Magazine article Issues in Science and Technology

Genetic Goose Chase

Magazine article Issues in Science and Technology

Genetic Goose Chase

Article excerpt

In "The Search for Schizophrenia Genes" (Issues, Winter 2016), Jonathan Leo notes that genetic searches at best have revealed a set of small-effect genomic variants that together explain less than 4% of schizophrenia liability. He adds that these susceptibility variants are close to equally common in the general population, with the differences appearing significant only when enormous sample sizes are assembled.

Although a 4% explained variance in schizophrenia liability has no practical utility, as Leo asserts, it is also a question whether even this small number is reliable. In a study central to the field--the 108 single nucleotide polymorphism (SNP) study, conducted by the Schizophrenia Working Group--results from the replication sample are telling, if one takes the effort to find them in the extensive online supplementary material. In the replication sample, as compared with the discovery sample, 13 of the 108 SNPs differed in the opposite direction between cases and controls; 87 failed to reach an uncorrected significance level of p < 0.05; and only three reached the adequate Bonferonni corrected significance level. Replication failures plague psychiatric genetics in general and indicate that the full hypothesis of no effect remains to be adequately rejected.

The validity of genomic sequence alterations in schizophrenia would be strengthened by demonstrating associations with the well-documented volumetric changes (usually shrinkage) that are seen in prefrontal cortical and subcortical brain regions in this disorder. In the hitherto largest attempt to do this, the Schizophrenia Working Group recently published a mega-analysis (Franke et al., Nature Neuroscience) of SNP associations to brain volumes in 11,840 patients and controls from 35 countries. No single SNPs or "polygenic risk scores" from the 108 study were significantly associated with size of the amygdala, hippocampus, thalamus, nucleus accumbens, or any other subcortical region. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.