MicroPheno: predicting environments and host phenotypes from 16S rRNA gene sequencing using a k-mer based representation of shallow sub-samples

Abstract

MotivationMicrobial communities play important roles in the function and maintenance of various biosystems, ranging from the human body to the environment. A major challenge in microbiome research is the classification of microbial communities of different environments or host phenotypes. The most common and cost-effective approach for such studies to date is 16S rRNA gene sequencing. Recent falls in sequencing costs have increased the demand for simple, efficient and accurate methods for rapid detection or diagnosis with proved applications in medicine, agriculture and forensic science. We describe a reference- and alignment-free approach for predicting environments and host phenotypes from 16S rRNA gene sequencing based on k-mer representations that benefits from a bootstrapping framework for investigating the sufficiency of shallow sub-samples. Deep learning methods as well as classical approaches were explored for predicting environments and host phenotypes.ResultsA k-mer distribution of shallow sub-samples outperformed Operational Taxonomic Unit (OTU) features in the tasks of body-site identification and Crohn’s disease prediction. Aside from being more accurate, using k-mer features in shallow sub-samples allows (i) skipping computationally costly sequence alignments required in OTU-picking and (ii) provided a proof of concept for the sufficiency of shallow and short-length 16S rRNA sequencing for phenotype prediction. In addition, k-mer features predicted representative 16S rRNA gene sequences of 18 ecological environments, and 5 organismal environments with high macro-F1 scores of 0.88 and 0.87. For large datasets, deep learning outperformed classical methods such as Random Forest and Support Vector Machine.Availability and implementationThe software and datasets are available at https://llp.berkeley.edu/micropheno.Supplementary information Supplementary data are available at Bioinformatics online.

Application of Machine Learning in Microbiology

“…Asgari et al (2018) used shallow subsample representation based on k -mer and deep learning, random forests, and SVMs to predict environmental and host phenotypes from 16S rRNA gene sequencing using the MicroPheno system. They found that the shallow subsample representation based on k -mer is superior to OTU in terms of body location recognition and Crohn’s disease prediction.…”

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

“…ASVs are commonly generated using the Divisive Amplicon Denoising Algorithm 2 (DADA2), and the resultant ASVs represent true biological sequences obtained from reads (Callahan et al, 2016). In addition, there have been recent efforts to use the occurrence of short-chain k-mer (15-30mer) (Molik et al, 2020), and very short-chain k-mers (<10mer) (Asgari et al, 2018(Asgari et al, , 2019, within reads that offer a unique reference-free and alignment-free approach to provide a data representation upon which a phenotype prediction model is built. We have included both of these k-mer approaches in our review to compare them directly with the OTU/ASV assignment methods.…”

EpitopeVec: Linear Epitope Prediction Using Deep Protein Sequence Embeddings

“…Segmentation of biological sequences into bag-of-or sequenceof-overlapping fixed length k-mers is one of the most favorable representations in bioinformatics research. K-mer representations are widely used in the areas of proteomics (Grabherr et al, 2011;Asgari and Mofrad, 2015), genomics (Jolma et al, 2013;Alipanahi et al, 2015), epigenomics (Awazu, 2016;Giancarlo et al, 2015), and metagenomics (Wood and Salzberg, 2014;Asgari et al, 2018).…”