Splice sites detection using chaos game representation and neural network.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate detection of splice sites is crucial for understanding gene structure and function.
• Existing methods for splice site detection can be complex and computationally intensive.

Purpose of the Study:

• To propose a novel and accurate method for detecting acceptor and donor splice sites.
• To leverage Chaos Game Representation (CGR) for effective DNA sequence numerical representation.
• To develop a simplified artificial neural network (ANN) model for splice site detection.

Main Methods:

• Utilized Chaos Game Representation (CGR) to map DNA sequences into numerical feature vectors.
• Employed an artificial neural network (ANN) with CGR-derived features for splice site detection.
• Validated the method on the NN269 dataset.

Main Results:

• The proposed method achieved good accuracy in detecting splice sites.
• The CGR-based ANN approach demonstrated simplicity compared to existing literature methods.
• The model requires only a single neural network component for splice site identification.

Conclusions:

• Chaos Game Representation provides an effective one-to-one numerical representation for DNA sequences.
• The integration of CGR with ANNs offers a promising, accurate, and simplified approach to splice site detection.
• This method contributes to advancing computational tools in genomic analysis.