What is a genetic association study?

A genetic association study is, in essence, a case-control study (Hattersley et al, Lancet 2005) It examines the frequency of an allele in a particular gene in patients with the disease, compared against controls. Genetic association studies have proliferated in the past 5 years, particularly in the field of epilepsy, with the aim of understanding the common genes and polymorphisms that may increase risk of common epilepsies. Results of such studies have however been inconsistent (Tan et al, Epilepsia 2004)

Why epiGAD?

The goal of epiGAD is to collate all association studies in epilepsy, whether published or unpublished. This will help researchers in this area identify all the available gene-disease associations, as well as facilitate future meta-analyses and studies on publication bias (Munafo et al, Trends Genet 2004). It is also hoped that epiGAD will foster collaboration between the different epilepsy genetics groups around the world, and faciliate formation of a network of investigators in epilepsy genetics (Ioannidis et al, Am J Epi 2005)

epiGAD is funded by NMRC, Singapore, and also through an educational grant from UCB Pharma.

Navigating through epiGAD

There are 4 databases within epiGAD:

the susceptibility genes database
the epilepsy pharmacogenetics database
the meta-analysis database
the genome-wide association studies (GWAS) database

The susceptibility genes database compiles all studies related to putative epilepsy susceptibility genes (eg. interleukin-1-beta in TLE), while the pharmacogenetics studies in epilepsy (eg. ABCB1 studies) are stored in 'phamacogenetics'.

The meta-analysis database compiles all existing published epilepsy genetic meta-analyses, whether for susceptibility genes, or pharmacogenetics. The GWAS database is currently empty, but will be filled once GWAS ( Mullen et al, Neurology 2009) are published.

The navigation bar on the top allows you to select and browse through the epiGAD database of your choice. It also allows you to search through the database using a search phrase.

Structure of epiGAD

There are several data fields in epiGAD.

For the Susceptibility Gene database, from left to right, the fields are:

Broad phenotype: The cases in the association study are categorized into 3 main groups - Generalized epilepsy, Focal epilepsy, and Febrile convulsions
Narrow phenotype: The specific epilepsy syndrome eg. TLE, IGE
Gene Family: Gene family, as defined by the HGNC
Gene: The gene examined
Allele: The allele examined
No. of cases/ controls: as shown
Control source: The type of controls used - population-based (P), family-based (F), or both (PF). If both are used (PF), the number of cases and controls shown will be the number used for the population-based analysis.
Country of origin: The geographic origin of the study patients
P value: this is the p value of the allelic association
Pub Year: year this study was published
Ref:Right-clicking on this hyperlink will take you to the Pubmed entry for the study

For the Epilepsy Pharmacogenetics database, the fields are the same, except that "Broad Phenotype' is replaced by 'Cases', and 'Narrow Phenotype' is replaced by 'Controls'. This allows us to describe the studies more precisely - for example, is drug-resistant epilepsy being compared to drug-responsive epilepsy, or normal controls?

For the Meta Analysis database, the fields are 'cases' and 'controls', 'gene' and 'allele', followed by the number of cases&controls. The Summary Odds Ratio is then presented, together with the 95% confidence intervals, to 2 decimal places.

For all databases, clicking on the header for the column will allow you to sort alphabetically for that data field; a repeat click reverses the sort order

Searching epiGAD

You can search using the Basic or Advanced mode in all databases.

The Basic search function will search through all data field for the phrase entered.

The advanced search searches using phrases in specific data fields. For example, you may want to find out if there have been any studies of TLE in the UK. You would enter 'TLE' under 'Narrow phenotype', and 'UK' under 'Country of origin'. Besides the Boolean operator of 'AND', you can also use 'OR' to search.

Uploading & downloading data

epiGAD will also permit investigators to download the database in CSV format, which can easily be imported into spreadsheets or databases.

For approved investigators, it also allows you to upload files (in CSV format) containing data on association studies.

Version history