Our cautionary tale about functional neuroimaging genetics of language is a featured article in this week's Journal of Neuroscience @SfNJournals. Here's a thread that breaks down the story. 1/9 https://www.jneurosci.org/content/39/44/8778 …
Within a population, DNA sequences vary from one person to another. Millions of positions in the genome are polymorphic i.e. different people carry different nucleotides at the site. We can test whether variation at a polymorphic site is associated with a trait of interest. 2/9
Unlike rare mutations that dramatically disrupt gene function, the effect size of a single common DNA variant on a trait is usually tiny at best, even if in aggregate all polymorphisms across the genome make a big contribution. Welcome to the world of multifactorial genetics. 3/9
Behavioural/cognitive traits are genetically complex. To study roles of common DNA variants we need large cohorts, especially when testing multiple polymorphisms. Small samples are underpowered to detect a true association & they're also more susceptible to false positives. 4/9
Some argue that neuroimaging measures are closer to underlying biology & easier to link to gene variants than e.g. cognitive tests. An influential study (Pinel et al 2012) associated DNA polymorphisms with activation levels of language-related brain areas in 94 healthy adults.5/9
We tried to validate claims in independent cohorts of 427 people, 4 times size of prior study. Despite having power to detect much smaller effects than those reported by Pinel et al, no associations replicated. Bayesian stats => substantial/strong evidence supporting the null.6/9
Multiple studies took Pinel et al (2012) as motivation to test the highlighted polymorphisms for association with questionnaire data or behavioural/cognitive measures; results have not been compelling. None attempted a direct replication with a matching experimental paradigm.7/9
Our paper flags several factors that can yield false positives in neuroimaging genetics, e.g. i) not fully adjusting for testing multiple different brain areas & phenotypic definitions, ii) circularity of confirming top hits by reanalysing same data with alternative methods. 8/9
Sadly, your genes don't care whether you were tested with a pen-and-paper word list or an expensive sophisticated brain scanning technique. Neuroimaging genetics with task-based fMRI should be subject to same rigorous standards as studies of other complex human traits. 9/9
You can follow @ProfSimonFisher.
Tip: mention @threader_app on a Twitter thread with the keyword “compile” to get a link to it.
Enjoy Threader? Sign up.
Threader is an independent project created by only two developers. The site gets 500,000+ visits a month and our iOS Twitter client was featured as an App of the Day by Apple. Running this space is expensive and time consuming. If you find Threader useful, please consider supporting us to make it a sustainable project.