The future of microbiome analysis: Biosensor methods for big data collection and clinical diagnostics

The invisible realm of the human microbiota contains patterns that, when properly detected and interpreted, could indicate much about the health or disease of its host, the human body. Biosensing techniques for the detection of the human microbiota have the potential to transform clinical diagnostics, yet point-of-care (POC) biosensors for direct detection of disturbances in microbial communities are not presently available in clinical settings. The objective of this review paper is to explore the potential for biosensors to usher the study of the microbiome into the spaces of clinical diagnostics and big data collection. To achieve this goal, we first outline the types of biosensor methods that have been used to detect multiple targets from clinical and field samples, discuss the challenges inherent in multiplex detection from complex samples and examine the potential for biosensors to integrate microbiome analysis with the diagnostic process. We then consider the potential pitfalls of biosensor-based microbiome analysis and highlight the anticipation for machine-learning techniques to address the unique challenges associated with the large variability in microbiota composition between individuals. We finally conclude that biosensor technologies with integrated machine learning algorithms will shape the future of microbiome analysis by allowing for acquisition of vast amounts of microbiome data that can eventually be harnessed in clinical settings for more rapid and accurate diagnoses.