Tokyo, July 6 (SocialNews.xyz) For the first time, researchers at the University of Tokyo used a special artificial intelligence (AI), called Bayesian Neural Networks, to investigate datasets of intestinal bacteria, to find relationships that current analytical tools cannot reliably identify.
Intestinal bacteria are known to be an important factor in many health-related concerns. The human body is made up of approximately 30 to 40 trillion cells, while the intestines contain approximately 100 trillion intestinal bacteria.
“The problem is that we are beginning to understand which bacteria produce which human metabolites and how these relationships change in different diseases,” said Prince, a project researcher at the Faculty of Biology Sciences, Hounda Lab, in a paper published at a biology briefing.
Accurately mapping these bacterial chemical relationships could potentially develop personalized treatments, Dan said. “Imagine that you can grow certain bacteria to produce beneficial human metabolites, or you can design targeted therapies that will modify these metabolites to treat diseases.”
This system, Vbayesmm, automatically distinguishes key players whose metabolites are heavily influenced from the microbial background where they are less relevant, acknowledging uncertainty about the predicted relationship rather than providing an overly confident but potentially incorrect answer.
“When tested with real data from sleep disorders, obesity and cancer studies, our approach consistently prioritized existing methods, providing the confidence to identify specific bacterial families that are consistent with known biological processes and discover real biological relationships rather than meaningless statistical patterns,” explained Dan.
Vbayesmm can handle and communicate uncertainty issues, which gives researchers more confidence than tools that do not. Although systems are optimized to deal with heavy analytical workloads, mining such a huge dataset is computationally expensive. However, over time, this becomes an increasingly barrier for those who want to use it.
“We plan to use a more comprehensive chemical dataset that captures the full range of bacterial products, which creates a new challenge to determine whether chemicals come from external sources like bacteria, the human body, or diet,” Dan said.
– Anne
Na/
Source: IANS
About Gopi
Gopi Adumilli is a programmer. He is the editor of SocialNews.xyz and the president of Agk Fire Inc.
He publishes news articles about website design, mobile application development, and current events from a variety of certified news sources.
As for writing, he likes to write about current world politics and Indian films. His future plans include the development of socialnews.xyz, which is developed on a news website with no bias or judgment.
He can contact gopi@socialnews.xyz
