Rapid detection of microbiota cell type diversity using machine-learned classification of flow cytometry data

Birge D Özel Duygan,Ambrin Farizah Babu,Markus Seyfried,Noushin Hadadi,Jan R Van Der Meer

doi:10.1038/s42003-020-1106-y

Abstract

The study of complex microbial communities typically entails high-throughput sequencing and downstream bioinformatics analyses. Here we expand and accelerate microbiota analysis by enabling cell type diversity quantification from multidimensional flow cytometry data using a supervised machine learning algorithm of standard cell type recognition (CellCognize). As a proof-of-concept, we trained neural networks with 32 microbial cell and bead standards. The resulting classifiers were extensively validated in silico on known microbiota, showing on average 80% prediction accuracy. Furthermore, the classifiers could detect shifts in microbial communities of unknown composition upon chemical amendment, comparable to results from 16S-rRNA-amplicon analysis. CellCognize was also able to quantify population growth and estimate total community biomass productivity, providing estimates similar to those from 14C-substrate incorporation. CellCognize complements current sequencing-based methods by enabling rapid routine cell diversity analysis. The pipeline is suitable to optimize cell recognition for recurring microbiota types, such as in human health or engineered systems.

Highlights

The study of complex microbial communities typically entails high-throughput sequencing and downstream bioinformatics analyses
We test to what extent CellCognize can analyze microbial cell diversity of unknown communities and changes thereof, imposed by addition of low concentrations of phenol or 1-octanol, which we compare with data from 16S rRNA gene amplicon diversity analysis and further interpret based on the probability of class assignment
We developed a pipeline (CellCognize) using a supervised artificial neural network (ANN), which classifies cell types in microbial community samples based on flow cytometry (FCM) multiparametric signature similarities with a predefined set of standards (Fig. 1)

Summary

Introduction

The study of complex microbial communities typically entails high-throughput sequencing and downstream bioinformatics analyses. We present a new pipeline to facilitate microbiota diversity analysis by providing rapid absolute quantification of cells and recognition of cell type diversity and physiology based on flow cytometry data. Our results demonstrate the ability to recognize and quantify known microbial cell types, their physiology and growth, amidst a known or unknown community background, and even infer community diversity changes in unknown microbial communities This suggests that with appropriate standards and optimization, CellCognize can reliably analyze recurring microbiota cell types in environmental, animal or clinical settings, thereby considerably accelerating and simplifying microbiota studies

Methods

Results

Conclusion