{"title":"结合代谢活动、分类和群落结构,改进基于微生物组的宿主表型预测模型。","authors":"Mahsa Monshizadeh, Yuzhen Ye","doi":"10.1080/19490976.2024.2302076","DOIUrl":null,"url":null,"abstract":"<p><p>We developed MicroKPNN, a prior-knowledge guided interpretable neural network for microbiome-based human host phenotype prediction. The prior knowledge used in MicroKPNN includes the metabolic activities of different bacterial species, phylogenetic relationships, and bacterial community structure, all in a shallow neural network. Application of MicroKPNN to seven gut microbiome datasets (involving five different human diseases including inflammatory bowel disease, type 2 diabetes, liver cirrhosis, colorectal cancer, and obesity) shows that incorporation of the prior knowledge helped improve the microbiome-based host phenotype prediction. MicroKPNN outperformed fully connected neural network-based approaches in all seven cases, with the most improvement of accuracy in the prediction of type 2 diabetes. MicroKPNN outperformed a recently developed deep-learning based approach DeepMicro, which selects the best combination of autoencoder and machine learning approach to make predictions, in all of the seven cases. Importantly, we showed that MicroKPNN provides a way for interpretation of the predictive models. Using importance scores estimated for the hidden nodes, MicroKPNN could provide explanations for prior research findings by highlighting the roles of specific microbiome components in phenotype predictions. In addition, it may suggest potential future research directions for studying the impacts of microbiome on host health and diseases. MicroKPNN is publicly available at https://github.com/mgtools/MicroKPNN.</p>","PeriodicalId":12909,"journal":{"name":"Gut Microbes","volume":null,"pages":null},"PeriodicalIF":12.2000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10793686/pdf/","citationCount":"0","resultStr":"{\"title\":\"Incorporating metabolic activity, taxonomy and community structure to improve microbiome-based predictive models for host phenotype prediction.\",\"authors\":\"Mahsa Monshizadeh, Yuzhen Ye\",\"doi\":\"10.1080/19490976.2024.2302076\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>We developed MicroKPNN, a prior-knowledge guided interpretable neural network for microbiome-based human host phenotype prediction. The prior knowledge used in MicroKPNN includes the metabolic activities of different bacterial species, phylogenetic relationships, and bacterial community structure, all in a shallow neural network. Application of MicroKPNN to seven gut microbiome datasets (involving five different human diseases including inflammatory bowel disease, type 2 diabetes, liver cirrhosis, colorectal cancer, and obesity) shows that incorporation of the prior knowledge helped improve the microbiome-based host phenotype prediction. MicroKPNN outperformed fully connected neural network-based approaches in all seven cases, with the most improvement of accuracy in the prediction of type 2 diabetes. MicroKPNN outperformed a recently developed deep-learning based approach DeepMicro, which selects the best combination of autoencoder and machine learning approach to make predictions, in all of the seven cases. Importantly, we showed that MicroKPNN provides a way for interpretation of the predictive models. Using importance scores estimated for the hidden nodes, MicroKPNN could provide explanations for prior research findings by highlighting the roles of specific microbiome components in phenotype predictions. In addition, it may suggest potential future research directions for studying the impacts of microbiome on host health and diseases. MicroKPNN is publicly available at https://github.com/mgtools/MicroKPNN.</p>\",\"PeriodicalId\":12909,\"journal\":{\"name\":\"Gut Microbes\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":12.2000,\"publicationDate\":\"2024-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10793686/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Gut Microbes\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/19490976.2024.2302076\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2024/1/12 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"GASTROENTEROLOGY & HEPATOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Gut Microbes","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/19490976.2024.2302076","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/12 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"GASTROENTEROLOGY & HEPATOLOGY","Score":null,"Total":0}
Incorporating metabolic activity, taxonomy and community structure to improve microbiome-based predictive models for host phenotype prediction.
We developed MicroKPNN, a prior-knowledge guided interpretable neural network for microbiome-based human host phenotype prediction. The prior knowledge used in MicroKPNN includes the metabolic activities of different bacterial species, phylogenetic relationships, and bacterial community structure, all in a shallow neural network. Application of MicroKPNN to seven gut microbiome datasets (involving five different human diseases including inflammatory bowel disease, type 2 diabetes, liver cirrhosis, colorectal cancer, and obesity) shows that incorporation of the prior knowledge helped improve the microbiome-based host phenotype prediction. MicroKPNN outperformed fully connected neural network-based approaches in all seven cases, with the most improvement of accuracy in the prediction of type 2 diabetes. MicroKPNN outperformed a recently developed deep-learning based approach DeepMicro, which selects the best combination of autoencoder and machine learning approach to make predictions, in all of the seven cases. Importantly, we showed that MicroKPNN provides a way for interpretation of the predictive models. Using importance scores estimated for the hidden nodes, MicroKPNN could provide explanations for prior research findings by highlighting the roles of specific microbiome components in phenotype predictions. In addition, it may suggest potential future research directions for studying the impacts of microbiome on host health and diseases. MicroKPNN is publicly available at https://github.com/mgtools/MicroKPNN.
期刊介绍:
The intestinal microbiota plays a crucial role in human physiology, influencing various aspects of health and disease such as nutrition, obesity, brain function, allergic responses, immunity, inflammatory bowel disease, irritable bowel syndrome, cancer development, cardiac disease, liver disease, and more.
Gut Microbes serves as a platform for showcasing and discussing state-of-the-art research related to the microorganisms present in the intestine. The journal emphasizes mechanistic and cause-and-effect studies. Additionally, it has a counterpart, Gut Microbes Reports, which places a greater focus on emerging topics and comparative and incremental studies.