iORI-Euk

iORI-Euk identifies origins of replication (ORIs) across multiple eukaryotic species using sequence-based feature extraction and machine learning to support studies of DNA replication initiation and its regulation.

Key Features:

• Integrated Prediction Model: Represents an integrated predictor for ORIs across seven eukaryotic species: Homo sapiens, Mus musculus, Drosophila melanogaster, Arabidopsis thaliana, Pichia pastoris, Schizosaccharomyces pombe, and Kluyveromyces lactis.
• Benchmark Dataset Construction: Constructs comprehensive benchmark datasets by collecting and curating ORI data from public databases for each of the seven species.
• Feature Extraction Strategies: Employs three sequence feature extraction strategies—k-mer, binary encoding, and a combined k-mer plus binary encoding approach—to encode DNA sequence samples.
• Classification Algorithm Performance: Evaluates multiple classifiers and identifies the support vector machine (SVM) as the optimal model based on 5-fold cross-validation and independent dataset evaluations.

Scientific Applications:

• DNA Replication Research: Enables identification of ORIs to facilitate study of DNA replication initiation and replication timing across eukaryotes.
• Gene Expression Regulation: Supports analyses linking ORI locations to local gene expression patterns and regulatory effects.
• Drug Development: Provides ORI annotations that can inform research into therapeutics targeting aberrant DNA replication processes.

Methodology:

Collects and curates ORI data from public databases to build species-specific benchmark datasets; applies k-mer, binary encoding, and combined feature extraction to transform DNA sequences; evaluates multiple classification algorithms using 5-fold cross-validation and independent dataset testing, selecting SVM as the best-performing predictor.