Bipolar disorder tends to run in families, but its genetics aren’t clear cut. The answer to the puzzle is unlikely to be a single gene; rather a mix of a number of different mutations, with environmental influences too. To try to solve the genetic puzzle, a group of Danish and Norwegian researchers have used data mining to see if they could find any significant patterns in combinations of genotypes. Any patterns or clusters could be useful for diagnosis of the disease, or as targets for new treatments.
The symptoms of bipolar disorder include mood swings and fluctuations in levels of energy and drive. It can range from mild to severe, and people with a parent or sibling with bipolar disorder are four to six times more likely to develop the mental illness.
Previous research has linked signal transduction pathways with bipolar disorder. The Scandinavian researchers focused on 803 SNPs (single nucleotide polymorphisms) in 55 genes related to signal transduction. In a genome-wide association study (GWAS), the team used GenoKey’s data mining software and technology to calculate all the combinations of three genotypes from the 3×803 SNP genotypes, in 1355 healthy people and 607 patients with bipolar disorder, and found four clusters of combinations of three SNPs. They checked their results against the independent WTCCC (Wellcome Trust Case Control Consortium) dataset for bipolar disorder, as well as testing it against random permutations on the general population. The researchers found different clusters of three SNPs in the two sets of data, which showed that there is a lot of variation in genetic mutations in bipolar disorder. The research was published in PLoS ONE.
As a next step, the researchers looked at these combinations and validated them against clinical data in a second paper in PLoS ONE. In this research, they found statistically-significant connections between the clusters of SNPs and the symptoms of bipolar disorder, particularly in patients with clustered manic episodes and alcohol-related bipolar episodes. There could also be a link between one of the clusters and more severe disease.
This could mean that it is possible to divide bipolar disorder into different genetic types and predict their symptoms and severity, but as this was a small study it’s not possible to draw any firm conclusions as yet. More studies with more clinical data, using powerful data mining technologies like GenoKey’s, are needed.