越来越多的证据表明，人类微生物组在癌症风险和治疗中发挥着重要作用。无针对性的“组学”技术加速了微生物组与癌症相互作用的研究，支持新的关联和机制的发现。然而，这些技术需要仔细选择和使用，以避免偏差和其他陷阱。在本文中，我们讨论了癌症背景下微生物组数据分析所涉及的选定挑战，包括机器学习 (ML) 的应用。我们专注于基于 DNA 测序（例如宏基因组学）的方法，但许多陷阱和机遇也适用于其他组学技术。我们倡导扩大培训机会、社区标准以及共享数据和代码的最佳实践，以提高癌症微生物组研究的透明度和可重复性。版权所有 © 2024 作者。由爱思唯尔有限公司出版。保留所有权利。

Increasing evidence suggests that the human microbiome plays an important role in cancer risk and treatment. Untargeted 'omics' techniques have accelerated research into microbiome-cancer interactions, supporting the discovery of novel associations and mechanisms. However, these techniques require careful selection and use to avoid biases and other pitfalls. In this essay, we discuss selected challenges involved in the analysis of microbiome data in the context of cancer, including the application of machine learning (ML). We focus on DNA sequencing-based (e.g., metagenomics) methods, but many of the pitfalls and opportunities generalize to other omics technologies as well. We advocate for extended training opportunities, community standards, and best practices for sharing data and code to advance transparency and reproducibility in cancer microbiome research.Copyright © 2024 The Author(s). Published by Elsevier Ltd.. All rights reserved.