iSS-PseDNC: identifying splicing sites using pseudo dinucleotide composition.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Eukaryotic gene expression involves precise removal of introns and ligation of exons.
• Accurate splice site identification is crucial for gene structure annotation and understanding RNA splicing.
• Existing computational methods often overlook intrinsic local DNA structural properties.

Purpose of the Study:

• To develop an automated computational method for rapid and effective splice site detection.
• To incorporate DNA local structural properties into splice site prediction.
• To enhance gene structure annotation and RNA splicing research.

Main Methods:

• Developed a predictor named iSS-PseDNC.
• Formulated DNA sequences using a novel feature vector, pseudo dinucleotide composition (PseDNC).
• Incorporated six DNA local structural properties into the PseDNC feature vector.

Main Results:

• iSS-PseDNC achieved high success rates in identifying splice sites.
• Overall success rates were 85.45% for splice donor sites and 87.73% for splice acceptor sites.
• Rigorous cross-validation tests were performed on two benchmark datasets.

Conclusions:

• iSS-PseDNC shows promise as a valuable tool for splice site identification.
• The incorporated DNA local structural properties may offer new insights into RNA splicing mechanisms.
• The study contributes to advancing automated gene annotation and splicing research.