Science: Mobile Elements Are More Common in Autism Genes

Post Reply New Topic

It really depends on the SNP. Some have a stronger relationship with autism than others. That core data set will probably be most helpful to you since they have a stronger relationship with autism. But there are truly few mutations that have known functional significance with autism and usually they are considered syndromic like fragile X. Of late I've seen much more of the literature that focuses on CNVs, but then again that could be simply because it's the latest thing. From what I've noticed many high risk autism genes house multiple types of mutations. It seems that the same mechanisms which underlie copy number variant occurrence may also underlie single-nucleotide polymorphisms.

A paper about TE frequency in autism genes, but since when have those genes been identified?

It depends on what you mean by "identified". There are MANY genes that have had weak-to-strong association with the conditions. The AutismKB database alone houses over 3,000 genes. Meanwhile, its core dataset of higher-risk associated genes (i.e., more reliably associated with the conditions) houses 171 genes in their most current version (#2).

That isn't to say that scientists know how even a minority of these genes may play some role in the phenotype (or whether these mutations actually do!), although there's plenty of hypotheses about synaptogenesis, cell proliferation, cell mobility, etc. But there has been a lot of work finding links between occurrence of the condition and occurrence of a given mutation or mutations within a given gene.

So they know there are likely relationships between autism and some of these genes, most likely with the stronger-association genes. BUT they still don't really know WHAT that relationship is. So have they "identified" genes? In a looser sense, definitely yes. Do they know how those genes play a role in the phenotype? Generally no, with some notable exceptions in syndromic cases like Fragile X.

Truthfully, autism genetics research is comparably lightyears behind fields like cancer research. So what the field could likely benefit from, aside from novel investigations, is a lot of translational research, meaning that you apply something you know in one field to another. From my vantage point, it seems that a lot of the people doing genetics work in the autism field weren't necessarily geneticists to begin with, OR that they're unfamiliar with a wide range of clinical genetics work. Just remember that a large number of autism researchers, especially those who are currently at the heights of their careers in their 40s, 50s, and 60s, are primarily clinicians. And that has a considerable effect on where the research leads. As newer generations of scientists take over though that may slowly change. At least I hope so.

While I really love studying genetics, I'm much more interested in how the genome interacts and adapts to its environment, affecting rates of mutations. I'm fascinated by how mutation rates and types can actually be driven by epigenetics and ecology. Which was why I focused in on TEs because they are a means for the DNA to alter its mutability based upon its molecular interactions, potentially providing a means for more rapid genetic adaptation.

Quite frankly, it is hard for me to see the functional value behind such a study in terms of understanding autism. I can see that it may be interesting to a biologist, maybe, to try to look at brain function from this angle, but imo it is a pipe dream.. You wrote on your website about having a holistic approach, as I recall, but this approach does not sound holistic. I do not see how looking at genes alone can help anyone figure out much about how the brain works, but yes, there may be some kind of factor of instability (or whatever you guys call it) in genes that plays into broader transactions in an integral way. I can see that to discover how that works might be exciting, but first I think it might be necessary to approach the subject from a holistic view, even if it is just a sketch and/or a theory, but I am not a scientist, so if I am missing something, please clue me in. ...

I do agree in approaching something from a broader perspective. Scientists have a habit of focusing in so closely on something that they can lose the bigger picture. (This is in part due to cognitive style, trees-before-the-forest thinking, and also the ways in which we actually get paid to do the work we do. Grants are not "general" and so scientists are financially rewarded for specializing and becoming "an expert" in their tiny little field.)

Despite what it may seem, I do actually have the bigger picture in mind. I don't believe all answers lie at the level of the gene-- in fact I believe most do not. Only a minority of autistics, after all, house any genes significantly associated with the condition, indicating that the majority of cases are likely due to phenotypic plasticity and less so genetic predisposition. But in order to know what IS actually genetic or not, we need to take a few steps back and figure out the basics. For one, we don't know under what specific circumstances these mutations arise, therefore learning how they mutate can give us some pointers as to why. Genes interact with the environment, not only expressively but mutationally. In the case of autism, it's important to know whether these mutations arose 1) in the sperm or egg in which case you need to be studying the parents' histories, not necessarily the child's, 2) early post-conception, pinpointing a different time frame altogether, or 3) may even be somatically mosaic, i.e., the mutation is not seen throughout organ systems but is instead relegated to a single or few systems.

Mutations can reasonably tell you what was occurring in the cell at the time of mutation. For instance, if Gene A is mutated in a specific way resulting in a SNP, a CNV, or some sort of repeat expansion, it is quite possible that Gene A was transcriptionally active (being expressed and possibly even overexpressed), because a common means for such mutations is expression, as is seen in cancer. As I mentioned in the blog post, transcriptional activation increases the torsional strain on the local DNA sequence, increasing the likelihood of acquiring mutation. The more frequent the expression, the greater the chances of mutation. Repetitive and regulatory content such as is seen in endogenous transposable elements play important roles in a gene's vulnerability to mutation; therefore if you combine transcription initiation with larger transposable content, the probability of acquiring mutation is comparably greater.

So while I most certainly agree with you that it's important for scientists to keep the bigger picture in mind at all times when asking questions, designing projects, and analyzing data, I do disagree that this is likely to be an unfruitful line of research into understanding how autism arises. Understanding how these mutations arise, under what circumstances, may help us trace back what originally occurred during the time of mutation. In circumstances in which an exogenous agent was present, say for instance some type of endocrine disruptor or an infection, we may better come to understand what was the domino that may've started the whole thing off. Because if you can know "what" you can start to figure out "how" and "why". In my mind, I don't see the mutation being the end-all-be-all of a phenotype. I see it merely as one player amongst many-- and in fact, just one target amongst many. Most agencies which alter cellular phenotype do not just promote mutation and then go on their merry ways. Valproic acid (VPA) for instance is a potential mutagen but it also acts, amongst other things, as a histone deacetylase (HDAC) inhibitor which subsequently upregulates the activities of a variety of pathways and can permanently alter the development of an organ system. In other words, it acts above genetics. Said organ system may house a mutation that is a result of VPA exposure but it's phenotype is not necessarily due to that mutation but to the inhibition of HDAC, or it may be a combination of both.

I like looking at mutations as the canary in the coal mine. Just one indication of what earlier took place. And that's what I'm trying to do: use mutations to read into the past.

Yes, in your mind or my mind:-) That is the key to intelligence---comprehensively processing data, and many people of just about any ilk cannot do it very well.. So if we understand that life evolved from some kind of primal slime, will that make us more intelligent? In effect you are making the argument that it will.

(Sorry I did not respond to this sooner....for me it is a tough subject to enquire into...)

Last edited by littlebee on 09 Jul 2013, 12:40 pm, edited 1 time in total.

If you are using Firefox, hold down the control key and keep on pressing the +/= key. Gradually, only the text will be visible. With IE, you can do the same thing from the menu or by holding down the control key and spinning the mouse wheel. Other browsers (and the Mac) have their own variations.

Sophist, thank you for sharing your work here. Your blog is really good, shows you're both a researcher and a great science communicator.
Thank you for having such an awesome background image, too.

A further comment: I am not discounting that this work has some potential value, perhaps in terms of preventing some kinds of brain damage or whatever you want to call it---this may not be quite the right term-- and I can see why a biologist might be passionately interested in this kind of study---in fact I find it very interesting, and Sophist, if you are reading, I am probably more in sync with you than it appears---much more actually, but I do not think this is the cause of most autism. and surely not the autism we see in the average person on this forum. I think to present this material as if it is the solution to certain problems such as lack of social adjustment is harmful to aspies. In fact you are not even necessarily doing that---but there is a slant from that angle and then people are kind of blowing it up...imo you do not have to justify your research---it is interesting and sounds valid, but some of the conclusions you come to by extrapolating various data, even though the data itself is valid, may be worth enquring into.

It seems you are talking about what is called secondary autism, and you surely know a lot more about that than me, but that is a rather small part of the autistic population, and yet it seems in the way you are characterizing your studies you are trying to include all of autism into it, but I have not reread this thread, so maybe I have forgotten....

littlebee