PRIME-BSPre: a genome-wide protein-RNA binding sites prediction method based on templates

TL;DR

PRIME-BSPre is a new method that predicts where proteins bind to RNA in the genome using RNA structure and protein data.

Contribution

PRIME-BSPre introduces the use of low Shannon entropy to model RNA binding preferences of RBPs.

Findings

PRIME-BSPre uses RNA sequence, secondary structure, and RBP tertiary structure for genome-wide predictions.

The method includes LS-PEAK to optimize alignment screening for RNA binding site prediction.

PRIME-BSPre shows excellent performance and robustness across different cell lines.

Abstract

In this paper, we present PRIME-BSPre, a template-based genome-wide method for predicting protein-RNA binding sites that incorporates the RNA sequence and secondary structure as well as the tertiary structure of corresponding RBPs. We are pioneers in introducing low Shannon entropy algorithm in PRIME-BSPre to describe the binding preferences of RBPs on RNA motifs. The LS-PEAK derived from LS-Scores in PRIME-BSPre is utilized to optimize the alignments screening. PRIME-BSPre has been successfully benchmarked on the human genome, demonstrating its excellent prediction performance on independent RBP datasets and its robustness across different cell lines.

The online version contains supplementary material available at 10.1186/s12864-026-12657-3.

Figures