Biological sequence retrieval method and device, electronic equipment and storage medium

Abstract

The invention provides a biological sequence retrieval method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a known biological sequence; taking the known biological sequence as initial state information of a reinforcement learning network, and performing at least one time of site variation on the known biological sequence by using the reinforcement learning network to obtain a variation biological sequence after each time of site variation; and determining a target biological sequence based on the variation biological sequence after each time of site variation. According to the method, the characteristics of the selected variant biological sequence are better by increasing the number of reinforcement learning training times, so that the generated target biological sequence can better meet the requirements of the biological field.

Description

technical field [0001] The present disclosure relates to the field of computer technology, in particular, to a biological sequence retrieval method, device, electronic equipment and storage medium. Background technique [0002] The search and prediction of biological sequences is an important issue in synthetic biology. The search and prediction based on protein amino acid sequence has broad application prospects in the fields of protein interaction and cell engineering. The prediction of nucleic acid (Ribonucleic Acid, RNA) sequence plays an important role in virus mutation and vaccine research. [0003] In recent years, artificial intelligence technology has begun to be applied to the biological field. In related technologies, most of them follow the idea of ​​supervised learning, construct a sequence prediction model by using the characteristics of a small number of biological sequences known to humans, and then use heuristic algorithms or evolutionary algorithms to sear...

AI Technical Summary

Problems solved by technology

[0004] However, the above methods usually cannot take advantage of the big data and high parallel computing advantages of current artificial intelligence, and have high requirements for the encoding and decoding of biological sequences and the establishment of sequence prediction models. The search logic is relatively simple, compared with random search Limited performance improvement

On the other hand, verifying the correctness of these sequences requires biological experiments, and the time period is long, making it difficult to conduct large-scale experiments without counting

Images