Genome-wide analysis identifies molecular systems and 149 genetic loci associated with income, 2019, Hill et al

Because individuals will vary genetically as a function of their ancestral lineage. If variation in lineage is genuinely random in a sample - and unconfounded with socioeconomic status - then it won't confound the results of a GWAS that looks at in income or SES (obviously these two a closely correlated).

I'm saying that in the UK, variation due to ancestral lineage is probably not random, that higher income earners are still likely to share more genetic material with historical groups such as the normans, and lower income earners with historically anglo saxon and celtic groups. This has nothing to do with which groups are smarter, its just that SES is still partly determined in the UK by the family you're born into. If this is the case, then the genetic variation attributable to these lineage factors will appear in GWAS as being the "genetic basis" of income/SES.

The study only looked at caucasians, which is why I've focused on the various historical caucasian groups in the UK.

I don't think I explained it very well. I am in no way saying that race and intelligence are linked, I'm saying these factors may create a statistical artefact. My ancestors were straight up anglo saxon and celtic (mostly the latter), really low class, as is the case for most Australians. The argument wouldn't be as strong if the study were done in Australia, because the social classes are slightly more dynamic, and there's (relatively speaking) more social mobility.

 

I agree. Given there is an inherited element to talents, and that society tends to reward some talents more than others, it seems inevitable there will be some correlation between genetics and income. The inevitable risk of course is that some people will draw erroneous information from the data, and others will probably make unintelligent policymaking decisions, but that is true of pretty much everything.

 

FWIW the reason I found this so interesting is because P4HA1 not only encodes for a part of procollagen-proline dioxygenase, which catalyzes the formation of hydroxyproline (a collagen precursor), it also plays a role in the body’s switch from aerobic to anaerobic metabolism under conditions of hypoxia (eg, low blood flow) via the downregulation of pyruvate dehydrogenase. This coincidence was striking. I did not know at the time the gene was only found in one person. (I still don’t understand how the math works on that in terms of statistical significance unless they were the only person in the entire biobank, including healthy controls, with this variant. But even then, it’s very confusing why that was even reported.) To me, it doesn’t matter that this is not an explanatory factor for all or even a subset of patients. I still wonder how this could contribute to an ME phenotype in that one individual.