Integrating machine learning with OMICs data for early detection in breast cancer

IF 0.9 Q4 GENETICS & HEREDITY

Gene Reports Pub Date : 2025-08-19 DOI:10.1016/j.genrep.2025.102325

Jiaqi Mu , Aquib Nazar , Muhammad Asim Ali , Athar Hussain

引用次数: 0

Abstract

Breast cancer is one of the most common cancers that significantly affects a large population of women, emphasizing its importance in early detection for effective treatments. The advancement in technologies, especially in machine learning and its integration with multi-omics data, such as genomics, transcriptomics, proteomics, metabolomics, and imaging, is not only revolutionizing the diagnosis and prognosis of breast cancer at its early stages but also providing a door for personalized treatment plans to improve patient outcomes. However, achieving truly personalized treatment requires integration of causal inference methods into machine learning frameworks, as correlational models alone may not ensure effective or safe decision-making. The current study revisits the progress made in this research area, providing a comprehensive insight into the challenges of breast cancer early detection, machine learning (ML) in cancer detection, ML-Omics integration, clinical applications, case studies, and future directions and innovations.

Abstract Image

查看原文本刊更多论文

将机器学习与组学数据相结合，用于乳腺癌的早期检测

乳腺癌是最常见的癌症之一，严重影响大量妇女，强调了早期发现有效治疗的重要性。技术的进步，特别是机器学习及其与多组学数据的整合，如基因组学、转录组学、蛋白质组学、代谢组学和成像，不仅在乳腺癌早期阶段的诊断和预后方面发生了革命性的变化，而且为个性化治疗计划提供了一扇门，以改善患者的预后。然而，实现真正的个性化治疗需要将因果推理方法整合到机器学习框架中，因为单独的相关模型可能无法确保有效或安全的决策。本研究回顾了这一研究领域的进展，全面了解了乳腺癌早期检测、机器学习（ML）在癌症检测中的应用、ML- omics整合、临床应用、案例研究以及未来的发展方向和创新。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Gene Reports Biochemistry, Genetics and Molecular Biology-Genetics

CiteScore

3.30

自引率

7.70%

发文量

246

审稿时长

49 days

期刊介绍： Gene Reports publishes papers that focus on the regulation, expression, function and evolution of genes in all biological contexts, including all prokaryotic and eukaryotic organisms, as well as viruses. Gene Reports strives to be a very diverse journal and topics in all fields will be considered for publication. Although not limited to the following, some general topics include: DNA Organization, Replication & Evolution -Focus on genomic DNA (chromosomal organization, comparative genomics, DNA replication, DNA repair, mobile DNA, mitochondrial DNA, chloroplast DNA). Expression & Function - Focus on functional RNAs (microRNAs, tRNAs, rRNAs, mRNA splicing, alternative polyadenylation) Regulation - Focus on processes that mediate gene-read out (epigenetics, chromatin, histone code, transcription, translation, protein degradation). Cell Signaling - Focus on mechanisms that control information flow into the nucleus to control gene expression (kinase and phosphatase pathways controlled by extra-cellular ligands, Wnt, Notch, TGFbeta/BMPs, FGFs, IGFs etc.) Profiling of gene expression and genetic variation - Focus on high throughput approaches (e.g., DeepSeq, ChIP-Seq, Affymetrix microarrays, proteomics) that define gene regulatory circuitry, molecular pathways and protein/protein networks. Genetics - Focus on development in model organisms (e.g., mouse, frog, fruit fly, worm), human genetic variation, population genetics, as well as agricultural and veterinary genetics. Molecular Pathology & Regenerative Medicine - Focus on the deregulation of molecular processes in human diseases and mechanisms supporting regeneration of tissues through pluripotent or multipotent stem cells.