RNAMethPre: A Web Server for the Prediction and Query of mRNA m6A Sites

Area of Science:

• Epigenetics and RNA Biology
• Bioinformatics and Computational Biology

Background:

• N6-Methyladenosine (m6A) is the most prevalent mRNA modification, influencing diverse biological processes across many species.
• Transcriptome-wide m6A site identification is limited in scope, necessitating robust computational prediction methods.

Purpose of the Study:

• To develop and validate a computational method for predicting mRNA m6A sites in mammalian sequences.
• To create a user-friendly web server for querying and visualizing predicted and curated m6A sites.

Main Methods:

• Integration of multiple mRNA features: flanking sequences, local secondary structure, and relative position.
• Training a Support Vector Machine (SVM) classifier for m6A site prediction.
• Validation using cross-validation and independent dataset tests.

Main Results:

• The SVM classifier achieved high performance in predicting m6A sites.
• The RNAMethPre web server was established, offering a database of predicted and literature-curated m6A sites for human and mouse.
• The server includes genome browser visualization for queried m6A data.

Conclusions:

• The developed computational method provides an accurate strategy for predicting mRNA m6A sites.
• The RNAMethPre web server serves as a valuable, publicly accessible resource for researchers studying m6A modifications.