DTA

DeepDTA is a software tool based on deep learning designed to predict drug-target interaction (DTI) binding affinities using only the sequence information of both targets and drugs. The prediction of DTI binding affinities is a crucial component of the drug discovery process, and computational methods have become more prevalent in recent years. Most of the computational methods proposed for predicting DT interactions have focused on binary classification, where the goal is to determine whether a DT pair interacts. However, protein-ligand interactions assume a continuum of binding strength values, also called binding affinity, and predicting this value remains a challenge.

One of the main advantages of DeepDTA is that it requires only the sequence information of both targets and drugs, which makes it a more efficient and cost-effective method for predicting DTI binding affinities. Most existing methods use either 3D structures of protein-ligand complexes or 2D features of compounds, which can be time-consuming and expensive. Another novel approach used in this software tool is the modeling of protein sequences and compound 1D representations with convolutional neural networks (CNNs).

The proposed model is effective for DTI binding affinity prediction, and it outperforms the KronRLS algorithm and SimBoost, a state-of-the-art method for DT binding affinity prediction. The model in which high-level representations of a drug and a target are constructed via CNNs achieved the best Concordance Index (CI) performance in one of our larger benchmark datasets. The results of this study demonstrate that DeepDTA is an effective approach for drug target binding affinity prediction.