Rosa Lundbye Allesøe, Agnete Troen Lundgaard, Ricardo Hernández Medina

и другие.

Nature Biotechnology, Год журнала: 2023, Номер 41(3), С. 399 - 408

Опубликована: Янв. 2, 2023

The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response treatment. However, scale heterogeneous nature multi-modal data makes integration inference a non-trivial task. We developed deep-learning-based framework, multi-omics variational autoencoders (MOVE), integrate such applied it cohort 789 people with newly diagnosed type 2 diabetes deep phenotyping from DIRECT consortium. Using silico perturbations, we identified drug-omics associations across datasets for 20 most prevalent drugs given substantially higher sensitivity than univariate statistical tests. From these, among others, novel between metformin gut microbiota as well opposite molecular responses two statins, simvastatin atorvastatin. used quantify drug-drug similarities, assess degree polypharmacy conclude that drug effects are distributed modalities.

Язык: Английский

Процитировано

63