Small Molecule Approach to RNA Targeting Binder Discovery (SMARTBind) Using Deep Learning Without Structural Input

This dataset supports SMARTBind (Small Molecule Approaches to RNA Targeting Binder Discovery), a structure-agnostic ligand discovery framework that combines an RNA large language model with contrastive learning and a ligand-specific decoy enhancement strategy. Please cite the following publication when using the dataset: Jiang, Shiyu, Amirhossein Taghavi, Tenghui Wang, Kisu Sung, Samantha M. Meyer, Noah A. Springer, Jinhang Wei, Jessica L. Childs-Disney, Chenglong Li, Mattew D. Disney, and Yanjun Li. "Small Molecule Approach to RNA Targeting Binder Discovery (SMARTBind) Using Deep Learning Without Structural Input." bioRxiv (2025): 2025-09. Overview The dataset contains model checkpoint and training data for SMARTBind including RNAmigos1 and HARIBOSS 10-fold random-split, HARIBOSS 5-fold cross-validation splits. All data is organized under the archive file SMARTBind_dataset.zip. Contents SMARTBind_weight.zip: saved checkpoint of the 10-fold SMARTBind model for binding score and binding site prediction. 10-fold cross-validation.zip: the folder contains HARIBOSS 10-fold random split dataset and RNAmigos1 10-fold rnadom split dataset. 5-fold cross-validations.zip: the folder contains five HARIBOSS 5-fold cross-validation splits, including RNA sequence-based, RNA structure-based, RNA pocket-based, pair-based, ligand-existence splittings. 5-fold cross-validations gbsubset.zip: the folder contains five HARIBOSS 5-fold cross-validation splits (GerNA-Bind subset), including RNA sequence-based, RNA structure-based, RNA pocket-based, pair-based, ligand-existence splittings. Chemspace_Screening_Compounds_SMlLES_sampled1M.smi.zip: the virtual screening background library used for benchmarking time-dependent test set. Decoy library.smi.zip: a chemical diverse decoy library with 92,626 entries that is curated for the ligand-specific decoy enhancement strategy. decoys.zip: DecoyFinder- and DeepCoy-generated decoy datasets for virtual screening evaluation.