The limited success of GWAS for depression is in contrast with other psychiatric disorders, where established risk variants are accumulating through GWAS. For example, at the time of this writing, there are now more than 100 loci that have been associated with schizophrenia and bipolar disorder at stringent levels of statistical significance.101-106 Despite the fact that individual risk loci have not been identified for depression, we know that such variants will be found given adequate sample sizes. For example, it is now possible to use genome-wide complex trait analysis (GCTA) to estimate the common variant contribution to depression using genome-wide SNP data (these estimates are sometimes referred to as SNP-heritability).107 Through these methods, estimates of the common variant contribution to depression have ranged from a high of 32%108 to a low of 21%.109 It should be noted that these are lower bound estimates because SNP-chip heritability only reflects the effect of common variation that is captured on genotyping arrays.
