Thread by @ewanbirney

Some epigenetics reminders post this really poorly written @BBCFuture piece. First off, epigenetics means (sadly) many different things, even to established scientists. It might mean DNA and Histone modification (a mechanism). It might mean cell state information (liver vs hair)

It might mean cell state as cellular memory. It might mean "some funky inheritance pattern from parents to child or complex, non standard genetic behaviour." and it might mean "Non DNA based inheritance".

(there is one final definition - everything which is not genetic, which is a really weird definition as it is pretty much all of biology).

Even well established scientists often have to stop each other and ask "what do you really mean by epigenetics here" and loop back to confirm which of the above definitions the other scientist is using. Yup, it's exhausting.

Pretty much uniformly though the use of "epigenetics" in the broader, non science, world is associated with an often vague "there is more to life than DNA as a fixed destiny", often linked to transgenerational inheritance, or environmental impact

The @BBCFuture one was on this topic of "inherieted trauma" via epigenetics (it is so misguided...)

First off, the popular understanding of genetics has a big overreach (and there is a whole possible thread that geneticists don't help this in our enthusiasm for our science); geneticists are completely comfortable with non-genetic stuff happening

DNA is *not* your destiny (as I've often stressed) - it is just often really useful - firstly for research and steadily coming into healthcare, but ... it doesn't mean you shouldn't exercise, choose what you eat, be happy, do what you want.

So - you don't need a fancy science term to say you are more than your DNA - you are - (and this is from a geneticist) - and you don't need a fancy science term to describe the impact of behaviour in families being passed down. We call that "behaviours" or "family behaviours"

If you want to know, multi-generational transgenerational epigenetics is rare (but observable) in plants. It is very rare in mammals, with only one good locus (Agouti coat colour) that is

Anne Fergusson-Smith, head of genetics in Cambridge, one of the world's leading epigeneticists and an absolutely no nonsense glaswegian has looked super carefully for more loci like this in Mouse. Not found one more *trans generational*. ncbi.nlm.nih.gov/pubmed/30454646

Some caveats here. Mammals have two well known properties. The first is that the female germline is laid down in the developing foetus and kept in stasis until the menstrual cycle. This means there is a direct cellular connection from every mammal to their maternal grandmother

The oocyte (unfertilized egg) that made you had a key cell division whilst it was in your grandmother, in the developing body of your mother. There is a little bit of you that existed way back then (in my case... 1944). Weird, but true.

Second, nearly all our genomes go through "epigenetic reprogramming" in the germline (meaning - removing all DNA modification and then adding special "you are sperm or egg" modifications) with the extra twist that at some places the eggs are programmed one way, sperm another

This parent-of-origin reprogramming has some striking and usually lethal effects when it goes wrong and some specific subtle effects on some phenotypes (the most famous being certain types of baldness which come only from the mother's line due to this).

This "parent of origin" imprinting is executed via epigenetics (in the "DNA modification" sense) and both this and the oocyte aspect means 1 or 2 generational information transfer happens all the time. But not as far as we know, multi-generational (except Agouti).

The Dutch famine, Holocaust survivors and other stories have more straightforward and clear cut explanations via both change in behaviour in parents (most simple) or these 1 or 2 generational exposure aspects.

Moral of the story - scientists find the word "epigenetics" complex. Your baseline is that you should be very skeptical about human trans generational epigenetics (its not formally impossible; it is though super unlikely).