Determining the three-dimensional shape of biological molecules is the most challenging problem in modern biology.
The authors of the new work used the methods of machinelearning and developed an approach that solves this problem, as it predicts their exact structure. The researchers note that their approach can be applied even to molecules and structures that are most difficult to determine experimentally.
Algorithm of graduate students predicts accuratemolecular structures, and on the basis of this, one can understand how they work in different fields: this will help both in basic biological research and in the development of drugs.
Proteins - these are such machines in the molecularscale, which perform all sorts of functions. In order to do something, they can bind to other proteins. If you know that a pair of proteins are involved in a disease, and also understand how they both interact, then you can try to make a drug that will hit the same target.
Stefan Eismann, PhD student at Stanford University
AI learned to find fundamental conceptsproteins that are key to the formation of molecular structure. And, importantly according to the authors, it was not pre-loaded with data on specific proteins, which could make the algorithm biased towards them and confuse the global analysis.
Therefore, the algorithm finds data and characteristics of proteins that scientists were not aware of before.
During the experiment, the algorithm successfully copedwith proteins and graduate students tested it on RNA molecules. As a result, the AI managed the puzzle in every case, although it was not designed specifically for RNA structures.
To read Further:
A spaceship several kilometers away: everything that is known about China's new project
COVID-19 mutates, and vaccines are modernized: how to deal with new strains
A supercapacitor the size of a speck of dust has appeared: it is 3 thousand times smaller than its analogues