Predicting Drug-Target Interactions with Electrotopological State Fingerprints and Amphiphilic Pseudo Amino Acid Composition

Abstract

The task of drug-target interaction (DTI) prediction plays important roles in drug development. The experimental methods in DTIs are time-consuming, expensive and challenging. To solve these problems, machine learning-based methods are introduced, which are restricted by effective feature extraction and negative sampling. In this work, features with electrotopological state (E-state) fingerprints for drugs and amphiphilic pseudo amino acid composition (APAAC) for target proteins are tested. E-state fingerprints are extracted based on both molecular electronic and topological features with the same metric. APAAC is an extension of amino acid composition (AAC), which is calculated based on hydrophilic and hydrophobic characters to construct sequence order information. Using the combination of these feature pairs, the prediction model is established by support vector machines. In order to enhance the effectiveness of features, a distance-based negative sampling is proposed to obtain reliable negative samples. It is shown that the prediction results of area under curve for Receiver Operating Characteristic (AUC) are above 98.5% for all the three datasets in this work. The comparison of state-of-the-art methods demonstrates the effectiveness and efficiency of proposed method, which will be helpful for further drug development.

Quantitative prediction model for affinity of drug–target interactions based on molecular vibrations and overall system of ligand-receptor

“…The three molecular forces, dispersion, dipole moment and hydrogen bonding, which influence the strength of DTIs affinity, are closely related to the electronic relationships characterized by E-state descriptors [ 49 , 50 ]. Due to their above natures, E-state descriptors have been widely used in the analysis of DTIs [ 51 ]. This suggests that E-state descriptors are a good choice for analyzing and predicting DTIs affinity.…”

Comparative Studies on Resampling Techniques in Machine Learning and Deep Learning Models for Drug-Target Interaction Prediction

“…To evaluate the performance of each of the machine learning and deep learning models with or without resampling strategies, we have used the scikit-learn library to compute the performance metrics of the models, mainly the Accuracy, Precision, Recall and F1 values. The formula for accuracy is (Equation (3)): where TP means True Positive (the number of drug-target pairs predicted as interactions correctly), TN stands for True Negative (the number of negative pairs predicted as non-interactions correctly) [ 62 ], FP means False Positive (the number of negative drug-target pairs classified as interactions incorrectly, and FN means False Negative (the number of positive drug-target pairs classified as non-interactions incorrectly) [ 62 ]. Hence, the accuracy value of a model means the number of correct predictions of interactions over the total number of predictions.…”

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

“… Most machine learning-based methods have poor descriptive features. Therefore, it is difficult to distinguish a potential drug mechanism from its function considering a pharmacological perspective 31 , 32 . …”