XG-PseU

XG-PseU predicts pseudouridine (Ψ) sites using eXtreme Gradient Boosting (xgboost) to improve computational identification of post-transcriptional modifications.

Key Features:

• Algorithm: Implements eXtreme Gradient Boosting (xgboost) as the core classifier.
• Target: Predicts pseudouridine (Ψ) modification sites in RNA sequences.
• Feature selection: Selects optimal features using a combination of forward feature selection and increment feature selection methods.
• Performance: Demonstrates enhanced accuracy and efficiency relative to existing computational approaches.
• Predictive modeling: Builds a predictive model based on selected features and xgboost classification.

Scientific Applications:

• Pseudouridine site identification: Computational identification of pseudouridine (Ψ) sites to support studies of RNA modification patterns.
• Functional analysis: Facilitates exploration of the biological roles and functional significance of pseudouridine in post-transcriptional regulation.

Methodology:

Uses an eXtreme Gradient Boosting (xgboost) classifier and selects optimal features via a combination of forward feature selection and increment feature selection methods.