Schizophrenia and Autism Genetics

If you've kept your eat to the proverbial structural variation ground you've undoubtedly noticed the continual attention that copy number variation (CNV) has been getting with respect to schizophrenia and autism. The very crude one-liner summary: Schizophrenic and autistic patients have a higher occurrence of rare and de novo copy number changes (Mental retardation also appears to show the same phenomenon). When these discoveries were made some years ago they were met with considerable hype in the scientific community. The major problem, however, is that we have known for a long time that schizophrenia is highly heritable. That means if your sibling is schizophrenic, you have a higher likelihood of showing symptoms also. So how do you reconcile the new information that says these conditions arise from rare and de novo mutations with the older information which supports a common allele phenomenon? The contradiction is not overly difficult to explain, because of the underlying complexity of these conditions, where both common and rare alleles can be pathogenic. Interestingly, in both reviews about 10% of schizophrenia and autism cases are attributed to rare CNVs, which is higher that what some other papers claim.

There are two review papers in press at Current Opinion in Genetics and Development detailing the genetics of these two conditions. In the schizophrenia review, the authors describe the situation:

Schizophrenia, like other common diseases, is a complex genetic disorder. It is likely that these result from a combination of relatively common alleles of small effect and some rare alleles with relatively large effects.

They summarize that up quite nicely in this plot:





The key parts are the upper-left (low frequency in the population but highly penetrant) and the lower-right (more frequent but with lower penetrance). GWAS stands for Genome-Wide Association Studies, which is basically a technique with tries to like specific alleles that are over-represented in a certain study group.

For Autism, the picture is drawn a little differently. From the autism review:

Currently, there are over 25 different loci that may be considered autism susceptibility candidate genes (ASCG) and many more implicated loci are under investigation. Most of these are rare Mendelian mutations, including CNVs or syndromic forms of autism, and only a few are dur to common genetic variation.

So there is the similar case of rare and common pathogenic alleles. In this review the authors cautiously point out the direct associations between ASCG gene products:




Such network diagrams are both much loved and hated and so the authors do themselves a favour to state "This is a first attempt to explore the connectivity between these genes, and therefore is not comprehensive. We do not mean to imply a single pathway as causative, as there may be many pathways that could be implicated." The review is a nice summary of what is known of ASCGs but given the complexities of both of these conditions, we are still a far cry from understanding their genetic etiologies.