Alvarez-Torres Maria Del Mar, Fu Xi, Rabadan Raul
Program for Mathematical Genomics, Department of Systems Biology, Columbia University, New York, New York.
Department of Biomedical Informatics, Columbia University, New York, New York.
Cancer Res. 2025 Jul 2;85(13):2368-2375. doi: 10.1158/0008-5472.CAN-25-0482.
Understanding the vast noncoding cancer genome requires cutting-edge, high-resolution, and accessible strategies. Artificial intelligence is revolutionizing cancer research, enabling advanced models to analyze genome regulation. This review examines illustrative examples of noncoding mutations in cancer, focusing on both key regulatory elements and risk-associated variants that remain poorly understood, and compares key artificial intelligence models developed over the last decade for identifying functional noncoding variants, predicting gene expression impacts, and uncovering cancer-associated mutations. The discussion of the goals, data requirements, features, and outcomes of the models offers practical insights to help cancer researchers integrate these technologies into their work, regardless of computational expertise. This article is part of a special series: Driving Cancer Discoveries with Computational Research, Data Science, and Machine Learning/AI.
了解庞大的非编码癌症基因组需要前沿、高分辨率且易于使用的策略。人工智能正在彻底改变癌症研究，使先进模型能够分析基因组调控。本综述研究了癌症中非编码突变的示例，重点关注关键调控元件和仍知之甚少的风险相关变异，并比较了过去十年开发的用于识别功能性非编码变异、预测基因表达影响以及发现癌症相关突变的关键人工智能模型。对这些模型的目标、数据要求、特征和结果的讨论提供了实用见解，以帮助癌症研究人员将这些技术融入他们的工作，无论其计算专业知识如何。本文是一个特别系列的一部分：通过计算研究、数据科学和机器学习/人工智能推动癌症发现。
