Artificial Intelligence (AI) in microbiome-directed biotherapeutics development

The human gut is home to trillions of microbial lineages, which collectively along with their genomic content form the human microbiome. This community of microbes is crucial to maintaining our health. Imbalance in this community, a state referred to as 'Dysbiosis', has been linked to many diseases. A large domain of research in the human microbiome thus focuses on identifying and administering microbes to restore this imbalance and potentially ameliorate diseases linked to the gut microbiome. These 'microbial medicines', known as live biotherapeutics, are basically living organisms, such as novel types of probiotics or specifically designed groups of bacteria to enhance health. A significant challenge in this context is discovering the right set of microbes. Variability in the baseline gut microbiome even across normal individuals based on demographic factors and context-dependent behavior of specific microbes and strain-specific variations are amongst the various factors that make the microbiome highly individual-specific, resulting in highly personalized responses to different therapeutics. In this, advanced artificial intelligence (AI) derived tools like, Machine Learning (ML) and Deep Learning (DL), can potentially provide major breakthroughs, by facilitating complex analysis of massive amount of microbiome and host OMICs. In this chapter, we discuss ways in which AI can be leveraged to identify patterns "signatures", in microbiome data can facilitate microbiome-derived diagnostics and therapeutics. Using examples, we explore how AI-based models can also look at the complex interactions between microbes and our bodies to discover new, promising bacteria that could be turned into "Live Biotherapeutic Products" (LBPs). This chapter will cover the primary ways of how AI is being utilized, the challenges we still face (such as the need for improved data), and the promising future of using AI to develop a new generation of microbiome-based medicines.