You might be interested in epigenetic inheritence. We do know that some epigenetic marks are passed down but its still very much unknown how much heritable information is encoded in epigenetics.
While histones and methylation aren't DNA themselves, they're certainly incapable of functioning without DNA. I'd assume the parent poster was referring to further still mechanics.
> Does the fundamental limit of the amount of classical information encodable in the human genome (even with epigenetics & simultaneous encoding) imply a vast capacity for learning survival-beneficial patterns in very little time, with very few biasing priors?
> [Fundamental 'gbit' requirement 1: “No Simultaneous Encoding”:] if a gbit is used to perfectly encode one classical bit, it cannot simultaneously encode any further information. Two close variants of this are Zeilinger’s Principle (10) and Information Causality (11).
> Is there a proved presumption that genes only code in sequential combinations? Still overestimating the size of the powerset of all [totally-ordered] nonlocal combinations?
Still trying to understand counterfactuals in re: constructor theory
Observe the human response to dandelions; are they weeds or are they edible?
Do they have lobed leaves? What [neurons,] do mammals have to heuristically generalize according to visual and gustatory-olfactory features, and counterfactually which don't they have?
Or it's entirely learned, and then the coding for the substrate is still relevant