Important New New Research Article on Autism Genetic Markers

[Loomis]

A research paper was just published last week that identifies 237 different genes that can be used to identify if an individual is diagnosed as Autistic or would be a NT. In my opinion this is one of the most significant genetic research articles on autism I have seen.

Preceding the abstract I define the term Single Nucleotide Polymorphism and give an example of a biochemical pathway. (I am not the best writer and struggled all my life with writing clear organized paragraphs. Hopefully I have done an adequate job here.) I have simplified the definitions and terminology so they are substantially accurate in general but a biologist would take exception to some of my word usage. My purpose is to make this information available to as many people as possible without requiring them to take a course in genetics. I have also edited the article's abstract to remove scientific jargon and make it more understandable to a non-scientific audience. I bracketed any terms I either added or changed from the original. An ellipse (...) indicates where I deleted material from the original.

Scientifically minded readers can go directly to the link at the bottom of the post.

Term defined:
SNP = Single Nucleotide Polymorphism. Genes are sequences of DNA that code for proteins. A gene occupies a specific place on a chromosome and consists of a stretch of DNA with a beginning and an end. For any given gene there are usually polymorphisms. This means that in a population a gene has more then one form. For example we all know that many different forms of the gene for eye color exist in human populations: brown, blue, green etc. The eye color gene is polymorphic in the human population. Or to say it a different way: more than one DNA sequence of the eye color gene exists in the human population. If there is only one spot (a nucleotide) in the DNA sequence that is different among the forms of a gene in a population it is called a single nucleotide polymorphism. This means there is only one coding point of DNA in the gene form that is different from the other forms. In general this means that when a gene is expressed in individuals with different forms of a gene each will have a slightly different protein in one of the biochemical pathways in their body.

Many processes in our body have very complex biochemical pathways. For example when a human body produces estrogen or testosterone, the process starts with cholesterol and goes through many intermediate molecules before it gets to its final form. Each step in the biochemical pathway is catalyzed (facilitated) by a protein enzyme that is coded by a unique stretch of DNA called a gene:

Cholesterol --> Intermediate A--> Intermediate B --> Intermediate C --> Intermediate D --> etc --> Estrogen or Testosterone. (As an aside these two hormones have very similar molecular structure.)

Abstract

"[The diagnosis of] Autism spectrum disorder depends on a clinical interview [because there are no known genetic markers] to aid [in the] diagnosis. [This research study looks at] single-nucleotide polymorphisms (SNPs) of individuals with ASD. [The] SNPs [we looked at] were mapped to ...[biochemical] pathways [that] identify [specific] cellular processes [in individuals with ASD]....

We created a genetic diagnostic [system] consisting of 237 SNPs in 146 genes that correctly predicted ASD diagnosis in 85.6% [individuals] of Central European [ancestry] .... [There were] eight SNPs in [particular]... [that] had the largest effect... with some [SNPs] acting [to make an individual more likely to have ASD], whereas others were protective [individuals having this SNP were more likely to be NT]....[Using a smaller subset of SNPs] our diagnostic [system] correctly predicted ASD diagnosis with an accuracy of 71.7% in [individuals] of Central European [ancestry].

In conclusion, we have developed an accurate diagnostic test for a [specific human population]...[and this]... approach identified cellular processes common to ASD across ethnicities."

Molecular Psychiatry - Predicting the diagnosis of autism spectrum disorder using gene pathway analysis

I just found this article today and have not had time to fully investigate the specific biochemical pathways involved in autism. I hope to do this in the future and present my results in a future post.

 

[Loomis]

I have put together a summary of the paper eliminating most of the scientific jargon while keeping the main data and inserting a few comments.

The researchers first assigned ethnicity to individual samples by comparing SNPs and grouping them to their closest ethnicity. Individuals of mixed ethnicity were excluded.

There were 3 groups:
Central European or Utah USA ancestry,
Tuscan Italian ancestry and
Hans Chinese ancestry.

Only 627 SNPs were different between the Tuscan Italian ancestry and the Central European/Utah ancestry groups. This shows there was no significant differences between SNPs in these two ancestry groups. However there were 116,753 different SNPs between the Central European/Utah ASD group and Han Chinese ASD group indicating these two populations differ significantly in SNPs.

The researchers next proceeded to only analyze and create a test specific to the Central European/Utah group.
For each individual of Central European/Utah ancestry 775-SNPs were examined to see if they could predict an individual?s clinical status (ASD versus non-ASD). From the 775 SNPs identified within the Central European/Utah ancestry group, an accurate genetic classification of ASD versus non-ASD was possible using 237 SNPs in 127 different genes determined to be highly significant.

Using the 237 SNPs in 127 genes comprising 13 biochemical pathways from the Central European/Utah analysis the researchers tried to identify ASD individuals in the following three test groups.

Three cohorts were used for validation:
1. 243 diagnosed ASDs and 42 controls of Central European/Utah ancestry;
2. 65 ASD and 88 controls of Tuscan Italian ancestry;
3. 33 ASDs and 169 controls of Han Chinese ancestry.

Classification accuracy for the 285 Central European/Utah ancestry validation cohort was 85.6% and 84.3% for the Tuscan Italian while accuracy for the Han Chinese cohort was only 56.4%.

The researchers looked at the biochemical pathways comprising the ASD SNPs for both Central European/Utah groups and the Han Chinese groups. They discovered the Hans Chinese ASD groups and the Central European/Utah ASD groups have different biochemical pathways that identify their ASD individuals. The researchers identified 13 ASD biochemical pathways in the Central European/Utah group and 16 ASD biochemical pathways in the Hans Chinese group but only six of these biochemical processes were common across both ethnic groups. Because of this disparity the genetic tests these researches created could not identify the Hans Chinese ASDs using the data they compiled from the Central European/Utah group. Their test was sub-optimal predicting only 56.4% of the Hans Chinese ASDs,, only slightly better than chance. It is likely that an additional set of SNPs may be predictive of an ASD diagnosis in Han Chinese and that methods used for developing the Central European/Uta test could be applicable to other ethnicities.

Nevertheless it is very important that six common pathways were identified across all ethnicities and further research can be expected to generate data to allow identification of ASDs in Hans Chinese populations. These researchers did not pursue that line of investigation in this paper. While the researchers did not address the question of mixed raced individuals, future research in my opinion will generate results to make this distinction.

A very important finding in this paper is that Eight SNPs in three genes, GRM5, GNAO1 and KCNMB4 were highly indicative of ASD status. The biochemical functions of these genes correspond to the ASD neuroanatomical differences that have been reported in many other scientific papers.

The KCNMB4 gene is a potassium channel that is important in neuronal excitability and has been implicated in epilepsy and dyskinesia. It is highly expressed within the fusiform gyrus, as well as in superior temporal, cingulate and orbitofrontal regions Microarray Data :: Allen Brain Atlas: Human Brain which are areas implicated in face identification and emotion face processing deficits seen in ASD.

The GNAO1 gene codes for a protein that is a subgroup of a G-protein that couples with many neurotransmitter receptors. Mice with the GNAO1 knocked out exhibit ?autism-like? features, including impaired social interaction, poor motor skills, anxiety and stereotypic turning behavior. GNAO1 has also been shown to have a role in nervous development at neuronal dendrites and synapses, and interacting with the GAP-43 gene at neuronal growth cones, with increased levels of GAP-43 demonstrated in the white matter adjacent to the anterior cingulate cortex in brains from ASD patients.

Different GRM5 SNPs have either an ASD contributory or a protective effect. the GRM5 gene is highly expressed in hippocampus, inferior temporal gyrus, inferior frontal gyrus and putamen regions implicated in ASD brain MRI studies. GRM5 has a role in synaptic plasticity, modulation of synaptic excitation, innate immune function and microglial activation. GRM5-effedts include stereotypies, sensory motor gating deficits and deficits in working, spatial and recognition memory. features described in ASD. With regard to GRM5?s involvement with neuroimmune function, this receptor is expressed on microglia with microglial activation demonstrated by us and others in frontal cortex in ASD. Further GRM5 may contribute to glutamatergic dysregulation in ASD.

 

[Margaretha]

Interesting, Loomis. Thanks for all the work you did on this.