PPUS

PPUS predicts pseudouridine (Ψ) modification sites and assigns likely pseudouridine synthases (PUS) responsible by applying a support vector machine to nucleotide sequence context.

Key Features:

• Support Vector Machine (SVM): PPUS employs a support vector machine as its core classifier.
• Sequence-context features: It leverages nucleotide sequences surrounding potential Ψ sites as predictive features.
• Enzyme-specific prediction: The method identifies which specific PUS enzyme modifies predicted Ψ sites.
• Species and enzyme coverage: PPUS provides predictions for PUS1, PUS4, and PUS7 in yeast and for PUS4 in humans.
• Novel site prediction: The approach enables accurate prediction of new pseudouridine sites.

Scientific Applications:

• Discovery of novel Ψ sites: Enable identification of previously unannotated pseudouridine modifications in RNA sequences.
• Enzyme-specific mapping: Support assignment of predicted Ψ sites to specific pseudouridine synthases (PUS1, PUS4, PUS7).
• RNA structure-function and regulation studies: Facilitate investigation of the structural and regulatory roles of pseudouridine modifications.

Methodology:

PPUS uses a support vector machine classifier that leverages nucleotide sequences surrounding candidate pseudouridine sites as predictive features.