iNuc-PseKNC: a sequence-based predictor for predicting nucleosome positioning in genomes with pseudo k-tuple...

Area of Science:

• Genomics
• Computational Biology
• Molecular Biology

Background:

• Nucleosome positioning is crucial for regulating cellular processes.
• Automated methods for nucleosome positioning prediction are needed due to vast genomic data.
• Existing methods are often species-specific and overlook DNA structural properties.

Purpose of the Study:

• To develop an automated, accurate, and species-general computational method for nucleosome positioning prediction.
• To incorporate DNA local structural properties into feature-vector representation for improved prediction.
• To provide a user-friendly web server for the developed prediction tool.

Main Methods:

• Developed a novel predictor named 'iNuc-PseKNC'.
• Formulated DNA sequence samples using a 'pseudo k-tuple nucleotide composition' feature-vector.
• Incorporated six DNA local structural properties into the feature-vector formulation.

Main Results:

• Achieved high success rates in predicting nucleosome positioning: 86.27% (Homo sapiens), 86.90% (Caenorhabditis elegans), and 79.97% (Drosophila melanogaster).
• Demonstrated superior performance compared to existing methods on various benchmark datasets.
• Validated the predictor through rigorous cross-validation tests on three stringent datasets.

Conclusions:

• iNuc-PseKNC is an effective computational tool for predicting nucleosome positioning across different species.
• The incorporation of DNA local structural properties significantly enhances prediction accuracy.
• The freely accessible web server facilitates broader application in genomic research.