RECON

RECON: Nucleosome Positioning Prediction via Discriminant Analysis and Genetic Algorithm Optimization

RECON predicts nucleosome formation potential across DNA sequences by profiling dinucleotide frequency patterns and classifying genomic regions based on statistical discriminant analysis.

Key Features:

• Nucleosome Potential Profiling: Generates probability profiles identifying genomic regions with high or low likelihood of nucleosome formation.
• Discriminant Analysis: Classifies DNA sequences into potential nucleosome-forming and non-nucleosome-forming regions.
• Genetic Algorithm Optimization: Optimizes model parameters influencing nucleosome positioning using a genetic algorithm.

Scientific Applications:

• Chromatin Structure Analysis: Predicts nucleosome positions to infer chromatin organization and gene expression regulation.
• Epigenetics Research: Analyzes nucleosome distribution to study epigenetic regulatory mechanisms.
• Genomic Sequence Annotation: Identifies nucleosome-forming regions to support genome annotation.

Methodology:

RECON analyzes dinucleotide frequencies within local DNA regions and applies discriminant analysis to classify sequences. A genetic algorithm iteratively refines model parameters to improve nucleosome positioning prediction accuracy.