viernes, 21 de febrero de 2020

Beyond the hype, machine learning might yet find a usable drug

The Readout
Damian Garde & Meghana Keshavan

Beyond the hype, machine learning might yet find a usable drug

With the caveat that it has been tested only in mice, scientists have found a potential treatment for deadly superbugs, and they owe it to artificial intelligence.

As STAT’s Casey Ross reports, researchers at the Massachusetts Institute of Technology used machine learning to comb a database of drugs in search of something that might kill drug-resistant bacteria. They got a surprising result: a molecule meant to treat diabetes, structurally distinct from any existing antibiotic. And then they tested it in mice, and it effectively treated two drug-resistant infections.

The molecule, named in honor of the less-than-benevolent AI in "2001: A Space Odyssey," has a long road ahead, but its very discovery bodes well for the future of machine learning in medicine, researchers said.

“Now we’re finding leads among chemical structures that in the past we wouldn’t have even hallucinated that those could be an antibiotic,” said Nigam Shah, professor of biomedical informatics at Stanford University. “It greatly expands the search space into dimensions we never knew existed.”

Read more.

No hay comentarios: