Polygenic and developmental profiles of autism differ by age at diagnosis ?

Your experience pretty much maps onto what happened to me. I was "normal" in primary school and when I could do things surrounded by parents. This really broke down in high school and has had really bad knock on effects for my whole life. It almost like I "become" autistic in certain situations but if I am in the right environment I can be "normal". I definitely fit into this late diagnosed category from this study and also identify with how the genetic pattern are more likely map onto conditions like depression. 

It is reflective of an increased interest among researchers in the autism field to move back towards formal sub-classification of autism. Recent decades have seen 'lumping' in the ascendance, hence the generalised diagnosis of ASD and the overarching idea of the spectrum. Now the 'splitters' are reasserting themselves and the genetic data, incomplete and tentative as it is, is tending to support this. Different genetic causation suggests different resulting conditions. I feel that autism will be increasingly viewed as a portmanteau term used to cover a number of distinct genetic conditions with some phenotypic overlap.

Reading through the report it seems the essence is that they believe 2 different genetic aspects are present for autists (for the purpose of this study) and that people with one aspect are most likely to present as autistic when in early childhood while the other group are likely to present much later, sometimes only in adulthood when pressures cause the autistic traits to become more pronounced.

Males showed a higher genetic corrolation for autism than females.

Concerns for me are that the childhood detection is a bit of a lottery from what I have read on this forum. Some schools have good SENCOs and some don't which can lead to later diagnoses based on which school they attend.

Hopefully this will improve with time so long as the government don't cut the budgets ( a bit of a big ask at the moment).

I note that they did not find any sort of corresponding age related presentation profle for ADHD people which is odd.

They do admit that the older diagnosed sample may be "contaminated" by misdiagnosed patients:
The higher genetic correlation between later-diagnosed autism and other mental-health conditions may indicate diagnostic misclassification (in which individuals with other conditions incorrectly receive an autism diagnosis) or diagnostic overshadowing (where the presence of co-occurring mental-health conditions can delay an autism diagnosis)

The research seem solid with a good sample size and they admit that they don't know what the real genetic links for autism are but the fact they can take a genetic profile and identify pobable autists can be a good and a bad thing.

Good because it gives parents advance warning to understand it, learn how to cope with it and hopefully to setup a proper care pipeline through education to allow the child to reach their best potential.

Bad because it gives a big neon sign over the babies head that they will probably be a net burdon on society and have a lower chance of being economically non-contributory in their adult life ( I'm basing this on the fact only 22% of autistic adults are in any kind of employment - NAS statistics).

The government under leadership like Trump and RFK Jr can then have the option to send them to "special care facilities" should they choose. A bit of a distopial extrapolation of things but I believe this is withing their potential to do.

It is a bit like the pre-crime division in Minority Report - they can now spot potential autists and remove them before they become a "problem".

However, this is all conjecture. The report is good research and has some interesting results.

You are welcome!

For me, it was pretty hard to follow, so I read the intro, the discussion, and bit and pieces of the rest.

It worked - thanks for that.

It was the Wired article that was paywall restricted but I have the report to read now

You can press download and it downloads, at least that seems to be the case for me. But also the first link I shared is to the complete online and open version of article.

Unknown said:

I am unsure the little overlap in genes involved fits this explanation that you provided.

I suspect there could be a link between autists ability to mask their behaviour when young (ie those who become later diagnised) and genetics which could explain this finding.

I'm skeptical but am keeping an open mind until I have seen the research itself (seems to be behind a paywall at the moment).

That is plausible. But I am unsure the little overlap in genes involved fits this explanation that you provided. What do you think?

It reflects the difference between children who spontaneously display autistic behaviours and children who only display autistic behaviours when their abilities to cope with social/societal pressures are exceeded.

Even if autism without intellectual disability had been diagnosable when I was a child, I would not have been diagnosed before I started school at four and a half, because I was a happy and sociable child at home and within my small close-knit community. When I started school I reacted to the situation by becoming selectively mute. This carried on for three months. Had this happened more recently, I would have been referred for assessment.