Researchers at the Indian Institute of Science Education and Research (IISER) in Bhopal have developed an artificial intelligence (AI) tool to predict how the bacteria in our guts break down food and medicines, potentially helping design personalised diets and treatments for health conditions.
Dubbed ‘GutBug’, the tool uses a combination of machine learning, neural networks, and chemoinformatics to predict the enzymes that may be produced by gut bacteria when they act on molecules found in foods and drugs that are ingested orally.
“GutBug can help us better understand how the food we eat or medicines we consume orally are processed by our gut bacteria and how this affects our health,” said Dr Vineet Sharma, associate professor in the Department of Biological Sciences at IISER Bhopal, and co-founder of the study.
He added that such information can be used to design better diets, develop prebiotics and nutraceutical products, and personalising treatments by understanding how the gut bacteria of an individual will react to particular medications.
Also read
A healthy gut could boost cell therapies for cancer
From gut to lung: how your digestive system affects your breathing
The gut-exercise connection: building a diverse microbiome with physical activity
The IISER researchers hope the tool will help in understanding the complex relationships between hosts and their microbiomes —the colony of trillions of bacteria in our guts. So far, researchers in this field have been challenged given the vastness of the microbiome and that it differs from person to person.
“We used a curated database of 363,872 enzymes from ~700 human gut bacterial strains and a substrate database consisting of 3,457 enzymes to train the AI model,” added Dr Sharma. The researchers said that they had used the largest enzyme substrate database available to train the model.
Bacteria in the gut secrete enzymes to process what we eat and provide the body with various metabolites that are essential for health and body functions. So far, research has shown that the health of the brain, liver, lungs, stomach, skin, and heart can be influenced by the gut microbiome.
The researchers tested the tool with 27 molecules, including complex carbohydrates, flavonoids, and drugs, and it showed an accuracy range of 78 to 97 per cent in its predictions. The results were published in the Journal of Molecular Biology.