We expect that the DSPN will improve accuracy mainly for complex traits with a highly polygenic architecture, but not necessarily for traits that are strongly determined by only a few variants, such as Mendelian disorders, or are closely correlated with population structure, such as ethnicity. However, even when the DSPN performance is low, it may still provide insights about intermediate phenotypes; for instance, in our analysis, the PFC transcriptome appears substantially less predictive with respect to gender (after removing the sex chromosome genes) than age, but this very fact highlights the similarity of the transcriptome between sexes (60). Finally, although our focus has been on common SNPs, the DSPN may be able to capture the effects of rare variants, such as those known to be implicated in ASD (51), through their influence on intermediate phenotypes.
