DeepDTA is a deep learning model that predicts drug-target (DT) binding affinities using only the sequence information of drugs and targets. The model employs convolutional neural networks (CNNs) to learn representations from the raw sequence data of proteins and compounds, followed by fully connected layers to predict binding affinities. The model was evaluated on two benchmark datasets: the Davis kinase binding affinity dataset and the KIBA large-scale kinase inhibitors bioactivity dataset. The results showed that the DeepDTA model outperformed existing methods such as KronRLS and SimBoost, particularly on the KIBA dataset, achieving a higher Concordance Index (CI) and lower Mean Squared Error (MSE). The model's performance was further validated using various evaluation metrics, including the CI and MSE, as well as the $ r_{m}^{2} $ index and Area Under Precision Recall (AUPR) scores. The study highlights the effectiveness of deep learning in capturing hidden patterns in drug-target interactions and demonstrates that using only sequence information can achieve competitive results compared to methods that rely on 3D structures or 2D features. The model's success underscores the potential of deep learning in drug discovery for predicting binding affinities with high accuracy.DeepDTA is a deep learning model that predicts drug-target (DT) binding affinities using only the sequence information of drugs and targets. The model employs convolutional neural networks (CNNs) to learn representations from the raw sequence data of proteins and compounds, followed by fully connected layers to predict binding affinities. The model was evaluated on two benchmark datasets: the Davis kinase binding affinity dataset and the KIBA large-scale kinase inhibitors bioactivity dataset. The results showed that the DeepDTA model outperformed existing methods such as KronRLS and SimBoost, particularly on the KIBA dataset, achieving a higher Concordance Index (CI) and lower Mean Squared Error (MSE). The model's performance was further validated using various evaluation metrics, including the CI and MSE, as well as the $ r_{m}^{2} $ index and Area Under Precision Recall (AUPR) scores. The study highlights the effectiveness of deep learning in capturing hidden patterns in drug-target interactions and demonstrates that using only sequence information can achieve competitive results compared to methods that rely on 3D structures or 2D features. The model's success underscores the potential of deep learning in drug discovery for predicting binding affinities with high accuracy.