A new approach improves prioritization of disease-associated SNPs

December 5, 2008

The more often a gene is differentially expressed, the more likely it is to contain disease-associated DNA variants. Research published today in BioMed Central's open access journal Genome Biology shows how a list of SNPs in genes that are repeatedly implicated across many publicly-available gene expression microarray experiments (so-called, 'fitSNPs'), based on differential expression rates, can be used to successfully prioritize candidate genes for further research.

Atul Butte from Stanford University School of Medicine, USA, led a team of researchers who developed the new way to prioritize candidate SNPs from genome-wide association studies (GWAS). He said, "fitSNPs successfully distinguished true disease genes from false positives in genome-wide association studies looking at multiple diseases, and can serve as a powerful and convenient tool to prioritize disease genes from this type of study."

The hypothesis that there is an association between gene expression and disease-associated variants has never before been demonstrated with such clarity and at this global scale. The authors have robustly demonstrated that the likelihood of having variants associated with disease was 12 times higher among differentially expressed genes compared to constantly expressed genes. According to Butte, "As a case study, we looked at type 1 diabetes mellitus. We derived a list of fitSNPs to analyze the top seven loci of the Wellcome Trust Case Control Consortium type 1 diabetes mellitus (T1DM) genome-wide association studies. We then rediscovered all T1DM genes, and predicted a novel gene for a previously unexplained locus."

There are many candidate gene and SNP prioritization methods, and while the authors acknowledge that no single method is perfect, they suggest that using fitSNPs in a complementary fashion with other prioritization methods will significantly lower experimental costs.

Citation: FitSNPs: Highly differentially expressed genes are more likely to have variants associated with disease, Rong Chen, Alex A Morgan, Joel Dudley, Tarangini Deshpande, Li Li, Keiichi Kodama, Annie P Chiang and Atul J Butte, Genome Biology (in press)
genomebiology.com/

Source: BioMed Central

Explore further: OCD-like behavior linked to genetic mutation

Related Stories

OCD-like behavior linked to genetic mutation

February 22, 2017

A new Northwestern Medicine study found evidence suggesting how neural dysfunction in a certain region of the brain can lead to obsessive and repetitive behaviors much like obsessive-compulsive disorder (OCD).

Obesity reprogrammes muscle stem cells

February 22, 2017

Obesity is associated with reduced muscle mass and impaired metabolism. Epigenetic changes that affect the formation of new muscle cells may be a contributing factor, according to new research from Lund University, Sweden.

Heart risks in middle age boost dementia risk later in life

February 22, 2017

People who have heart disease risks in middle age - such as diabetes, high blood pressure or smoking - are at higher risk for dementia later in life, according to research presented at the American Stroke Association's International ...

Recommended for you

Neural networks promise sharpest ever images

February 22, 2017

Telescopes, the workhorse instruments of astronomy, are limited by the size of the mirror or lens they use. Using 'neural nets', a form of artificial intelligence, a group of Swiss researchers now have a way to push past ...

2 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

E_L_Earnhardt
not rated yet Dec 05, 2008
I assume "gene expression" is by electrons with specific speed and spin. More specific understanding is needed!
rong
not rated yet Dec 10, 2008
Gene expression means the expression levels of genes in the human body, which controls all physilogical property of our body, such as disease. We already have the expression levels of all genes in human for more than 300 diseases in the public domain. This article makes use of the data set and predicted disease-causing genes for 597 unknown diseases.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.