Computer Vision News - September 2021

3 Summary 25 Deep Learning and Microbiome The human microbiota consists of the 10-100 trillion symbiotic microbial cells which can be found in each human’s gut (mostly); the human microbiome consists of all those genes living inside the human cells. That’s the reason many different projects have been started to understand the role and its impact on human health (in a primary level) but also use them as a guide of what’s happening inside a human body. The drastic reduction in sequencing costs experienced over the past few years has made it possible to identify specific microbial taxa found within the human gut that are difficult or impossible to culture. Researchers are now able to generate millions of sequences per sample to assess differences in microbial communities between body sites and individuals. You can readmore and a bit of historical info, in the great article “Defining the Human Microbiome” by Luke K Ursell et al in doi: 10.1111/j.1753-4887.2012.00493.x Role Previous research on the microbiota has shown to be useful in the following distinct ways: 1. Aids the human digestive system as it can be very efficient in extracting energy from food and harvesting nutrients, as microbiota provides humans with enzymes and biochemical pathways produced by versatile metabolic microbial genes that are far more than found in human genome. 2. Protects the system against invasive pathogens by providing a physical barrier, producing antimicrobial substances, or involving in competitive exclusion. 3. It is essential in the induction, training, and function of the host immune system. Modern regression techniques have been applied to solve the problem of microbiome prediction using the high dimensional features. They usually incorporate a sparse penalty they tend to be computationally expensive and require a significant parameter tuning. More importantly, they are limited in exploring the phylogenetic relationship among taxa. To explain, with some diseases (e.g., inflammatory bowel disease) the overall composition of the microbiota is significant rather than granular marker taxa. The opposite is true for diseases like