Autoencoder-based drug–target interaction prediction by preserving the consistency of chemical properties and functions of drugs

Abstract

Motivation Exploring the potential drug-target interactions (DTIs) is a key step in drug discovery and repurposing. In recent years, predicting the probable DTIs through computational methods has gradually become a research hot spot. However, most of the previous studies failed to judiciously take into account the consistency between the chemical properties of drug and its functions. The changes of these relationships may lead to a severely negative effect on the prediction of DTIs. Results We propose an autoencoder-based method, AEFS, under spatial consistency constraints to predict DTIs. A heterogeneous network is established to integrate the information of drugs, proteins and diseases. The original drug features are projected to an embedding (protein) space by a multi-layer encoder, and further projected into label (disease) space by a decoder. In this process, the clinical information of drugs is introduced to assist the DTI prediction. By maintaining the distribution of drug correlation in the original feature, embedding and label space, AEFS keeps the consistency between chemical properties and functions of drugs. Experimental comparisons indicate that AEFS is more robust for imbalanced data and of significantly superior performance in DTI prediction. Case studies further confirm its ability to mine the latent drug-target interactions. Availability The code of AEFS is available at https://github.com/JackieSun818/AEFS. Supplementary information Supplementary data are available at Bioinformatics online.

A novel method for drug-target interaction prediction based on graph transformers model

“…Drugrelated similarity matrices include the drug-drug similarity matrix, drug-disease similarity matrix, drug-side effect similarity matrix and drug similarity matrix. Protein-related (6) e c,ij = W c,e e i,j + b c,e (7)…”

“…With the rapid development of deep learning methods, DTIs methods have been proposed as a deep learning approach for target prediction and drug repurposing in heterogeneous drug-gene-disease networks, which greatly facilitates target identification and advances the process of drug repurposing. Sun et al [6] proposed an autoencoder-based DTI prediction method that projects drug features to the protein space via a multi-layer encoder and then to the disease space via a decoder. Xuan et al [7] proposed methods to integrate multi-scale adjacent topologies, multiple similarities, associations, and drug-and protein-related interactions, which used a fully connected self-encoder learning framework to learn low-dimensional feature representations of nodes in heterogeneous networks, and then applied a multilayer convolutional neural network to generate the final predictions.…”

Drug–target interaction prediction based on protein features, using wrapper feature selection

“…In 15 sequence-based deep learning, 16 deep neural multi-function learning, 17 deep convolution neural networks, 18 light deep convolution neural networks, 19 end-to-end deep learning approaches are applied to predict interactions between drug and target. In using Autoencoders, we can also mention 20 and 21 that were done in 2021.…”

An inductive graph neural network model for compound–protein interaction prediction based on a homogeneous graph

“…the parameters involved in the network propagation algorithms cannot be optimized by the CPI prediction task [ 26 ]. In recent years, graph neural networks (GNNs) have been utilized in extracting representations for heterogeneous graphs, such as graph convolutional networks in NeoDTI [ 23 ], and graph convolutional autoencoders and generative adversarial networks (GANs) in GANDTI [ 28 ]. Deep models have shown stronger performance than these two-step methods in CPI prediction.…”