cs.LGAug 13, 2026

Recent Advances in Deep Learning-Based Drug-Target Binding Affinity Prediction

Authors: Jafin Khan, Md Hossain Shuvo

Organizations: Department of Computer Science, Prairie View A&M University, Prairie View, Texas, USA.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Computational approaches to drug discovery involve multiple sub-problems, and among them, drug-target binding affinity prediction plays an important role. Despite recent advances, accurately predicting binding affinity remains an open research area. The major objective of our paper is to perform a comprehensive review and comparative analysis of recent machine learning methods for drug-target binding affinity prediction, with a focus on identifying strengths, limitations, and research gaps. We review representative recent deep learning approaches that use common benchmark datasets and evaluation metrics, covering a range of neural network architectures and representation strategies. In addition, we analyze seven widely used benchmark datasets and commonly adopted evaluation metrics for drug-target binding affinity prediction. Our analysis indicates that although many methods report strong performance on standard benchmarks, their effectiveness is often influenced by dataset bias and limited evaluation settings. Furthermore, most methods exhibit reduced performance in cold-start scenarios, highlighting challenges in generalization. We identify several limitations of current approaches, including dataset imbalance, the lack of standardized evaluation, limited real-world applicability, and challenges in cold-start scenarios. We also discuss future research directions, including better dataset design, more robust evaluation methods, improved handling of cold-start problems, and the integration of multimodal representations.

Figures & tables

Figure 1: General process of drug–target binding affinity prediction: (A) Input of drug and protein information (e.g., SMILES and sequences), (B) Learning patterns using deep learning models, (C) Prediction of binding affinity values.
Dataset# Proteins# Drugs/Ligands# RecordsBinding Affinity Range
DAVIS3796830,056pKd: 5.0–10.80
KIBA (Original)46752,498243,251KIBA score: −3.10-3.10 –17.80
PDBbind (Total)19,44315,47719,443pKd/pKi: 0.40–15.22
PDBbind (General Set)14,12711,95314,127pKd/pKi: 0.40–15.22
PDBbind (Refined Set)5,3164,2035,316pKd/pKi: 2.00–11.92
Metz1701,42335,259Ki: 4.0–11.1
Table 1: Summary of commonly used drug–target binding affinity datasets.
Figure 2: Length and distribution statistics across the collected drug–target binding affinity datasets. (A) Distribution of drug SMILES lengths, where circles indicate minimum and maximum values and triangles denote average length. (B) Distribution of target protein sequence lengths, where circles indicate minimum and maximum values and triangles denote average length. (C) Distribution of binding affinity values across datasets.
Figure 3: (A) Pairwise sequence identity among protein target sequences across commonly used drug–target affinity benchmark datasets (B) Frequency of drug and target feature representations utilized in existing drug–target binding affinity prediction methods. The left side represents drug-related features, while the right side represents target/protein-related features.
CategoryMethod NameNeural Network ArchitectureVariants
GNNCM-DTA Yang and Ren (2025)GNNImproved GIN
 MSGNN-DTA Wang et al. (2023)GNNGCN and Graph Attention Network
 HCAF-DTA Li and Yao (2025)GNNCross-attention fusion hypergraph neural network
 HGRL-DTA Chu et al. (2022)GNNHierarchical GNN
 GSAML-DTA Liao et al. (2022)GNNGCN + GAT
 3DProtDTA Voitsitskyi et al. (2023)GNN3D Structure-aware GNN
Table 2: Categorization of Deep Learning-Based DTA Methods by Neural Network Architecture
CategoryMethodDescription
Evaluation MetricsConcordance Index (CI)Measures the ranking agreement between predicted and observed affinity values. Higher values indicate better performance.
 Pearson Correlation Coefficient ( rr )Measures the linear correlation between predicted and observed affinity values. Values closer to 1 indicate stronger agreement.
 Coefficient of Determination ( R2R^{2} )Measures how well the predicted values explain the variation in the observed data. Higher values indicate better fit.
 Mean Absolute Error (MAE)Measures the average absolute difference between predicted and observed affinity values. Lower values indicate better accuracy.
 Mean Squared Error (MSE)Measures the average squared difference between predicted and observed affinity values. Lower values indicate better predictive performance.
Evaluation StrategiesStandard EvaluationThe dataset is randomly divided into training and testing sets. Drugs and targets may appear in both sets but in different interaction pairs.
Table 3: Common evaluation metrics and strategies used in drug–target affinity prediction studies.
Figure 4: Distribution of the reported performance of representative deep learning-based drug–target binding affinity (DTA) prediction methods on the Davis and KIBA benchmark datasets using (A) Concordance Index (CI), (B) Mean Squared Error (MSE), and (C) Coefficient of Determination (R²).
Figure 5: Reported performance of representative deep learning-based drug–target binding affinity (DTA) prediction methods on the Davis and KIBA benchmark datasets using (A) Concordance Index (CI) and (B) Mean Squared Error (MSE), grouped by neural network architecture.

Explore similar work

CardsList
Mar 16, 2026

LaPro-DTA: Latent Dual-View Drug Representations and Salient Protein Feature Extraction for Generalizable Drug--Target Affinity Prediction

Drug--target affinity prediction is pivotal for accelerating drug discovery, yet existing methods suffer from significant performance degradation in realistic cold-start scenarios (unseen drugs/targets/pairs), primarily driven by overfitting to training instances and information loss from irrelevant target sequences. In this paper, we propose LaPro-DTA, a framework designed to achieve robust and generalizable DTA prediction. To tackle overfitting, we devise a latent dual-view drug representation mechanism. It synergizes an instance-level view to capture fine-grained substructures with stochastic perturbation and a distribution-level view to distill generalized chemical scaffolds via semantic remapping, thereby enforcing the model to learn transferable structural rules rather than memorizing specific samples. To mitigate information loss, we introduce a salient protein feature extraction strategy using pattern-aware top-kk pooling, which effectively filters background noise and isolates high-response bioactive regions. Furthermore, a cross-view multi-head attention mechanism fuses these purified features to model comprehensive interactions. Extensive experiments on benchmark datasets demonstrate that LaPro-DTA significantly outperforms state-of-the-art methods, achieving an 8% MSE reduction on the Davis dataset in the challenging unseen-drug setting, while offering interpretable insights into binding mechanisms.

Zihan Dun, Liuyi Xu, An-Yang Lu +2

College of Information Science and Engineering, Northeastern University, Shenyang 110819, China · School of Artificial Intelligence, Beihang University, Beijing 100000, China · School of Computer Science and Engineering, Northeastern University, Shenyang 110819, China

Protein-Ligand Binding Affinity PredictionMolecular Representation LearningProtein Representation Learning
AskPDF
Mar 11, 2026

Co-Diffusion: An Affinity-Aware Two-Stage Latent Diffusion Framework for Generalizable Drug-Target Affinity Prediction

Predicting drug-target affinity is fundamental to virtual screening and lead optimization. However, existing deep models often suffer from representation collapse in stringent cold-start regimes, where the scarcity of labels and domain shifts prevent the learning of transferable pharmacophores and binding motifs. In this paper, we propose Co-Diffusion, a novel affinity-aware framework that redefines DTA prediction as a constrained latent denoising process to enhance generalization. Co-Diffusion employs a two-stage paradigm: Stage I establishes an affinity-steered latent manifold by aligning drug and target embeddings under an explicit supervised objective, ensuring that the latent space reflects the intrinsic binding landscape. Stage II introduces modality-specific latent diffusion as a stochastic perturb-and-denoise regularizer, forcing the model to recover consistent affinity semantics from noisy structural representations. This approach effectively mitigates the reconstruction-regression conflict common in generative DTA models. Theoretically, we show that Co-Diffusion maximizes a variational lower bound on the joint likelihood of drug structures, protein sequences, and binding strength. Extensive experiments across multiple benchmarks demonstrate that Co-Diffusion significantly outperforms state-of-the-art baselines, particularly yielding superior zero-shot generalization on unseen molecular scaffolds and novel protein families-paving a robust path for in silico drug prioritization in unexplored chemical spaces.

Yining Qian, Pengjie Wang, Yixiao Li +4

School of Computer Science and Engineering, Northeastern University, Shenyang 110819, China · College of Information Science and Engineering, Northeastern University, Shenyang 110819, China · Westlake University, Hangzhou 310000, China +2

Drug DiscoveryLatent Diffusion Models
AskPDF
Apr 25, 2026

HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction

Accurate prediction of drug-target binding affinity accelerates drug discovery by prioritizing compounds for experimental validation. Current methods face three limitations: sequence-based approaches discard spatial geometric constraints, structure-based methods fail to exploit hydrogen bond features, and conventional loss functions neglect prediction-target correlation, a key factor for identifying high-affinity compounds in virtual screening. We developed HBGSA (Hydrogen Bond Graph with Self-Attention), a 3.06M-parameter model that encodes hydrogen bond spatial features. HBGSA uses graph neural networks to model hydrogen bond spatial topology with self-attention enhancement and Pearson correlation loss. Experimental results on PDBbind Core Set and CSAR-HiQ dataset demonstrate that HBGSA outperforms baseline methods with strong generalization capability. Ablation studies confirm the effectiveness of hydrogen bond modeling and Pearson correlation loss.

Junxiao Kong, Chupei Tang, Di Wang +4

School of Mathematics and Statistics, Lanzhou University, 222 South Tianshui Road, 730000, Gansu, China · Department of Computer Science, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon Tong, Hong Kong 999077, China

Protein-Ligand Binding Affinity PredictionGraph Neural NetworksMolecular Property PredictionDrug Discovery
AskPDF