Thirty-year drug company veteran Derek Lowe slams AlphaFold: "Pure self-importance" to make drugs based on structural predictions

DeepMind recently announced the latest progress of AlphaFold: it has predicted 214 million protein structures of more than 1 million species, covering almost all known proteins on earth, refreshing our expectations for it once again.

The enthusiastic scene when AlphaFold 2 was launched was recreated and once again sparked hot discussions on social media both at home and abroad. However, researchers in the life science field, who are "insiders", have mixed feelings about the results announced by AlphaFold.

Previously, several scholars in the field of bioinformatics, such as Pan Yi, Zhou Yaoqi and Xu Dong, said that the massive amount of data in the AlphaFold Protein Structure Database updated this time had problems such as the structure of some results was unstable and could not be applied to research.

Professor Tang Jian from the MILA laboratory at the University of Montreal, Canada, also told Medical Health AI Nuggets that the proteins predicted by AlphaFold have limited impact on drug development.

Recently, Dr. Derek Lowe, a senior expert in the pharmaceutical industry in the United States, posted an article to tattle on AlphaFold, which is being hailed by the outside world, especially by the media.

A graduate of Duke University, Dr. Derek Lowe spent more than three decades working for major pharmaceutical companies on drug discovery programs for schizophrenia, Alzheimer's, diabetes, osteoporosis, and other diseases.

Last week, Derek Lowe posted an article on the Royal Society of Chemistry's website. He made a clear statement: AlphaFold will not revolutionize drug discovery.

The following article by Derek Lowe has been compiled by Thunderbird without altering the original meaning.

Protein structure prediction has long been considered one of the most difficult problems in computational biology.