As someone who has recently started learning about how we study the brain, the two most common forms of data I've encountered are neuroimaging data (fMRI scans, etc.) of the brain's distribution of activity in response to various external conditions, and a lesser extent gene expression, the degree to which genes are expressed in certain areas of the brain such as those responsible for serotonin transport. While I have encountered very detailed gene expression datasets such as the Allen Institute's heatmaps of the human brain, the vast majority of experimental research I've looked at seems to rely on neuroimaging data rather than RNA-Seq for obtaining its data.
Websites such as PsychENCODE, for instance, display data that seems to indicate genomic analysis can be used to trace the expression of genotypes in the brain to observed phenotypes associated with disorders such as ASD, for instance, yet for the vast majority of disorders like OCD or Tourette's Syndrome I'm primarily directed to research involving the use of neuroimaging data.
Suppose we identify genes strongly correlated with a particular disorder via a genome-wide association study that doesn't take into consideration the regions where said genes are expressed. What are the differences in the types of information obtained searching for where those genes are most strongly expressed in the brain vs. analyzing where the most significant degree of live activity occurs when searching for neurological bases of that disorder in the brain?
One distinction I had considered was the potential for predictive models created from mapping gene expression in the brain: for example, if a gene marker was found that correlated with hyperactive processing in regions it was expressed in, could mapping the areas in a sensory cortex where such processing occurs help create a predictive model for what behaviors sensory processing disorders may cause in an individual? And would a similar method be possible using neuroimaging data?
