R-DTI: Drug Target Interaction Prediction Based on Second-Order Relevance Exploration
Authors: Yang Hua, Tianyang Xu, Xiaoning Song, Zhenhua Feng, Rui Wang, Wenjie Zhang, Xiaojun Wu
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results prove the merits and superiority of our R-DTI against the state-of-the-art, achieving 1.4% and 1.9% higher AUCROC on the Binding DB and Drug Bank datasets, respectively. |
| Researcher Affiliation | Academia | 1School of Artifcial Intelligence and Computer Science, Jiangnan University, Wuxi, Jiangsu, P.R. China 2Sino-UK Joint Laboratory on Artificial Intelligence, Ministry of Science and Technology, China 3International Joint Laboratory on Artificial Intelligence, Ministry of Education, China {7211905018; wenjie.zhang}@stu.jiangnan.edu.cn; {tianyang.xu; x.song; fengzhenhua; cs wr; wu xiaojun}@jiangnan.edu.cn; |
| Pseudocode | No | The paper describes the proposed R-DTI method, its components (Multi-Mode Protein Feature Extractor, Multi-Mode Drug Feature Extractor, Riemannian Classifier) and their mathematical formulations (equations 1-5), but does not present a dedicated pseudocode or algorithm block. |
| Open Source Code | Yes | Code https://github.com/JU-Hua Y/R-DTI |
| Open Datasets | Yes | We evaluate the proposed method on six datasets, including four classical DTI benchmarks, Human, C.elegans (Tsubaki, Tomii, and Sese 2019), Binding DB (Chen et al. 2020), Drug Bank (Zhao et al. 2022), and two DTA datasets, Davi and KIBA (Davis et al. 2011). ... Specifically, we use a structure-based extractor and 3D PDB data (Berman et al. 2000) containing protein sequence and validated 3D structure to empower the sequence-based extractor. |
| Dataset Splits | Yes | To evaluate the generalization of DTI models, we split the test data of Binding DB (Chen et al. 2020) into three subsets, Unseen drug, Unseen target, and Unseen pair. The number of positive and negative samples is shown in Figure 3 (d). |
| Hardware Specification | No | The paper does not provide specific details about the hardware (e.g., GPU models, CPU types, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions several tools and models like Prot-Bert, Dense Net, Smiles-Bert, and Rdkit tools, but does not provide specific version numbers for these software dependencies or the overall development environment (e.g., Python, PyTorch/TensorFlow versions). |
| Experiment Setup | Yes | Our loss functions, hyperparameter settings and measure metrics are shown in Appendix B. |