CGR

CGR transforms DNA sequences into numerical representations that are used with an artificial neural network to detect acceptor and donor splice sites.

Key Features:

• Chaos Game Representation (CGR): An iterative mapping technique that converts DNA sequences into numerical sequences in a one-to-one manner by assigning each nucleotide a specific position on a plane.
• Artificial Neural Network: Numerical sequences produced by CGR serve as feature vectors that are input to an artificial neural network to classify splice sites.
• Simplicity and Efficiency: The approach uses a single neural network component and computational experiments report good accuracy on the NN269 dataset.

Scientific Applications:

• Splice Site Detection: Identification of acceptor and donor splice sites within DNA sequences to support analyses of splicing mechanisms and exon–intron boundaries.
• Genomic Research: Precise splice site identification for applications in studies of gene expression regulation, genetic disorder research, and evolutionary biology.

Methodology:

DNA sequences are converted into numerical sequences using the chaos game representation, an iterative mapping that assigns nucleotides to positions on a plane in a one-to-one manner. The resulting numerical sequences serve as feature vectors. These feature vectors are input into an artificial neural network that classifies acceptor and donor splice sites.