DrugPred_RNA-A Tool for Structure-Based Druggability Predictions for RNA Binding Sites

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular biology

Background:

• Ribonucleic acid (RNA) is a significant emerging target for novel drug discovery.
• Not all RNA binding sites are amenable to conventional drug-like ligands, similar to protein targets.
• Identifying druggable RNA binding sites is crucial for advancing RNA-targeted therapeutics.

Purpose of the Study:

• To develop a structure-based computational tool, DrugPred_RNA, for identifying druggable RNA binding sites.
• To overcome the challenge of limited annotated RNA binding site data for predictor training.

Main Methods:

• Developed DrugPred_RNA, a structure-based predictor for RNA binding site druggability.
• Trained the predictor on protein pockets using shared descriptors applicable to both RNA and protein binding sites.
• Validated the predictor's performance on both protein and RNA datasets.

Main Results:

• DrugPred_RNA effectively discriminated between druggable and non-druggable protein binding sites.
• Predictions for RNA binding sites showed strong agreement with manual assessments.
• Drug-like ligands in an RNA test set were predominantly located in sites predicted as druggable by DrugPred_RNA.

Conclusions:

• DrugPred_RNA is a robust predictor of RNA binding site druggability, validated against protein and RNA data.
• The method demonstrates resilience to conformational and sequence variations in binding sites.
• DrugPred_RNA can significantly aid and direct drug discovery efforts focused on RNA targets.