The human gut microbiome (HGM), which is made up of more than 3.3 million distinct genes from more than 1,000 different bacterial species in the human gut, gives the host extra metabolic activity so that it can metabolise substances that are not broken down by the host. HGM is a potential target to modulate health and illness because it breaks down ingested food to produce a wide variety of microbial metabolites that are essential to human physiology and health.
Due to the vast diversity and complexity of the gut microbiome, it is difficult to experimentally determine the metabolism of compounds taken in orally. To solve this issue, Dr Vineet Sharma and his team from IISER Bhopal (Indian Institute of Science Education and Research) created ‘GutBug’, a web-based tool that predicts all potential bacterial metabolic enzymes that may be able to biotransform biotic and xenobiotic compounds. Bioactive substances including polyphenols, oligosaccharides (fructo- and galacto-oligosaccharides), pigments, etc. are abundant in functional foods. They are indicated as nutraceuticals with anti-tumor, anti-inflammatory, and antioxidant activities that also alter gut microbiota by lowering pathogenic bacteria and increasing the abundance of good bacteria, although they have a low bioavailability. Thus, consuming probiotics and prebiotics together to stimulate the growth of gut bacteria that benefit the host, is now receiving attention as a tool against several metabolic disorders.
‘GutBug’, provides information on specific bacterial enzymes, reactions, and bacteria involved in the process of digestion and absorption of nutrients by the human intestine.
The study of complex host–microbial relationships is challenging given the vastness of the microbiome and the varying bacterial collections from individual to individual. The researchers say that the ‘GutBug’ helps identify potential bacterial enzymes and bacterial strains for enhanced metabolism.
Researchers say that the number of bacterial cells in the gut is greater than the number of cells in the human body. The large number of bacterial cells and diversity in the bacterial communities of the gut provide additional metabolic capacity, which can metabolise drugs and undigested food. Bacterial metabolism of orally taken drugs or nutraceuticals affects their efficacy and toxicity, which can harm the body.
Using computational methods such as machine learning, it is possible to gain deeper insight into how specific bacterial species in the gut function. It may also help to understand how this interaction can positively or negatively affect host microorganisms in the human body.
Researcher Dr Vineet Sharma, IISER Bhopal said, “The new ‘GutBug’ tool uses a combination of machine learning, neural networks and chemoinformatic methods. A curated database of 363,872 enzymes from approximately 700 bacterial strains of the human gut and a substrate database containing 3,457 enzymes were used to train the AI model.”
The device has been tested on 27 different molecules, with success rates ranging from 0.78 to 0.97. “GutBugs can help us better understand how the food we eat, or the drugs we take orally, are processed by our gut bacteria, and how this can affect our health,” said Dr Sharma, who was assisted by Aditya S Malve and Gopal N Srivastava.
