Small Animal Multivariate Brain Analysis (SAMBA) – a High Throughput Pipeline with a Validation Framework

Abstract

While many neuroscience questions aim to understand the human brain, much current knowledge has been gained using animal models, which replicate genetic, structural, and connectivity aspects of the human brain. While voxel-based analysis (VBA) of preclinical magnetic resonance images is widely-used, a thorough examination of the statistical robustness, stability, and error rates is hindered by high computational demands of processing large arrays, and the many parameters involved therein. Thus, workflows are often based on intuition or experience, while preclinical validation studies remain scarce. To increase throughput and reproducibility of quantitative small animal brain studies, we have developed a publicly shared, high throughput VBA pipeline in a high-performance computing environment, called SAMBA. The increased computational efficiency allowed large multidimensional arrays to be processed in 1–3 days—a task that previously took ~1 month. To quantify the variability and reliability of preclinical VBA in rodent models, we propose a validation framework consisting of morphological phantoms, and four metrics. This addresses several sources that impact VBA results, including registration and template construction strategies. We have used this framework to inform the VBA workflow parameters in a VBA study for a mouse model of epilepsy. We also present initial efforts towards standardizing small animal neuroimaging data in a similar fashion with human neuroimaging. We conclude that verifying the accuracy of VBA merits attention, and should be the focus of a broader effort within the community. The proposed framework promotes consistent quality assurance of VBA in preclinical neuroimaging, thus facilitating the creation and communication of robust results. Electronic supplementary material The online version of this article (10.1007/s12021-018-9410-0) contains supplementary material, which is available to authorized users.

Identifying Vulnerable Brain Networks in Mouse Models of Genetic Risk Factors for Late Onset Alzheimer’s Disease

“…Images were processed using a high-performance computing pipeline (Anderson et al, 2017(Anderson et al, , 2018a, to perform diffeomorphic mapping of a symmetric mouse brain atlas, containing 332 regions, based originally of the one presented in Calabrese et al (2015). To perform these processes we employed at the core of our pipeline advanced normalization tools (Avants et al, 2008(Avants et al, , 2011.…”

“…We have implemented code for tract based analyses 2 . The tracts connecting pairs of atlas regions (Anderson et al, 2018a) were used to build connectomes based on a constant solid angle (Q-Ball method) method implemented in DIPY (Garyfallidis et al, 2014). We used a relative peak ratio of 0.5, separation angle 25 • , and 4 parallel compute threads.…”

Optimizing Diffusion Imaging Protocols for Structural Connectomics in Mouse Models of Neurological Conditions

“…Further reductions in acquisition times come from compressed sensing [28], which several groups have implemented for mouse MRI [29][30][31]. Such advances can help translate diffusion protocols into population studies [32] incorporating multiple biomarkers from morphometry [33], microstructural properties based on diffusion [13] or magnetic susceptibility [34], or network properties [35]. Such integrative studies may better predict changes in behaviors, modeling those observed in humans with neurodegenerative conditions [34].…”

Theanine, the Main Amino Acid in Tea, Prevents Stress-Induced Brain Atrophy by Modifying Early Stress Responses

“…MDT computation was performed using the script in ANTs named "antsMultivariateTemplateConstruction2.sh" with the following SyN parameters: gradient step size = 0.1 voxels, update field variance = 3 voxels, and total field variance = 0.5 voxels. These SyN parameters were suggested by a recent mouse ex vivo brain study to account for a balance between registration quality and computation time [44]. Other parameters were as follows: iteration of template creation = 4, maximum iterations for each pairwise registration = 30 × 20 × 10, shrink factors = 4 × 2 × 1 voxels, smoothing factors = 2 × 1 × 0 voxels, similarity metric = cross-correlation, radius in brackets = 4 voxels, N4BiasFieldCorrection = on, and type of transformation model = Greedy SyN.…”