One step closer to knowing the organism (Photo: JUAN MABROMATA via .)
What makes up a human protein, one of the great (immense) mysteries of life, closer than ever to having an answer. Artificial intelligence is behind a fundamental finding that has its own names: DeepMind and the European Molecular Biology Laboratory (EMBL) have used Google’s proprietary AlphaFold system to publish the most complete and accurate database of predictions of human protein structures.
The prestigious journal Nature has echoed this work and describes how these predictions work to provide the most complete picture of the proteins that make up the human proteome (the set of proteins encoded by the human genome).
“Our goal at DeepMind has always been to build AI and then use it as a tool to help accelerate the pace of scientific discovery itself, thereby enhancing our understanding of the world around us,” said DeepMind Founder and CEO Demis. Hassabis.
For the expert, “AlphaFold generates the most complete and accurate image of the human proteome. We believe this represents the most significant contribution AI has made to the advancement of scientific knowledge to date, and is a great example of the kinds of benefits that AI can bring to society. “
We believe this is the most significant contribution artificial intelligence has made to the advancement of scientific knowledge to date Demis Hassabis, founder of DeepMind
This breakthrough, immense in terms of human knowledge, could have a vital effect on health and medicine. Knowing what makes up each protein, its structure and internal functioning can lead to the advancement of various fields and, in the long term, the development of drugs.
The proteins of E. coli, the mouse or the fruit fly have been discovered
The database, open to the scientific community and which will be hosted by the European Bioinformatics Institute (EMBL-EBI), will include around 20,000 proteins expressed by the human genome. Among the first 350,000 published structures, in addition to the human proteome, are the proteins of 20 biologically significant organisms such as E. coli, the fruit fly, the mouse, the zebrafish, the malaria parasite and the tuberculosis bacteria.
The study authors found that the AlphaFold algorithm was able to “confidently” predict the structural position of 58% of the amino acids in the human proteome. Of these, the position of a subset of 35.7% was predicted with a “very high” degree of confidence, which is twice the number covered by the experimental structures, the journal explained.
The researchers clarify that the use of artificial intelligence, with its ability to computationally predict the shape of a protein from its amino acid sequence, does not have to be determined experimentally with the use of laborious and sometimes expensive techniques.
It’s a true revolution for life sciences Edith Herad, CEO of EMBL
AlphaFold has been developed with data from public resources created by the scientific community, so it makes sense for its predictions to be public, said the CEO of EMBL, Edith Herad. For her it is “a true revolution for the life sciences, just as genomics was several decades ago”, and is already being used by the Initiative of Drugs for Neglected Diseases.
The researchers believe that the large-scale and accurate prediction of structures will become “an important tool to address new scientific questions from a structural perspective,” and AlphaFold’s predictions will help further clarify the role of proteins. a group from the University of California at San Francisco has used the predictions of this algorithm to study the biology of SARS-CoV-2.
This article originally appeared on The HuffPost and has been updated.