Although there is growing interest in finding mouse models of human disease, no technique for quickly and quantitatively determining anatomical mutants currently exists. Magnetic resonance imaging (MRI) is ideally suited to probe fine structures in mice. This technology is three-dimensional, non-destructive and rapid compared to histopathology; hence MRI scientists have been able to create detailed three-dimensional images of 60 mum resolution or better. The data is digital which lends itself to sophisticated image processing algorithms. Here we show a variational MRI atlas constructed from nine excised brains of 8 week old 129S1/SvImJ male mice. This new type of atlas is comprised of an unbiased average brain--created from alignment of the individual brains--and the mathematical descriptors of anatomical variation across the individuals. We found that the majority of internal points in the individuals never varied more than 117 microm from equivalent points in the atlas. A three-dimensional annotation of the average image was performed and used to estimate the mean and standard deviation of volumes in a variety of structures across the individual brains; these volumes never differed by more than 5%. Our results indicate that variational atlases of inbred strains represent a well-defined basis against which mutant outliers can be readily compared.

No abstract

The neuronal glutamate transporter gene SLC1A1 is a candidate gene for obsessive-compulsive disorder (OCD) based on linkage studies and convergent evidence implicating glutamate in OCD etiology. The 3' end of SLC1A1 is the only genomic region with consistently demonstrated OCD association, especially when analyzing male-only probands. However, specific allele associations have not been consistently replicated, and recent OCD genome-wide association and meta-analysis studies have not incorporated all previously associated SLC1A1 SNPs. To clarify the nature of association between SLC1A1 and OCD, pooled analysis was performed on all available relevant raw study data, comprising a final sample of 815 trios, 306 cases and 634 controls. This revealed weak association between OCD and one of nine tested SLC1A1 polymorphisms (rs301443; uncorrected P = 0.046; non-significant corrected P). Secondary analyses of male-affecteds only (N = 358 trios and 133 cases) demonstrated modest association between OCD and a different SNP (rs12682807; uncorrected P = 0.012; non-significant corrected P). Findings of this meta-analysis are consistent with the trend of previous candidate gene studies in psychiatry and do not clarify the putative role of SLC1A1 in OCD pathophysiology. Nonetheless, it may be important to further examine the potential associations demonstrated in this amalgamated sample, especially since the SNPs with modest associations were not included in the more highly powered recent GWAS or in a past meta-analysis including five SLC1A1 polymorphisms. This study underscores the need for much larger sample sizes in future genetic association studies and suggests that next-generation sequencing may be beneficial in examining the potential role of rare variants in OCD.

No abstract