Research on Frequent Itemset Mining of Imaging Genetics GWAS in Alzheimer's Disease.
Hong LiangLuolong CaoYue GaoHaoran LuoXianglian MengYing WangJin LiWenjie LiuPublished in: Genes (2022)
As an efficient method, genome-wide association study (GWAS) is used to identify the association between genetic variation and pathological phenotypes, and many significant genetic variations founded by GWAS are closely associated with human diseases. However, it is not enough to mine only a single marker effect variation on complex biological phenotypes. Mining highly correlated single nucleotide polymorphisms (SNP) is more meaningful for the study of Alzheimer's disease (AD). In this paper, we used two frequent pattern mining (FPM) framework, the FP-Growth and Eclat algorithms, to analyze the GWAS results of functional magnetic resonance imaging (fMRI) phenotypes. Moreover, we applied the definition of confidence to FP-Growth and Eclat to enhance the FPM framework. By calculating the conditional probability of identified SNPs, we obtained the corresponding association rules to provide support confidence between these important SNPs. The resulting SNPs showed close correlation with hippocampus, memory, and AD. The experimental results also demonstrate that our framework is effective in identifying SNPs and provide candidate SNPs for further research.
Keyphrases
- genome wide
- genome wide association study
- dna methylation
- magnetic resonance imaging
- genome wide association
- copy number
- cognitive decline
- machine learning
- endothelial cells
- high resolution
- gene expression
- deep learning
- fluorescence imaging
- photodynamic therapy
- induced pluripotent stem cells
- blood brain barrier
- contrast enhanced