Anticancer peptide prediction method based on bidirectional long-short-term memory network and feature fusion

Abstract

The invention relates to the technical field of anticancer peptide prediction, in particular to an anticancer peptide prediction method based on a bidirectional long-short-term memory network and feature fusion, which comprises the following steps: reading four reference peptide sequence data sets, and carrying out amino acid composition analysis on the data sets; feature extraction is carried out on the data set through Bi-LSTM, and a Bi-LSTM feature vector is generated; performing feature extraction on the five amino acid feature vectors through a full-connection neural network; and carrying out feature fusion on the feature vectors through a Concatenate algorithm, obtaining a probability score through a full connection layer with a unit 1 and a Sigmoid activation function, and distinguishing the probability score into an anti-cancer peptide and a non-anti-cancer peptide through the score. According to the method, the anticancer peptide prediction with high accuracy, high Marius correlation coefficient, high sensitivity, high specificity and high ROC curve area is realized.

Description

technical field [0001] The invention relates to the technical field of anti-cancer peptide prediction, in particular to an anti-cancer peptide prediction method based on bidirectional long short-term memory network and feature fusion. Background technique [0002] The discovery of anticancer peptide (ACP) has broadened people's horizons of anticancer road, its specificity and tumor cannot develop resistance to it, solved some side effects caused by traditional anticancer therapy, and hopefully become a kind of cancer Anticancer peptides are usually composed of 5-40 amino acids; in order to further understand the mechanism of action of anticancer peptides, there have been many biological experimental methods for the identification of anticancer peptides. For example, Vidal et al. identified peptide cocktails targeting intracellular tumor proteins by the yeast two-hybrid system, and Peelle et al. identified novel targeting peptides that were not cell-type specific by screening...

AI Technical Summary

Problems solved by technology

For example, Vidal et al. identified peptide cocktails targeting intracellular tumor proteins through the yeast two-hybrid system, and Peelle et al. discovered novel targeting peptides that are not cell-type specific through mammalian cell screening; however, these identification methods are time-consuming and expensive , is very complex and difficult to achieve in a high-throughput manner, so rapid and effective identification of anticancer peptides is particularly important

[0003] Wu et al. proposed a PTPD model, using feature vectors extracted by k-mer and Word2vec (word vectors), and input them into a convolutional neural network (CNN) to predict peptides; Rao et al. applied graph convolutional networks (GCN) to anti-cancer In the prediction of peptides, the ACP-GCN model is proposed; however, these deep learning methods only consider the original sequence information and physical and chemical properties of amino acids, ignoring the long-term correlation information of anti-cancer peptides at the time level, and cannot be low-cost, fast and efficient. anticancer peptide

Images