MambaTransDTA: A Hybrid Mamba-Transformer Architecture for Accurate Drug-Target Binding Affinity Prediction
Bibliographic record
Abstract
In recent years, deep learning techniques have made significant advances in drug-target affinity (DTA) prediction. However, existing models still have considerable room for improvement in prediction accuracy, robustness, and generalization ability. To address these challenges, we present a novel hybrid model, MambaTransDTA, which integrates the Mamba architecture with the Transformer architecture to optimize drug-target interaction prediction. By combining Mamba's ability to capture long-range dependencies with the Transformer's modeling of local interactions, MambaTransDTA estimates drug-target affinity more comprehensively and achieves notable performance improvements. On the four benchmark data sets, MambaTransDTA demonstrates exceptional prediction accuracy, achieving mean squared error (MSE) values of 0.191 (Davis), 0.173 (KIBA), 0.302 (Metz) and 0.715 (BindingDB). These results represent a significant improvement over the next best existing model, with relative error reductions of 4.3% (Davis), 3.3% (KIBA), 4.7% (Metz) and 10.5% (BindingDB). Ablation experiments and analysis of different hybridization strategies confirm that the proposed MambaTransDTA architecture significantly improves prediction performance by fully leveraging the strengths of both architectures. Therefore, MambaTransDTA is poised to become an essential tool for AI-driven drug discovery. Our data and code are available at https://github.com/pdssunny/MambaTransDTA.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".