m6AmPred: Identifying RNA N6, 2'-O-dimethyladenosine (m⁶A_m) sites based on sequence-derived information

Area of Science:

• Molecular Biology
• Bioinformatics
• RNA Modifications

Background:

• N6,2 '-O-dimethyladenosine (m⁶A_m) is a prevalent, reversible RNA modification.
• The precise biological roles of m⁶A_m remain largely unelucidated.
• Accurate identification of m⁶A_m sites is crucial for understanding its functional significance.

Purpose of the Study:

• To develop the first computational tool for identifying m⁶A_m sites in RNA sequences.
• To provide researchers with a reliable method for predicting m⁶A_m modification sites.
• To facilitate the study of m⁶A_m's influence on cellular mRNA fate.

Main Methods:

• Development of m6AmPred, a web server utilizing the eXtreme Gradient Boosting with Dart algorithm (XgbDart).
• Implementation of the EIIP-PseEIIP encoding scheme for RNA sequence representation.
• Rigorous validation using 10-fold cross-validation, independent testing, and cross-validation with experimentally verified m⁶A_m sites.

Main Results:

• m6AmPred demonstrated high prediction accuracy, achieving Area Under the Curve (AUC) values exceeding 0.954.
• The tool was validated against independent datasets and experimental data, confirming its robust performance.
• The web server provides a user-friendly platform for in silico identification of m⁶A_m sites.

Conclusions:

• m6AmPred is the first web server for predicting N6,2 '-O-dimethyladenosine RNA modification sites.
• The tool offers high accuracy and reliability, serving as a valuable resource for RNA biology research.
• m6AmPred will aid researchers in investigating the functional roles of m⁶A_m in various RNA molecules.