New AI tool helps scientists identify bacteria hidden in soil
A handful of soil holds a crowded world of bacteria, fungi and other organisms. Scientists can collect their DNA, but much of it remains difficult to identify. Now, researchers at the Hebrew University of Jerusalem and partner institutions have developed an artificial intelligence tool that helps make sense of this genetic puzzle.
Scientists have developed an AI tool called BBERT to help identify bacteria hidden in soil samples. The tool learns patterns in bacterial DNA, allowing it to recognize fragments as short as 100 DNA letters. In tests, BBERT distinguished bacterial from nonbacterial DNA with about 90% accuracy, and identified protein-making instructions correctly 88% of the time.
Applied to 1,969 soil DNA data sets, BBERT identified many fragments that a commonly used comparison tool had left unclassified. The tool also helped reconstruct bacterial genomes from short DNA fragments faster and with fewer errors. While BBERT can miss some bacterial fragments or mistake other DNA for bacterial DNA, it could make exploring hidden soil life more accessible for researchers.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
