Recent advances of machine learning applications in human gut microbiota study: from observational analysis toward?causal inference and clinical intervention
Recent advances of machine learning applications in human gut microbiota study: from observational analysis toward?causal inference and clinical intervention
复制标题
机器学习在人类肠道微生物群研究中应用的最新进展：从观察分析到因果推理和临床干预
DOI：
10.1016/j.copbio.2022.102884
复制
发表时间：
2023
影响因子：
7.7
通讯作者：
Yamada Takuji
中科院分区：
文献类型：
--
作者：
Salim Felix;Mizutani Sayaka;Zolfo Moreno;Yamada Takuji
HighlightsML有助于检测疾病相关的微生物特征。开放数据存储库和工具开发提高了研究的可重复性。最近的方法集成了SNV和微生物组来进行微生物-疾病因果推断。统计方法也可用于设计基于微生物组的临床干预措施。统计方法，特别是机器学习、学习（ML），对于分析多组学人类肠道微生物组研究生成的大数据至关重要。这些分析导致发现微生物与疾病的关联。此外，最近为提高数据透明度和可访问的分析工具所做的努力提高了数据的可用性和研究的可重复性。我们最近积累的关于微生物-疾病关联的知识为接下来的问题带来了光明：微生物在疾病进展中的作用是什么，以及我们如何在临床环境中应用我们的微生物组知识？在这里，我们介绍了最近实施ML来回答因果推理和临床翻译问题的研究。
HighlightsML facilitates the detection of disease-associated microbial features.Open data repository and tool development improves study reproducibility.Recent methods integrate SNV and microbiome for microbe-disease causal inference.Statistical methods can also be used to design microbiome-based clinical intervention.Statistical methods, especially machine learning, learning (ML), are pivotal for the analyses of large data generated by multiomics human gut microbiota study. These analyses lead to the discovery of microbe-disease associations. Furthermore, recent efforts for more data transparency and accessible analytical tools improved data availability and study reproducibility. Our recent accumulated knowledge on microbe-disease associations brings light to the next questions: what is the role of microbes in disease progression and how can we apply our knowledge of microbiome in clinical settings? Here, we introduce recent studies that implemented ML to answer the questions of causal inference and clinical translation.