Systematic Comparisons for Composition Profiles, Taxonomic Levels, and Machine Learning Methods for Microbiome-Based Disease Prediction

Abstract

Microbiome composition profiles generated from 16S rRNA sequencing have been extensively studied for their usefulness in phenotype trait prediction, including for complex diseases such as diabetes and obesity. These microbiome compositions have typically been quantified in the form of Operational Taxonomic Unit (OTU) count matrices. However, alternate approaches such as Amplicon Sequence Variants (ASV) have been used, as well as the direct use of k-mer sequence counts. The overall effect of these different types of predictors when used in concert with various machine learning methods has been difficult to assess, due to varied combinations described in the literature. Here we provide an in-depth investigation of more than 1,000 combinations of these three clustering/counting methods, in combination with varied choices for normalization and filtering, grouping at various taxonomic levels, and the use of more than ten commonly used machine learning methods for phenotype prediction. The use of short k-mers, which have computational advantages and conceptual simplicity, is shown to be effective as a source for microbiome-based prediction. Among machine-learning approaches, tree-based methods show consistent, though modest, advantages in prediction accuracy. We describe the various advantages and disadvantages of combinations in analysis approaches, and provide general observations to serve as a useful guide for future trait-prediction explorations using microbiome data.

Whole-genome sequencing and gene sharing network analysis powered by machine learning identifies antibiotic resistance sharing between animals, humans and environment in livestock farming

“…To reduce the number of features analysed by the classifiers a two-step feature selection method was applied: firstly, k-mers with a p-value greater than 0.05 according to the chi-square test were discarded [ 45 , 46 ] and then those remaining were used as input to an Extra Tree Classifier. K-mers with a Gini feature importance above the overall mean were selected [ 47 , 48 ].…”

Conditional forest models built using metagenomic data could accurately predictSalmonellacontamination in Northeastern streams

“…We found that the microbiome data transformation had an inconsistent effect on a model performance, as it improved the AUC of some models (e.g., CF using genus- and family-level data, SVM using family-level data) while it decreased the AUC of other models (e.g., RRF using genus- and family-level data, SVM using genus-level data). Previous studies on gut microbiome data emphasized the importance of data transformation due to the compositionality of the microbiome data (28, 30, 32). However, they also reported that the performance of tree-based algorithms (i.e., random forest and XGBoost) was not significantly affected by data transformation (32).…”

“…Previous studies on gut microbiome data emphasized the importance of data transformation due to the compositionality of the microbiome data (28,30,32). However, they also reported that the performance of tree-based algorithms (i.e., random forest and XGBoost) was not significantly affected by data transformation (32).…”

“…This suggests the importance of selecting and testing multiple algorithms to improve the prediction accuracy. In addition to model selection, the performance of a model may be affected by microbiome data pre-processing, such as data normalization (28). The latter is commonly carried out to account for potential differences in sample library sizes, hence its effect on prediction accuracy needs to be assessed (31)(32)(33).…”

Benchmark of Data Processing Methods and Machine Learning Models for Gut Microbiome-Based Diagnosis of Inflammatory Bowel Disease

“…In this article, we propose a standardized approach for evaluating the performance and generalizability of data processing pipelines and ML models with microbiome data to classify patients with IBD. Previous microbiome ML benchmarking studies focused on performance of various combinations of model type, normalization or transformation, and microbiome compositional features using variations of fivefold cross validation (Song et al, 2020;Topçuoğlu et al, 2020). Five-fold cross validation fails to assess the generalizability to new, unseen sample batches as each split potentially contains samples from all batches present in the dataset.…”