The autism sequencing consortium: large-scale, high-throughput sequencing in autism spectrum disorders

Abstract

Research during the past decade has seen significant progress in the understanding of the genetic architecture of autism spectrum disorders (ASDs), with gene discovery accelerating as the characterization of genomic variation has become increasingly comprehensive. At the same time, this research has highlighted ongoing challenges. Here we address the enormous impact of high-throughput sequencing (HTS) on ASD gene discovery, outline a consensus view for leveraging this technology, and describe a large multisite collaboration developed to accomplish these goals. Similar approaches could prove effective for severe neurodevelopmental disorders more broadly.

Figure 1. Expected yield of identified ASD risk genes as a function of the number of trios (cases) evaluated for de novo LOF variants

Analyses (not shown) have demonstrated that for 8000 families a threshold of 3 or more de novo LOF variants (right panel) is sufficient to declare genomewide significance. Using 2 or more de novo LOF as a criteria (left panel) would identify ~80 likely ASD genes with an FDR of 0.1. As we learn more about annotation of truly damaging missense de novo variants, the rate of discovery will rise dramatically – and we anticipate that this will happen soon by our work and that of others such as the 1000 genomes project. In addition, other forms of discovery are not modeled but will lead to additional gene discovery. The analyses in this figure were taken from (Sanders et al., 2012).

Figure 2. A pathway for discovering novel risk genes for autism spectrum disorders (ASD)

Beyond the genes identified by de novo events, as illustrated in Fig. 1, inherited variation will also prove useful for identifying risk genes, the simplest to interpret being variation acting recessively. Potentially imbuing far greater discovery power will be two other sources of data: genetic data of various kinds, such as genomic regions already implicated in risk for ASD by copy number variation; and meta-data from other complementary fields, such a neuroscience and so-called “omics” (e.g., pathways of functionally related genes). With these additional sources of information it will be possible to identify a substantial fraction of ASD genes, providing sufficient grist for clinical geneticists to predict risk and for pharmacologists to develop therapeutics