18/06/2026
Every time scientists sequence a genome or measure proteins in a diseased cell, they generate a list that can run to tens of thousands of entries. The hard part has always been working out which entries actually matter.
Researchers at ARMI have built a tool called ExIR that does exactly that. Unlike existing approaches, ExIR does not need to consult any database of prior knowledge to get there. It reads the data in front of it and identifies the molecules most likely to be driving a disease, flagging it, or connecting the two. Tested across more than a dozen published datasets covering multiple cancers and neurological conditions, it outperformed every comparable tool.
Read the full story via the link in the comments.