Final Report Summary - AUTISM MOLGEN (Using European and international populations to identify autism susceptibility loci)
The ultimate aim of the AUTISM MOLGEN project was to identify susceptibility alleles for Autism spectrum disorders (ASDs). For this purpose, the consortium aimed to pool data and laboratory resources, as well as exploit genetically isolated European populations, such as those from Finland and from northern Holland. These resources would then be used to examine ASD candidate loci using methods such as linkage, parent-of-origin analysis, association and mutation screening.
The project collection consisted of 102 multiplex and 372 singleton families would help achieve the additional sample numbers needed to dissect the genetic architecture of the ASDs and, thus, would provide a valuable resource for future studies. The molecular genetics portion of the project provided important results. This work was divided into linkage studies, association studies, as well as mutation screens. Recent advances in genotyping technology and subsequent decreases in cost allowed the advent of whole genome association studies. Despite this, linkage studies could still be of value in helping to identify and refine critical loci and, thus, identify putative candidate genes for further study. Towards this end, the project attempted to refine several linkage peaks previously identified by the project members. Disappointingly, LOD scores approaching suggestive evidence of linkage were only obtained for chromosome 2q. The lack of linkage signals in the analysis could be due to increased heterogeneity in the combined cohort. However, the positive result for chromosome 2q further confirmed the importance of this region in ASD susceptibility. Also of interest was that the subsequent Transmission disequilibrium test (TDT) analysis of the Single nucleotide polymorphisms (SNPs) genotyped as part of the linkage study in the Finnish and northern Dutch cohorts identified significant association in two genes, SND1 and MKL2. These associations survived correction for multiple testing and so represent interesting potential new candidate genes for future study.
In addition to the association study performed using the genotype results from the linkage study, 37 genes were examined for association with ASD. Of these, several nominal associations have been identified, but only a handful survived correction for multiple testing. This result was not surprising, as even for the most well-studied candidate genes in the scientific literature, such as RELN, multiple studies showing both association and lack of association were published. Therefore, for the majority of the genes investigated it could not be concluded that the data supported their role in ASD susceptibility, while it would not be appropriate to categorically reject them as unimportant in these disorders. However, evidence for the role of several genes in ASD susceptibility was amassed. The key genes amongst these were LRRTM3, LRRN1, LRRN3, SND1 and MKL2. These genes present interesting avenues for future research. In addition, while associations for genes such as JMJD2C, RELN, and GRIK2 may have not survived stringent correction for multiple testing, the associations observed were interesting and could be taken as part of a cumulative case for their involvement in ASDs.
The final molecular genetics aspect of the work was the mutation screening in candidate genes. This type of investigation became of particular relevance in ASD genetics due to the increasing numbers of reports implicating rare mutations of large effect influencing the development of these disorders, such as in the case of the neuroligins. Rare variants of possible functional effect were identified, including non-synonymous, splicesite and stop codon mutations, both in genes associated with ASD and in other plausible candidates. Such variants would serve as important starting points for future functional studies. In addition to such small variants, the importance of larger Copy number variants (CNVs) in ASD was also increasingly recognised. Although the genotyping data did not allow investigation for such variants, as it had in denser Single nucleotide polymorphism (SNP) genotyping studies, focussed CNV identification studies were performed. This resulted in discoveries such as the CNV at the IMMP2L-DOCK4 locus which had a putative functional effect.
Therefore, this project met its objectives and helped identify ASD susceptibility alleles, contributed to the evidence for previously recognised candidate genes and loci, generated vital new sample collections, as well as it identified avenues for possible future research.
The project collection consisted of 102 multiplex and 372 singleton families would help achieve the additional sample numbers needed to dissect the genetic architecture of the ASDs and, thus, would provide a valuable resource for future studies. The molecular genetics portion of the project provided important results. This work was divided into linkage studies, association studies, as well as mutation screens. Recent advances in genotyping technology and subsequent decreases in cost allowed the advent of whole genome association studies. Despite this, linkage studies could still be of value in helping to identify and refine critical loci and, thus, identify putative candidate genes for further study. Towards this end, the project attempted to refine several linkage peaks previously identified by the project members. Disappointingly, LOD scores approaching suggestive evidence of linkage were only obtained for chromosome 2q. The lack of linkage signals in the analysis could be due to increased heterogeneity in the combined cohort. However, the positive result for chromosome 2q further confirmed the importance of this region in ASD susceptibility. Also of interest was that the subsequent Transmission disequilibrium test (TDT) analysis of the Single nucleotide polymorphisms (SNPs) genotyped as part of the linkage study in the Finnish and northern Dutch cohorts identified significant association in two genes, SND1 and MKL2. These associations survived correction for multiple testing and so represent interesting potential new candidate genes for future study.
In addition to the association study performed using the genotype results from the linkage study, 37 genes were examined for association with ASD. Of these, several nominal associations have been identified, but only a handful survived correction for multiple testing. This result was not surprising, as even for the most well-studied candidate genes in the scientific literature, such as RELN, multiple studies showing both association and lack of association were published. Therefore, for the majority of the genes investigated it could not be concluded that the data supported their role in ASD susceptibility, while it would not be appropriate to categorically reject them as unimportant in these disorders. However, evidence for the role of several genes in ASD susceptibility was amassed. The key genes amongst these were LRRTM3, LRRN1, LRRN3, SND1 and MKL2. These genes present interesting avenues for future research. In addition, while associations for genes such as JMJD2C, RELN, and GRIK2 may have not survived stringent correction for multiple testing, the associations observed were interesting and could be taken as part of a cumulative case for their involvement in ASDs.
The final molecular genetics aspect of the work was the mutation screening in candidate genes. This type of investigation became of particular relevance in ASD genetics due to the increasing numbers of reports implicating rare mutations of large effect influencing the development of these disorders, such as in the case of the neuroligins. Rare variants of possible functional effect were identified, including non-synonymous, splicesite and stop codon mutations, both in genes associated with ASD and in other plausible candidates. Such variants would serve as important starting points for future functional studies. In addition to such small variants, the importance of larger Copy number variants (CNVs) in ASD was also increasingly recognised. Although the genotyping data did not allow investigation for such variants, as it had in denser Single nucleotide polymorphism (SNP) genotyping studies, focussed CNV identification studies were performed. This resulted in discoveries such as the CNV at the IMMP2L-DOCK4 locus which had a putative functional effect.
Therefore, this project met its objectives and helped identify ASD susceptibility alleles, contributed to the evidence for previously recognised candidate genes and loci, generated vital new sample collections, as well as it identified avenues for possible future research.
