Supervised selection of single nucleotide polymorphisms in chronic fatigue syndrome

Area of Science:

• Genetics
• Bioinformatics
• Medical Diagnostics

Background:

• Single nucleotide polymorphism (SNP) selection methods yield conflicting results for chronic fatigue syndrome (CFS) prediction.
• Existing methods lack a standardized approach for identifying optimal SNP profiles.

Purpose of the Study:

• To evaluate the predictive efficacy of SNPs selected via a supervised approach for chronic fatigue syndrome (CFS).
• To identify an optimal SNP profile for enhanced CFS risk stratification.

Main Methods:

• Polymorphisms not meeting Hardy-Weinberg equilibrium were excluded.
• A supervised approach was used to derive a SNP profile.
• Prediction accuracy was compared against a linkage disequilibrium-based profile and evaluated for risk stratum efficacy.

Main Results:

• The supervised approach yielded a valid SNP profile with significantly higher prediction accuracy (72.8%) than linkage disequilibrium (62.2%).
• The profile included NR3C1_11159943 major and 5HTT_7911132 minor alleles.
• Combining the profile with clinical data (muscular pain, sinus/nasal symptoms) further increased prediction accuracy.

Conclusions:

• A supervised approach successfully identified a reliable SNP profile for chronic fatigue syndrome (CFS).
• This profile achieved the highest reported accuracy for the dataset and improved further with clinical data integration.