AGP
The Autism Genome Project
 "Investigating the genetic basis of autism"

AGP Research: Phase 1 Project

 

1. Current project: AGP Phase 2 (2007-):

CLICK HERE

2. AGP Phase 1 project (2004-2007):

2.1 Background

The first AGP project (Phase 1) was successfully concluded in early 2007.

In this first phase, the AGP achieved its initial goal of assembling the world’s largest gene bank for autism at this time.

This was also most comprehensive database of autism families, including both multiplex families (families with at least two affected individuals) and simplex families (one affected individual plus both parents).

The AGP also achieved its goal of carrying out the world’s most comprehensive genome scan into the genetics of autism at this time (see AGP publications).

2.2 Who did we study?

The AGP assembled a sample of over 1,400 multiplex families, with two or more individuals affected by Autism Spectrum Disorders (ASDs).

Extensive phenotypic information was gathered on the individuals in each family, including the following data:

DNA and phenotypic information from the families included in the AGP sample was collected by researchers from a large number of centres collaborating in the project.

2.3 How did we work?

Autism susceptibility loci have been identified on a number of chromosomal regions (including 2q, 7q and 17q).

Moreover, substantial evidence suggests that chromosomal abnormalities also contribute to autism risk.

According to the AGP model, autism risk is influenced by combinations of multiple loci that possibly interact, as well as microscopic or sub-microscopic chromosomal abnormalities.

This complicates the detection of individual autism risk loci and means that multiple strategies are required to localise autism susceptibility loci.

The AGP study was the first to attempt to merge linkage analysis with studies of fine-level chromosomal variation. (See: research techniques).

The AGP also developed an approach to identify sub-microscopic copy number variations (CNVs) as potential risk loci and as a tool to stratify the samples, in order to reduce genetic heterogeneity for linkage analyses.

A large sample of multiplex families was necessary to increase the likelihood of detecting susceptibility loci.

First, a complete-genome scan was carried out for the sample of over 1,400 multiplex families, using Affymetrix 10K v2 SNP arrays (10,000 SNPs).

A linkage analysis was then performed. This was the largest linkage scan to date for individuals with ASD.

Finally, a CNV analysis was performed to assess the CNV content of the samples. An algorithm was used to infer copy number from signal intensity of a SNP genotype, relative to intensity from other samples.

2.4 Our main research findings

See: AGP publications

Of the chromosomal regions that have been featured prominently in previous linkage analyses, the AGP study found only modest linkage support for regions 2q and 7q, and did not find support for region 17q.

However, the study did find evidence for linkage in the chromosomal region 11p12–p13, which had not previously been a major focus for discovery of autism risk loci. The linkage analyses also found support for linkage on regions 5p and 9p.

It is known that risk for ASD can in small part be attributed to chromosomal copy number abnormalities (CNAs).

The families in the AGP study were pre-screened for CNAs. However, the CNV analyses revealed evidence that relevant CNVs may also be a risk factor for autism.

Among a number of interesting discoveries, of particular note was the mizygous deletion of coding exons from the NRXN1 gene (2p16.3), whwhich was detected for a pair of affected siblings.

Other studies have also found rare variants of this gene in individuals with ASD. Moreover, this gene interacts with neuroligins, for which rare mutations appear to generate risk for ASDs and mental retardation.

There seems, therefore, to be accumulating evidence for a role for neurexins and neuroligins in ASDs. [See: neuroligin-3 (NLGN3) and neuroligin-4 (NLGN4) ].

TOP