Identifying N⁶-methyladenosine sites using extreme gradient boosting system optimized by particle swarm optimizer

Area of Science:

• Biochemistry
• Bioinformatics
• Molecular Biology

Background:

• N6-methyladenosine (m6A) is a crucial RNA modification influencing gene expression, mRNA stability, and cell differentiation.
• Accurate identification of m6A sites is vital for biomedical research and therapeutic development.
• Current laboratory methods for m6A site identification are costly and time-consuming, necessitating efficient computational approaches.

Purpose of the Study:

• To develop an advanced computational method for improved m6A site prediction.
• To enhance the accuracy and efficiency of identifying m6A modification sites in RNA sequences.

Main Methods:

• Proposed a novel prediction method, PXGB (Particle Swarm Optimization-optimized eXtreme Gradient Boosting).
• Integrated three feature types: position-specific nucleotide propensity (PSNP), position-specific dinucleotide propensity (PSDP), and nucleotide composition (NC).
• Utilized deep features combined with original sequence-based features for enhanced prediction.

Main Results:

• PXGB achieved an Area Under the Curve (AUC) of 0.8390 and a Matthews Correlation Coefficient (MCC) of 0.5234 via 10-fold cross-validation.
• Comparative analysis showed PXGB outperformed existing methods in predicting m6A sites, evidenced by higher MCC and AUC values.
• The proposed method demonstrates significant effectiveness in m6A site prediction.

Conclusions:

• The PXGB method offers a more accurate and efficient approach for identifying m6A sites compared to existing computational tools.
• This predictor can aid in identifying a larger number of m6A sites and guide subsequent experimental validation.
• The findings contribute to advancing basic biomedical research and drug discovery through improved m6A site identification.