Seems like I'm late to the game on this thread, but I must say that I don't love aspects of how genetics were sold by Dr. Harden.
She went kind of hard on a view of genetics that has come somewhat into vogue these days that we have done hundreds of GWAS studies and found links to thousands of genes so probably there aren't a small number of genes that control things but instead things are determined by the complex interaction of thousands of small differences. I think there is little evidence that this is true.
1) We don't know what most genes really do. Back in the day predicting planetary movements probably seemed super complicated with a bunch of esoteric rules where this one constellation moves one way unless another constellation stops then it moves the other way and the whole thing is reversed in winter. Then we discovered gravity and other explanations of planetary movement and it turns out predicting basic celestial movement is actually super explainable and follows reasonably simple rules. The same thing is happening with genetics - we don't know what all the genes do, so we talk about how a huge array of complicated and unrelated genetic forces are probably working to connect a bunch of GWAS hits in un-explainable ways, but we might figure out how one of the genes works tomorrow and suddenly it will all make sense and be super explainable and the crazy complicated hysteria will seem silly in retrospect.
2) GWAS studies look at these spots in the genome where people tend to differ by a single DNA letter, which actually tells you very little about what is happening biologically. Sometimes the change in DNA letter matters itself, but rarely in biology does a single letter change matter an insane amount (it happens, but not often). More often we track these single letters because bits of DNA that are close to each other are more likely to be inherited together so if a disease is always associated with a letter change then it is probably at least associated with a gene that lives close to that letter change in the genome. But there are a TON of things that could live close to that spot and figuring out the relevant difference is really hard and is usually only done for GWAS hits that are super gangbusters (like high risk alzheimer's GWAS hits). So when people say that we get a ton of hits and can't use them to explain certain traits, they are looking at reasonably low resolution data - data that basically says what chunks of the genome might be correlated with a trait - and it's not surprising when low resolution data can't explain why stuff works the way it does.
When people hear about GWAS studies, I think they often think we are looking at the genome, when in reality we are really looking at a relatively small number of hand-picked single letters throughout the genome. Sometimes scientists do something called "whole exome sequencing" and look at all the DNA that makes proteins, but that is much more expensive that the stuff GWAS uses and thus much rarer. Whole genome sequencing, which might actually get you somewhere, is more expensive yet. Epigenetic sequencing, which can tell you an extra level of complexity on top of whole genome sequencing, adds even more cost.
So all in all, I find it weird to claim that traits are the result of super complicated relationships between tons of genes when we are currently doing the genetic equivalent of super old school celestial mapping and wondering why it seems so complicated...
Whatever the reason for the end of your relationship with the Weeds, please consider starting your own podcast. I've really enjoyed your run as the Weeds host.
Seems like I'm late to the game on this thread, but I must say that I don't love aspects of how genetics were sold by Dr. Harden.
She went kind of hard on a view of genetics that has come somewhat into vogue these days that we have done hundreds of GWAS studies and found links to thousands of genes so probably there aren't a small number of genes that control things but instead things are determined by the complex interaction of thousands of small differences. I think there is little evidence that this is true.
1) We don't know what most genes really do. Back in the day predicting planetary movements probably seemed super complicated with a bunch of esoteric rules where this one constellation moves one way unless another constellation stops then it moves the other way and the whole thing is reversed in winter. Then we discovered gravity and other explanations of planetary movement and it turns out predicting basic celestial movement is actually super explainable and follows reasonably simple rules. The same thing is happening with genetics - we don't know what all the genes do, so we talk about how a huge array of complicated and unrelated genetic forces are probably working to connect a bunch of GWAS hits in un-explainable ways, but we might figure out how one of the genes works tomorrow and suddenly it will all make sense and be super explainable and the crazy complicated hysteria will seem silly in retrospect.
2) GWAS studies look at these spots in the genome where people tend to differ by a single DNA letter, which actually tells you very little about what is happening biologically. Sometimes the change in DNA letter matters itself, but rarely in biology does a single letter change matter an insane amount (it happens, but not often). More often we track these single letters because bits of DNA that are close to each other are more likely to be inherited together so if a disease is always associated with a letter change then it is probably at least associated with a gene that lives close to that letter change in the genome. But there are a TON of things that could live close to that spot and figuring out the relevant difference is really hard and is usually only done for GWAS hits that are super gangbusters (like high risk alzheimer's GWAS hits). So when people say that we get a ton of hits and can't use them to explain certain traits, they are looking at reasonably low resolution data - data that basically says what chunks of the genome might be correlated with a trait - and it's not surprising when low resolution data can't explain why stuff works the way it does.
When people hear about GWAS studies, I think they often think we are looking at the genome, when in reality we are really looking at a relatively small number of hand-picked single letters throughout the genome. Sometimes scientists do something called "whole exome sequencing" and look at all the DNA that makes proteins, but that is much more expensive that the stuff GWAS uses and thus much rarer. Whole genome sequencing, which might actually get you somewhere, is more expensive yet. Epigenetic sequencing, which can tell you an extra level of complexity on top of whole genome sequencing, adds even more cost.
So all in all, I find it weird to claim that traits are the result of super complicated relationships between tons of genes when we are currently doing the genetic equivalent of super old school celestial mapping and wondering why it seems so complicated...
Whatever the reason for the end of your relationship with the Weeds, please consider starting your own podcast. I've really enjoyed your run as the Weeds host.