Sampling learning based protein-ligand binding site prediction method

Abstract

The invention provides a sampling learning based protein-ligand binding site prediction method. The method comprises the steps of: firstly, utilizing PSI-BLAST and PSIPRED programs to obtain evolutionary information and secondary structure information of protein, and using a slide window technology to extract characteristics of each amino acid residue (sample); secondly, utilizing a random down-sampling technology to perform random down-sampling on non-binding site samples, and using obtained non-binding site sample subsets and binding site sample set to train an SVM for predicting all to-be-predicted samples; thirdly, according to characteristic information of each to-be-predicted sample, utilizing a KNN dynamic sampling learning technology to perform sampling learning on binding site samples and the non-binding site samples respectively, and combining binding site sample subsets and the non-binding site sample subsets after sampling to train a specific SVM for predicting the to-be-predicted samples; and finally, using a threshold based integration technology to integrate the two trained SVMs. The method has the advantages that: firstly, the use of the random down-sampling and KNN dynamic sampling learning technologies can effectively reduce the scale of training sets and accelerate the model training speed; secondly, the use of the KNN dynamic sampling learning technology can train different SVM models for different to-be-predicted samples and effectively infuse the difference among the to-be-predicted samples; and thirdly, the use of the SVM integration technology effectively reduces the information loss caused by sampling learning and improves the model prediction precision.

Description

technical field [0001] The present invention relates to the field of bioinformatics prediction of protein-ligand binding sites, in particular to a method for predicting protein-ligand binding sites based on sampling learning, especially a method based on random downsampling and KNN dynamic sampling A method for predicting protein-ligand binding sites with high precision using learning techniques and support vector machine ensemble strategies. Background technique [0002] In life activities, large and small ligands play an indispensable role, such as adenosine triphosphate (ATP), vitamins, etc.; among them, ATP is an important biological macromolecule, which is important for membrane transport, muscle contraction, Signal transmission, cell movement, DNA replication and transcription, and other life activities are of great significance. Most of these ligands interact with proteins through protein-ligand binding sites, and perform various biochemical functions by means of pro...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above-mentioned problem of protein-ligand binding site prediction due to the lack of generality caused by the lack of ligand types, the differences between different samples to be predicted have not been fully considered, resulting in a large gap between the prediction accuracy and the actual application. Due to the disadvantage of poor interpretability, the purpose of the present invention is to propose a protein-ligand binding method based on sampling learning with high prediction accuracy and strong model interpretability, which combines random downsampling, KNN dynamic sampling learning and integration technology. Anchor Point Prediction Method

Images