>>12370926Your picture doesn't include any p-values (before or after multiple comparison correction).
SNPs are single nucleotide polymorphisms. Your DNA is a sequence of nucleotides (A,T,G, and C). A SNP is a variant at one position of you DNA, ie A is replaced by T or C or G. Due to how DNA codes for proteins, sometimes SNPs have no effect on the protein produced (synonymous coding) and sometimes they do.
It may be a protein that metabolises fat, or a structural protein, or an ion channel. Maybe that SNP means the protein functions a little different compared to a different SNP at the same position on the DNA. Differences in proteins may have significant difference in function.
For example, in ashkenazim there are many sphingolipidoses. Genetic diseases which affect metabolism of sphingolipids. If you alter sphingolipid metabolism, you can cause neurons to branch out more, producing greater axon growth cones, and producing more dendrites, leading to greater rates of synaptogenesis, and this can be tested in a petri dish. Synaptogenesis has a lot to do with learning, since synapses are where neurons 'exchange information'. This is believed by some to explain the average IQ of ashkenazim being around 115.
In this case, there is a clear genetic difference in a small number of genes, leading to a clear functional difference in a few proteins, leading to a clear phenotypic difference in cellular activity.
Polygenic studies for SNPs by GWAS however are not as good evidence. They suggest genes we can look at, to see if the proteins differ and if this has an actual impact on he biological underpinnings of intelligence. On the other hand, anyone with a set of pairs of IQ scores and genomes will do an analysis like this. If a person doesn't find any SNPs they wont publish their paper, and if they do find some they will publish, but it could simply be false positives from statistical noise, or some confounding variable.
cont(1/?)
