A 2-phased approach for detecting multiple loci associations with traits

Area of Science:

• Genomics
• Bioinformatics
• Data Mining

Background:

• Advances in SNP genotyping reduce costs but increase data volume.
• High-dimensional SNP data poses challenges for conventional analysis techniques.

Purpose of the Study:

• To propose a novel method for analyzing large-scale SNP genotype data.
• To address the high-dimensionality problem in genomic data analysis.

Main Methods:

• Utilizes document-term and transaction analysis models.
• Phase 1: Dimensionality reduction via significant SNP extraction using data transformation.
• Phase 2: Association rule discovery between SNPs and traits in reduced-dimension data.

Main Results:

• Successfully reduced dimensions of SNP genotype data.
• Identified significant association rules between SNPs and traits.
• Validated findings through literature surveys and experiments on HGDP panel data.

Conclusions:

• The proposed method effectively performs dimensionality reduction.
• The method accurately identifies associations between multiple SNPs and traits.
• Offers a robust solution for analyzing large-scale SNP genotype data.