iRNA-Methyl

iRNA-Methyl predicts N6-methyladenosine (m6A) sites in RNA sequences to support mapping RNA methylation and studying its regulatory roles.

Key Features:

• Target modification: Predicts N6-methyladenosine (m6A) at adenine positions, taking into account the consensus motif GAC.
• Sequence encoding: Implements pseudo dinucleotide composition to represent RNA sequence information.
• Physicochemical integration: Integrates three RNA physicochemical properties into the predictive model.
• Scalability: Applicable to large RNA sequence datasets for genome- or transcriptome-wide identification of potential m6A sites.
• Validation: Evaluated by cross-validation on benchmark datasets with reported prediction accuracy of 65.59%.

Scientific Applications:

• Transcriptome-wide m6A mapping: Identification of m6A sites across transcriptomes to characterize modification landscapes.
• Regulatory mechanism studies: Supporting investigation of the roles of m6A in RNA regulation and related biological processes.
• Genome analysis of RNA modifications: Enabling genome-scale analyses of m6A distribution and potential functional impacts.

Methodology:

Sequence features are encoded using pseudo dinucleotide composition with integration of three RNA physicochemical properties; model performance was assessed by cross-validation on benchmark datasets yielding 65.59% accuracy.