RNAsnoop

RNAsnoop predicts thermodynamically optimal H/ACA snoRNA–RNA interactions to identify pseudouridylation guide targets.

Key Features:

• Dynamic Programming Interaction Model: Computes thermodynamically optimal H/ACA-RNA interactions using a dedicated dynamic programming algorithm for efficient genome-wide target scanning.
• SVM-Based Classification with Comparative Evaluation: Applies a support vector machine (SVM) to distinguish true binding sites from spurious predictions and incorporates comparative sequence information to refine and validate snoRNA–target interactions.

Scientific Applications:

• Pseudouridine Site Identification: Identifies H/ACA snoRNAs guiding pseudouridine modifications in human rRNAs and assigns targets to orphan H/ACA snoRNAs in species such as Drosophila.

Methodology:

RNAsnoop integrates dynamic programming to calculate thermodynamically favorable H/ACA snoRNA–target duplexes, filters candidates using an SVM classifier, and incorporates comparative sequence analysis to improve prediction robustness and specificity.