Caporaso, J. G. | Search bioRxiv

bioRxiv2018

Species-level microbial sequence classification is improved by source-environment information

Popular naive Bayes taxonomic classifiers for amplicon sequences assume that all species in the reference database are equally likely to be observed. We demonstrate that classification accuracy degrades linearly with the degree to which that assumption is violated, and in practice it is always violated. By incorporating environment-specific taxonomic abundance information, we demonstrate that species-level resolution is attainable.

bioinformatics↗

bioRxiv2018

q2-sample-classifier: machine-learning tools for microbiome classification and regression

Microbiome studies often aim to predict outcomes or differentiate samples based on their microbial compositions, tasks that can be efficiently performed by supervised learning methods. Here we present a benchmark comparison of supervised learning classifiers and regressors implemented in scikit-learn, a Python-based machine-learning library. We additionally present q2-sample-classifier, a plugin for the QIIME 2 microbiome bioinformatics framework, that facilitates application of the scikit-learn classifiers to microbiome data. Random forest, extra trees, and gradient boosting models demonstrate the highest performance for both supervised classification and regression of microbiome data. Automated feature selection and hyperparameter tuning enhance performance of most methods but may not be necessary under all circumstances. The q2-sample-classifier plugin makes these methods more accessible and interpretable to a broad audience of microbiologists, clinicians, and others who wish to utilize supervised learning methods for predicting sample characteristics based on microbiome composition. The q2-sample-classifier source code is available at https://github.com/qiime2/q2-sample-classifier. It is released under a BSD-3-Clause license, and is freely available including for commercial use.

bioinformatics↗

bioRxiv2017

Lack of Evidence that Ursodeoxycholic Acid’s Effects on the Gut Microbiome Influence Colorectal Adenoma Risk

ObjectiveWe previously reported that Ursodeoxycholic acid (UDCA), a therapeutic bile acid, reduces risk for advanced colorectal adenoma in men but not women. Interactions between the gut microbiome and fecal bile acid composition as a factor in colon cancer neoplasia have been postulated but evidence is limited to small cohorts and animal studies.\n\nDesignUsing banked stool samples collected as part of a phase III randomized clinical trial of UDCA for the prevention of colorectal neoplasia, we compared change in the microbiome composition after 3 years intervention in a subset of participants randomized to 8-10 mg/kg of body weight UDCA (n=198) to placebo (n=203). UDCA effects on the microbiome, sex and adenoma outcome were investigated.\n\nResultsStudy participants randomized to UDCA experienced compositional changes in their microbiome that were statistically more similar to other individuals in the UDCA arm than to those in the placebo arm. This change reflected an UDCA-associated shift in microbial community distance metrics (P <0.001), independent of sex, with no evidence of UDCA effect on microbial richness (P > 0.05). These UDCA-associated shifts in microbial community distance metrics from baseline to end-of-study were not associated with risk of any or advanced adenoma (all P> 0.05) in men or women.\n\nConclusionDespite a large sampling of randomized clinical trial participants, daily UDCA use only modestly influenced the relative abundance of microbial species in stool with no evidence for effects of UDCA on stool microbial community composition as a modifier of colorectal adenoma risk.\n\nSUMMARYO_ST_ABSWhat is already known about this subject?C_ST_ABSO_LIUrsodeoxycholic acid (UDCA) is a therapeutic bile acid used in the treatment of primary biliary cirrhosis (PBC) and investigated for anti-cancer activity in the colon\nC_LIO_LIIn humans, UDCA is produced in the colon from the conjugation of primary bile acids by intestinal bacteria\nC_LIO_LIIntestinal bacteria play a critical role in human intestinal health and disease including a hypothesized role in the development of colorectal cancer.\nC_LIO_LIUDCA was found to reduce the risk of more advanced colorectal adenoma with effects present in men but not women.\nC_LIO_LITherapeutic UDCA was recently shown to reduce the extent of bacterial dysbiosis in patients with PBC\nC_LI\n\nWhat are the new findings?O_LIAmong a population of patients with colorectal adenoma, low dose oral UDCA taken daily produced modest changes in fecal bacterial composition\nC_LIO_LIUDCA associated changes in the gut microbiome were similar in men and women.\nC_LIO_LIUDCA associated changes in the gut micobiome were not associated with risk of any or advanced colorectal adenoma in the patient population.\nC_LI\n\nHow might it impact on clinical practice in the foreseeable future?O_LIThese findings confirm effects of oral UDCA on the microbiome that may be beneficial for patients with PBC.\nC_LIO_LIThese findings suggest that the anti-cancer effects of UDCA for colorectal adenoma prevention are not due to major effects of UDCA on the gut microbiome.\nC_LI

cancer biology↗

bioRxiv2017

q2-longitudinal: a QIIME 2 plugin for longitudinal and paired-sample analyses of microbiome data

Studies of host-associated and environmental microbiomes often incorporate longitudinal sampling or paired samples in their experimental design. Longitudinal sampling provides valuable information about temporal trends and subject/population heterogeneity, offering advantages over cross-sectional and pre/post study designs. To support the needs of microbiome researchers performing longitudinal studies, we developed q2-longitudinal, a software plugin for the QIIME 2 microbiome analysis platform (https://qiime2.org). The q2-longitudinal plugin incorporates multiple methods for analysis of longitudinal and paired-sample data, including paired differences and distances, linear mixed effects models, microbial interdependence test, first differencing, and volatility analyses. The q2-longitudinal package (https://github.com/qiime2/q2-longitudinal) is open source software released under a BSD-3-Clause license and is freely available, including for commercial use.

bioinformatics↗