... The variants discovered by the two groups, one led by Dr. Kari Stefansson of Decode Genetics in Iceland and the other by Dr. Pamela Sklar of Massachusetts General Hospital, are rare. They substantially increase the risk of schizophrenia but account for a tiny fraction of the total number of cases.
This finding, coupled with the general lack of success so far in finding common variants for schizophrenia, raises the possibility that the genetic component of the disease is due to a large number of variants, each of which is very rare, rather than to a handful of common variants...
... The new focus on rare mutations suggests that natural selection is highly efficient at removing schizophrenia-causing genes from the population. Despite selection against the disease, according to this new idea, schizophrenia continues to appear because it is driven by a spate of new mutations that occur all the time in the population....
“This may be the case in other brain diseases, too,” Dr. Goldstein said, “because successful cognitive functioning is a highly complex system and there are many independent ways to take it down.”
The search for common variants in schizophrenia, however, has not been very successful so far, though not for want of trying. There have been more than a thousand studies, implicating 3,608 genetic variants.
But when all the data are pooled, only 24 of those variants turn out to be statistically significant, according to an analysis in the current issue of Nature Genetics by a group led by Dr. Lars Bertram of Massachusetts General Hospital...I read this as consistent with recent literature telling us that the human brain has been undergoing rapid evolution for a few thousand years, perhaps with an unusually accelerated mutation rage and high selection pressures.
Rapidly evolving organs are often fragile things -- like a car slapped together in a rush. Thousands of thousands of things can go wrong. Some are manageable, others produce schizophrenia and perhaps autism.
The bad news is we were hoping gene studies would identify common genetic mechanisms; we could then use these to characterize disease subtypes, better evaluate therapies, and identify new drug targets. Now that door seems to have shut. We have to look further up the bioinformatics chain for common functions that meds could act on, and we may fear that they'll be hard to find.