Jyl Boline

Digital Atlasing and Standardization in the Mouse Brain

Digital brain atlases are used in neuroscience to characterize the spatial organization of neuronal structures [1]–[3], for planning and guidance during neurosurgery [4], [5], and as a reference for interpreting other modalities such as gene expression or proteomic data [6]–[9]. The field of digital atlasing is extensive, and includes high quality brain atlases of the mouse [10], rat [11], rhesus macaque [12], human [13], [14], and several other model organisms. In addition to atlases based on histology, [11], [15], [16], magnetic resonance imaging [10], [17], and positron emission tomography [11], modern digital atlases often use probabilistic and multimodal techniques [18], [19], as well as sophisticated visualization software [20], [21]. Whether atlases involve detailed visualization of structures of a single or small group of specimens [6], [22], [23] or averages over larger populations [18], [24], much of the work in developing digital brain atlases is from the perspective of the user of a single resource. This is often due largely to the challenges of data generation, maintenance, and resources management [25], [26]. A more recent goal of many neuroscientists is to connect multiple and diverse resources to work in a collaborative manner using an atlas based framework [2], [19]. This vision is appealing as, ideally, researchers would be able to share their data and analyses with others, regardless of where they or the data are located. An important step in this direction is the specification of a common frame of reference across specimens and resources (either as coordinate, ontology, or region of interest) that is adopted by the community. In this perspective, we propose a collaborative digital atlasing framework for coordinating mouse brain research that allows access to data, tools, and analyses from multiple sources.

Cerebellar cortical atrophy in experimental autoimmune encephalomyelitis

Brain atrophy measured by MRI is an important correlate with clinical disability and disease duration in multiple sclerosis (MS). Unfortunately, neuropathologic mechanisms which lead to this grey matter atrophy remain unknown. The objective of this study was to determine whether brain atrophy occurs in the mouse model, experimental autoimmune encephalomyelitis (EAE). Postmortem high-resolution T2-weighted magnetic resonance microscopy (MRM) images from 32 mouse brains (21 EAE and 11 control) were collected. A minimum deformation atlas was constructed and a deformable atlas approach was used to quantify volumetric changes in neuroanatomical structures. A significant decrease in the mean cerebellar cortex volume in mice with late EAE (48-56 days after disease induction) as compared to normal strain, gender, and age-matched controls was observed. There was a direct correlation between cerebellar cortical atrophy and disease duration. At an early time point in disease, 15 days after disease induction, cerebellar white matter lesions were detected by both histology and MRM. These data demonstrate that myelin-specific autoimmune responses can lead to grey matter atrophy in an otherwise normal CNS. The model described herein can now be used to investigate neuropathologic mechanisms that lead to the development of gray matter atrophy in this setting.

Anatomical landmarks for registration of experimental image data to volumetric rodent brain atlasing templates.

BACKGROUND Assignment of anatomical reference is a key step in integration of the rapidly expanding collection of rodent brain data. Landmark-based registration facilitates spatial anchoring of diverse types of data not suitable for automated methods operating on voxel-based image information. NEW TOOL Here we propose a standardized set of anatomical landmarks for registration of whole brain imaging datasets from the mouse and rat brain, and in particular for integration of experimental image data in Waxholm Space (WHS). RESULTS Sixteen internal landmarks of the C57BL/6J mouse brain have been reliably identified: by different individuals, independent of their experience in anatomy; across different MRI contrasts (T1, T2, T2(*)) and other modalities (Nissl histology and block-face anatomy); in different specimens; in different slice acquisition angles; and in different image resolutions. We present a registration example between T1-weighted MRI and the mouse WHS template using these landmarks and reaching fairly high accuracy. Landmark positions identified in the mouse WHS template are shared through the Scalable Brain Atlas, accompanied by graphical and textual guidelines for locating each landmark. We identified 14 of the 16 landmarks in the WHS template for the Sprague Dawley rat. COMPARISON WITH EXISTING METHODS This landmark set can withstand substantial differences in acquisition angle, imaging modality, and is less vulnerable to subjectivity. CONCLUSIONS This facilitates registration of multimodal 3D brain data to standard coordinate spaces for mouse and rat brain taking a step toward the creation of a common rodent reference system; raising data sharing to a qualitatively higher level.

Digital atlases as a framework for data sharing

Digital brain atlases are useful as references, analytical tools, and as a data integration framework. As a result, they and their supporting tools are being recognized as potentially useful resources in the movement toward data sharing. Several projects are connecting infrastructure to these tools which facilitate sharing, managing, and retrieving data of different types, scale, and even location. With these in place, we have the ability to combine, analyze, and interpret these data in a manner not previously possible, opening the door to examine issues in new and exciting ways, and potentially leading to speedier discovery of answers as well as new questions about the brain. Here we discuss recent efforts in the use of digital mouse atlases for data sharing.

Brain atlases and neuroanatomic imaging.

Quantifying the effect of a genetic manipulation or disease is a complicated process in a population of animals. Probabilistic brain atlases can capture population variability and be used to quantify those variations in anatomy as measured by structural imaging. Minimum deformation atlases (MDAs), a subclass of probabilistic atlases, are intensity-based averages of a collection of scans in a common space unbiased by selection of a single target image. Here, we describe a method for generating an MDA from a set of magnetic resonance microscopy images. First, the images are segmented to remove any non-brain tissue and bias field corrected to remove field inhomogeneities. The corrected images are then linearly aligned to a representative scan, the geometric mean of all the transformations is calculated, and a minimum deformation target (MDT) is produced by averaging the volumes in this new space. The brains are then non-linearly aligned to the MDT to produce the MDA. Finally, the images are linearly aligned to the MDA using a full-affine transformation to spatially and intensity normalize them, removing global differences in size, shape, and position but retaining anatomically significant differences.

Cyberinfrastructure for the digital brain: spatial standards for integrating rodent brain atlases.

Biomedical research entails capture and analysis of massive data volumes and new discoveries arise from data-integration and mining. This is only possible if data can be mapped onto a common framework such as the genome for genomic data. In neuroscience, the framework is intrinsically spatial and based on a number of paper atlases. This cannot meet todays data-intensive analysis and integration challenges. A scalable and extensible software infrastructure that is standards based but open for novel data and resources, is required for integrating information such as signal distributions, gene-expression, neuronal connectivity, electrophysiology, anatomy, and developmental processes. Therefore, the International Neuroinformatics Coordinating Facility (INCF) initiated the development of a spatial framework for neuroscience data integration with an associated Digital Atlasing Infrastructure (DAI). A prototype implementation of this infrastructure for the rodent brain is reported here. The infrastructure is based on a collection of reference spaces to which data is mapped at the required resolution, such as the Waxholm Space (WHS), a 3D reconstruction of the brain generated using high-resolution, multi-channel microMRI. The core standards of the digital atlasing service-oriented infrastructure include Waxholm Markup Language (WaxML): XML schema expressing a uniform information model for key elements such as coordinate systems, transformations, points of interest (POI)s, labels, and annotations; and Atlas Web Services: interfaces for querying and updating atlas data. The services return WaxML-encoded documents with information about capabilities, spatial reference systems (SRSs) and structures, and execute coordinate transformations and POI-based requests. Key elements of INCF-DAI cyberinfrastructure have been prototyped for both mouse and rat brain atlas sources, including the Allen Mouse Brain Atlas, UCSD Cell-Centered Database, and Edinburgh Mouse Atlas Project.

A High-Resolution Anatomical Framework of the Neonatal Mouse Brain for Managing Gene Expression Data

This study aims to provide a high-resolution atlas and use it as an anatomical framework to localize the gene expression data for mouse brain on postnatal day 0 (P0). A color Nissl-stained volume with a resolution of 13.3 × 50 × 13.3 μ3 was constructed and co-registered to a standard anatomical space defined by an averaged geometry of C57BL/6J P0 mouse brains. A 145 anatomical structures were delineated based on the histological images. Anatomical relationships of delineated structures were established based on the hierarchical relations defined in the atlas of adult mouse brain (MacKenzie-Graham et al., 2004) so the P0 atlas can be related to the database associated with the adult atlas. The co-registered multimodal atlas as well as the original anatomical delineations is available for download at http://www.loni.ucla.edu/Atlases/. The region-specific anatomical framework based on the neonatal atlas allows for the analysis of gene activity within a high-resolution anatomical space at an early developmental stage. We demonstrated the potential application of this framework by incorporating gene expression data generated using in situ hybridization to the atlas space. By normalizing the gene expression patterns revealed by different images, experimental results from separate studies can be compared and summarized in an anatomical context. Co-displaying multiple registered datasets in the atlas space allows for 3D reconstruction of the co-expression patterns of the different genes in the atlas space, hence providing better insight into the relationship between the differentiated distribution pattern of gene products and specific anatomical systems.

Registration workflows for the creation of INCF digital atlas hubs

Workflows are being developed around specific data sharing use cases (Figure 2a). At this time, the use cases focus on 2D brain slice images (some sparsely, others highly sampled) of various modalities. The goal is to create tools, recommendations, and standard operating procedures to aid in the registration of data to a known standard atlas space (Figure 2b) and ability to share that data through INCF atlas hubs or to create new hubs (Figures 2c and 3). More information can be found at http://atlasing.incf.org/wiki/Workflow.