Prokaryotic acetylation site prediction method based on information fusion and deep learning

Abstract

The invention discloses a prokaryotic acetylation site prediction method based on information fusion and deep learning, and relates to the technical field of biological information. According to the prediction method, multi-information fusion is introduced, feature coding is carried out on protein sequences from multiple aspects of sequence information, physicochemical information and evolution information, and the difference between acetylation site and non-acetylation site protein sequences is fully reflected. The original special diagnosis space is screened and optimized by Group Lasso, important features of an acetylation site recognition model are reserved, the optimal feature subset is obtained, and the model calculation speed and quality are improved. The deep neural network generates features with discrimination through hierarchical learning of the optimal feature subset, the acetylation site and the non-acetylation site in the protein sequence are effectively distinguished, the calculation time is saved at the same time, other costs and related limitations are avoided, deep understanding of an acetylation molecular mechanism can be facilitated, and valuable reference information can be provided for experimental verification related to acetylation site recognition.

Description

technical field [0001] The invention relates to the technical field of biological information, in particular to a prokaryotic acetylation site prediction method based on information fusion and deep learning. Background technique [0002] Protein post-translational modification is a regulatory mechanism that plays an important role in normal and pathological cell physiology. At present, hundreds of protein post-translational modification types have been discovered, among which the regulation of lysine acetylation on metabolism is one of the important advances in the field of post-translational modification research in recent years. The process of covalently attaching acetyl groups to lysine residues enzymatically. That is, the acetyl group is covalently linked to a specific lysine residue by lysine acetyltransferase, and the acetyl group is removed by lysine deacetylase. [0003] Acetylation modification is extremely conserved during the evolution of life, and the regulatio...

AI Technical Summary

Problems solved by technology

Secondly, when predicting acetylation sites, a single feature extraction algorithm cannot effectively extract protein sequence information, and does not integrate multiple feature information, which underestimates the importance of multi-information fusion

Finally, we found that classifiers for acetylation site prediction are limited by support vector machines, random forests, logistic regression, etc., with limited prediction accuracy

Images