Risk prediction and marker selection in nonsynonymous single nucleotide polymorphisms using whole genome sequencing...

Area of Science:

• Genetics
• Bioinformatics
• Genomics

Background:

• Nonsynonymous single nucleotide polymorphisms (nsSNPs) can alter protein structure and function, potentially leading to deleterious effects.
• Genome-wide studies focusing on the impact of nsSNPs remain relatively rare despite their known functional consequences.

Purpose of the Study:

• To develop a method for predicting individual risk scores based on the deleterious effects of nsSNPs.
• To identify an optimal subset of nsSNPs that effectively represents the cumulative nsSNP effect using machine learning.
• To explore the biological functions associated with selected nsSNPs through gene ontology analysis.

Main Methods:

• Prediction of deleterious effects for a large set of nsSNPs.
• Application of machine learning to select a representative nsSNP subset from a larger dataset.
• Gene ontology analysis to identify enriched biological processes among the selected nsSNPs.

Main Results:

• A total of 16,100 nsSNPs were identified as optimal representatives out of 89,519 initially regressed nsSNPs.
• Gene ontology analysis revealed significant enrichment in DNA metabolic processes, chemokine- and immune-related functions, and reproduction.
• A risk score prediction model was developed based on the identified nsSNPs.

Conclusions:

• The developed risk score prediction and nsSNP marker selection methods offer a novel approach for genome-wide association studies.
• These findings are expected to advance breeding science by providing better genetic markers.
• The study highlights the importance of nsSNPs in fundamental biological processes and individual risk assessment.