iDNA6mA-Rice

iDNA6mA-Rice predicts N6-methyladenine (6mA) sites in the rice genome to enable genomic-scale analysis of 6mA-mediated regulatory functions.

Key Features:

• Target modification: Detects N6-methyladenine (6mA) sites within the rice genome.
• Encoding scheme: Uses mono-nucleotide binary encoding to represent DNA sequences.
• Sample construction: Prepares positive and negative samples for supervised learning.
• Classification algorithm: Employs the Random Forest algorithm to classify 6mA versus non-6mA sites.
• Validation strategy: Evaluated by fivefold cross-validation achieving an AUC of 0.964 and accuracy of 0.917.
• Independent validation: Generalization confirmed on an independent dataset with an AUC of 0.981.

Scientific Applications:

• Genome-wide 6mA mapping: Enables identification of 6mA distribution across the rice genome.
• Strand discrimination studies: Facilitates investigation of 6mA roles in distinguishing original and newly synthesized DNA strands post-replication.
• Transcriptional regulation research: Supports analysis of 6mA influence on gene transcription.
• Transposable element repression: Aids studies on 6mA-mediated repression of transposable elements.
• Cell cycle and DNA stability: Permits exploration of 6mA effects on DNA duplex stability and cell cycle dynamics.
• Plant epigenomics: Assists characterization of DNA methylation patterns and regulatory roles in plant genomes.

Methodology:

Sequences are encoded by mono-nucleotide binary encoding to form positive and negative samples, classified using a Random Forest algorithm, and assessed by fivefold cross-validation and independent-dataset evaluation reporting AUC and accuracy metrics.