{"id":21441762,"url":"https://github.com/dmis-lab/arkdta","last_synced_at":"2026-03-12T09:27:01.711Z","repository":{"id":239670000,"uuid":"620636373","full_name":"dmis-lab/ArkDTA","owner":"dmis-lab","description":null,"archived":false,"fork":false,"pushed_at":"2023-04-11T13:40:45.000Z","size":3616,"stargazers_count":11,"open_issues_count":0,"forks_count":2,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-07-14T20:10:05.442Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/dmis-lab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2023-03-29T04:29:47.000Z","updated_at":"2024-02-05T10:08:26.000Z","dependencies_parsed_at":null,"dependency_job_id":"a162160a-3852-4e44-aa17-2f182c075052","html_url":"https://github.com/dmis-lab/ArkDTA","commit_stats":null,"previous_names":["dmis-lab/arkdta"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/dmis-lab/ArkDTA","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dmis-lab%2FArkDTA","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dmis-lab%2FArkDTA/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dmis-lab%2FArkDTA/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dmis-lab%2FArkDTA/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/dmis-lab","download_url":"https://codeload.github.com/dmis-lab/ArkDTA/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dmis-lab%2FArkDTA/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":270385467,"owners_count":24574556,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-14T02:00:10.309Z","response_time":75,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-23T01:42:04.630Z","updated_at":"2026-03-12T09:27:01.642Z","avatar_url":"https://github.com/dmis-lab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ArkDTA: Attention Regularization guided by non-Covalent Interactions for Explainable Drug-Target Binding Affinity Prediction\n\n## Abstract\n\nProtein-ligand binding affinity prediction is a central task in drug design and development. Cross-modal attention mechanism has recently become a core component of many deep learning models due to its potential to improve model explainability. Non-covalent interactions, one of the most critical domain knowledge in binding affinity prediction task, should be incorporated in protein-ligand attention mechanism for more explainable deep DTI models. We propose ArkDTA, a novel deep neural architecture for explainable binding affinity prediction guided by non-covalent interactions. Experimental results show that ArkDTA achieves predictive performance comparable to current state-of-the-art models while significantly improving model explainability. Qualitative investigation into our novel attention mechanism reveals that ArkDTA can identify potential regions for non-covalent interactions between candidate drug compounds and target proteins, as well as guiding internal operations of the model in a more interpretable and domain-aware manner. (*submitted to ISMB2023, under review*)\n\n## Overview of ArkDTA\n\n![img](./figures/0_arkdta.png)\n\n## Attention Regularization guided by non-Covalent Interactions\n\n![img](./figures/1_arkmab.png)\n\n## Prerequisites for running ArkDTA\n\n- Python 3.7.9\n- CUDA: 11.X\n- Download and extract data.tar.gz ([link](https://drive.google.com/file/d/1hmR5w47VUk6RW0br8BanJT94R2FPHgDL/view?usp=share_link)), 45MB) at current directory. These files are the preprocessed datasets PDBBind (ver.2020), Davis and Metz.\n- Download and extract saved.tar.gz ([link](https://drive.google.com/file/d/1iVttdzlAMXYeJ11JKVe19Dkvgpb8PZSS/view?usp=share_link)), 170MB) at directory **./saved**. These files are the model checkpoints for each fold of the PDBbind datset.\n\n## Installing the Python (3.8.12) Conda Environment\n\n```\nconda env create -f arkdta.yaml\nconda activate arkdta\n```\n\n## How to use the ArkDTA source code\n\n### Training ArkDTA on PDBBind Dataset\n\nRun the following code,\n```\npython run.py -pn {wandb_project_name} -sn arkdta -mg {multiple gpu indices}\n```\n\nIf you want to train ArkDTA on the IC50 subset, configure the **/sessions/arkdta.yaml** by editing the following,\n```\nba_measure: IC50 \n```\n\n### Evaluating ArkDTA on PDBBind Dataset (5CV)\n\nRun the following code,\n```\npython run.py -pn {wandb_project_name} -sn arkdta -mg {multiple gpu indices} -tm\n```\n\n### Finetuning ArkDTA on other datasets (Davis, Metz)\n\nConfigure the **/sessions/arkdta.yaml** by editing the following,\n```\ndataset_subsets: davis\ndataset_partition: randomsingle\n```\n\nThen run the following code,\n```\npython run.py -pn {wandb_project_name} -sn arkdta -mg {multiple gpu indices} -ft {davis or metz}\n```\n\n### Evaluating ArkDTA on other datasets\nRun the following code,\n```\npython run.py -pn {wandb_project_name} -sn arkdta -mg {multiple gpu indices} -tm -cn {your/saved/path_davis or _metz}\n```\n\n### Running model inference and extracting attention maps from ArkDTA\n\nRun the following script,\n```\n./arkdta.sh\n```\n\nYou can change the input SMILES (ligands) or FASTA sequence (proteins) by editting the **arkdta.sh** file.\n\n#### 4x6n, 3Y5\n\n![img](./figures/2_4x6n_3y5.png)\n\n#### 6n77, KEJ\n\n![img](./figures/3_6n77_kej.png)\n\n#### 8bq4, QZR\n\n![img](./figures/4_8bq4_qzr.png)\n\n\n## Contributors\n\n\u003ctable\u003e\n\t\u003ctr\u003e\n\t\t\u003cth\u003eName\u003c/th\u003e\t\t\n\t\t\u003cth\u003eAffiliation\u003c/th\u003e\n\t\t\u003cth\u003eEmail\u003c/th\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eMogan Gim\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003eakim@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eJunseok Choe\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003ejuns94@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eSeungheun Baek\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003etmdgms9417@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eJueon Park\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003ejueon_park@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eChaeeun Lee\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003echaeeunlee1997@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eMinjae Ju\u0026dagger;\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eLG CNS, AI Research Center, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003eminjae.ju@lgcns.com\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eSumin Lee\u0026dagger;\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eLG AI Research, Seoul South Korea\u003c/td\u003e\n\t\t\u003ctd\u003esumin.lee@lgresearch.ai\u003c/td\u003e\n\t\u003c/tr\u003e\n\t\u003ctr\u003e\n\t\t\u003ctd\u003eJaewoo Kang*\u003c/td\u003e\t\t\n\t\t\u003ctd\u003eData Mining and Information Systems Lab,\u003cbr\u003eKorea University, Seoul, South Korea\u003c/td\u003e\n\t\t\u003ctd\u003ekangj@korea.ac.kr\u003c/td\u003e\n\t\u003c/tr\u003e\n\n\u003c/table\u003e\n\n\n\n- \u0026dagger;: *This work was done while the author was a graduate student at Korea University Computer Science Department.*\n- \u0026ast;: *Corresponding Author*","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdmis-lab%2Farkdta","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdmis-lab%2Farkdta","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdmis-lab%2Farkdta/lists"}