|
 |
|
      |
|
|
|
 Research & Giving News Article  Now, a sophisticated statistical assessment has been published in the journal Nature Genetics of a database containing information on over 500 genes that have been evaluated in schizophrenia. The study winnows that long list of suspect genes to a handful of culprits deemed most likely to actually be involved in causation of the illness. Comprehensive schizophrenia database supported by NARSAD The database that gives rise to the new study, called SchizophreniaGene, or SzGene for short, has received major support from NARSAD since its inception several years ago. It is the work of several teams of investigators in the United States and overseas, led by Lars Bertram, M.D., Ph.D., of Harvard Medical School and Massachusetts General Hospital, the principal investigator of the project. The SzGene database resides online (www.schizophreniaforum.org/res/sczgene) within the website of the Schizophrenia Research Forum, or SRF, which also receives major support from NARSAD. Dr. Bertram and the team involved in compilation and interpretation of the SzGene database are co-authors of the Nature Genetics paper published in July. For the article, they systematically analyzed all of the genetic association papers catalogued in the SzGene database through April 2007, some 1,179 in all. These papers cumulatively reported 3,608 different gene variations with possible associations to schizophrenia. These variations involved 516 different genes. (In many cases, different variations were reported within the same genes.) The meaning of 1,179 different studies But what to make of the nearly 1,179 different studies? “Despite these efforts [by researchers], no single genetic variant has been established as a bona fide schizophrenia susceptibility gene,” Dr. Bertram and colleagues write, “at least not with the confidence accorded to other genes associated with susceptibility to complex disease[s],” including Alzheimer’s disease and macular degeneration. Most of the 1,179 studies identify specific genetic markers with possible links to schizophrenia risk. Such studies, which often involve several hundred schizophrenia patients and a like number of controls, “usually do not themselves have the power to reliably determine whether there is a significant effect in the sample,” Dr. Bertram says. The solution, in statistical terms, is to increase sample size, and in so doing, “increase the power to make an assertion” about whether or to what degree the data for specific markers are significant. The technique used by Dr. Bertram and colleagues in assessing past studies is called meta-analysis. It is an established statistical procedure that quantitatively summarizes data across a set of studies on a given genetic marker by combining their data and thereby effectively increasing sample size. Sifting and re-sifting the results Of the 3,608 genetic variations cited in SzGene’s 1,179 studies, only 118 variants appearing in 52 genes were deemed by Dr. Bertram and colleagues of sufficient “power” to warrant meta-analysis. These 118 gene variants were in turn sifted using various mathematical and other criteria. The team was left with 24 gene variants residing in 16 different genes that showed significant risk effects in the meta-analyses. (The complete human genome has over 20,000 genes). Overall, the effect exerted by these 24 variants was very small, increasing or decreasing the risk for schizophrenia by a mere 20 percent on average. In combination, however, co-inheritance of some of these risk genes, potentially together with exposure to certain environmental risk factors, could determine whether or not an individual is prone to develop schizophrenia. Despite these promising leads, more independent genetic and functional genomic data is needed to determine which of the current SzGene “top results” will prove to be genuine. A final sifting, using a stringent set of newly proposed criteria developed by collaborators from the Human Genome Epidemiology Network (HuGENet) was applied at this final stage. It reduced the number of highlighted genes to only four from the prior set of 16. These four genes were deemed by Drs. Bertram and colleagues to have the highest degree of “epidemiologic credibility” for relevance to schizophrenia, at least when applying the HuGENet criteria. One, called DRD1, directs brain cells to manufacture a type of cellular receptor for the neurotransmitter dopamine. “This receptor is thought to have a role in regulation of cognitive functions in the prefrontal cortex” of the brain, the team wrote. They also noted that the same receptor is believed to play an important role in the way patients respond to an important “atypical” anti-psychotic drug called clozapine. In this way, the team’s meta-analysis of a large portion of published research on schizophrenia genetics has the potential to help researchers focus on the most promising candidate susceptibility genes. Their method, they note, also “can be easily adapted to genetic association studies of other common diseases of public health significance.” This has already been done by Dr. Bertram and his team, who have created similar websites for genetic studies in Alzheimer’s disease (www.alzgene.org) and Parkinson’s disease (www.pdgene.org). |
Media Contact
Latest News from NARSAD
Spotlight
|
|
For the Media