Identify origin of replication in Saccharomyces cerevisiae using two-step feature selection technique.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• DNA replication ensures genetic continuity, with origins of replication (ORI) being crucial initiation sites.
• Experimental identification of ORI is costly and time-consuming.
• Computational methods offer a cost-effective and rapid alternative for ORI identification.

Purpose of the Study:

• To develop a computational predictor for identifying ORI in the Saccharomyces cerevisiae genome.
• To leverage sequence information and physicochemical properties for ORI prediction.

Main Methods:

• Developed iORI-PseKNC2.0, a predictor utilizing pseudo k-tuple nucleotide composition (PseKNC) with 90 physicochemical properties.
• Implemented a two-step feature selection process to refine predictive accuracy.
• Employed a support vector machine (SVM) algorithm for classification.

Main Results:

• Achieved an 88.53% overall success rate in 5-fold cross-validation.
• The PseKNC approach effectively formulated ORI and non-ORI samples.
• The developed predictor demonstrates high accuracy in identifying yeast ORI.

Conclusions:

• iORI-PseKNC2.0 provides an accurate and efficient method for ORI identification in yeast.
• The predictor can significantly aid genomic analysis for researchers.
• A user-friendly webserver is available for broader accessibility and application.