Loss of function gene-set analysis of personal genome using pathway-disease similarity.

personal genome;gene-set analysis;loss of function variant;gene loss;genetic variation;610

의과대학 의과학과 ;

University:서울대학교 대학원

关键词: personal genome;    gene-set analysis;    loss of function variant;    gene loss;    genetic variation;    610;   

Others  :  http://s-space.snu.ac.kr/bitstream/10371/132961/1/000000003558.pdf

美国|英语

来源: Seoul National University Open Repository

【 摘 要 】

Personal genome projects discovered usual and unusual human variations and genetic factors associated with certain diseases. Genetic and monogenic diseases are actively researched among researchers and they have discovered many causative loci. Accordingly, variations that affect human phenotype are focused and discovered in many cases, however, a few mutations is deleterious to human health. Besides, mutations of essential genes are purified by selections. However, recent study identified that human genome carries up to 100 LoF variants predicted to affect transcript and 20 completely inactivated genes. Those of who carry broken genes still survive without severe clinical abnormalities. Healthy people always carry deleterious variants – severe recessive disease allele in the heterozygous state – and the variants will become causal disease variants within a few generations.Here we examined the potential effects of a set of LoF genes that share biological functions in clinical context. First, we found LoF genes of individual and calculate the proportion of shared genes between personal broken genes and each pathway member genes. Second, we queried disease-gene association database to identify genes with known associations with diseases and then compute pathway-disease similarity. Lastly, Pearson’s correlation was applied to calculate between personage-pathway and disease-pathway similarity vectors and we ordered by highest to lowest coefficient values of correlations. The higher coefficient value, the higher proportion of shared genes between phenotype associated gene list and individual LoF gene list. It suggests that a subject is highly connected to a certain disorder based on pathway information. We found that there is a higher correlation between an individual similarity vector and Parkinson disease similarity vector although none of individual LoF gene is associated with Parkinson disease directly. Our results suggest that LoF gene-set wise approach is useful for finding potential disease risk of personal genome and relevant biological features are helpful for clinical assessment.