A bioimage informatics platform for high-throughput embryo phenotyping

Abstract

High-throughput phenotyping is a cornerstone of numerous functional genomics projects. In recent years, imaging screens have become increasingly important in understanding gene-phenotype relationships in studies of cells, tissues and whole organisms. Three-dimensional (3D) imaging has risen to prominence in the field of developmental biology for its ability to capture whole embryo morphology and gene expression, as exemplified by the International Mouse Phenotyping Consortium (IMPC). Large volumes of image data are being acquired by multiple institutions around the world that encompass a range of modalities, proprietary software and metadata. To facilitate robust downstream analysis, images and metadata must be standardized to account for these differences. As an open scientific enterprise, making the data readily accessible is essential so that members of biomedical and clinical research communities can study the images for themselves without the need for highly specialized software or technical expertise. In this article, we present a platform of software tools that facilitate the upload, analysis and dissemination of 3D images for the IMPC. Over 750 reconstructions from 80 embryonic lethal and subviable lines have been captured to date, all of which are openly accessible at mousephenotype.org. Although designed for the IMPC, all software is available under an open-source licence for others to use and develop further. Ongoing developments aim to increase throughput and improve the analysis and dissemination of image data. Furthermore, we aim to ensure that images are searchable so that users can locate relevant images associated with genes, phenotypes or human diseases of interest.

Keywords: automated analysis; bioimage informatics; embryonic phenotyping; high-throughput; image processing; software tools.

Figure 1

The IMPC informatics platform. Following acquisition and reconstruction, image data are processed to reduce file size and converted to a standard format. Specimen and experiment data are downloaded at the DCC (PhenoDCC), ensuring they adhere to the standard operating procedures specified in IMPReSS (https://www.mousephenotype.org/impress/). Three-dimensional image data are downloaded separately, and undergo additional processing before being made available for display on the IMPC portal [http://www.mousephenotype.org/data/search/gene?fq=(embryo_data_available:%22true%22)]. If sufficient data are available for a knockout line, automated image analysis [1] is performed, and the results are made displayable in the web-based viewer. All data are open access and freely downloadable at full resolution for desktop visualization. Quality control of images and analysis results is conducted before export to the CDA for long-term storage.

Figure 2

Accessing 3D image data on the portal. (A) From mousephenotype.org, users can search for genes or phenotypes of interest in the search box. (B) On the search results page, users may then filter the list of genes returned to only those with 3D images available. (C) On selecting a gene of interest (e.g. http://www.mousephenotype.org/data/genes/MGI:1890602), the user will be able to view all phenotyping data associated with that gene. Clicking the ‘Embryo Viewer’ button will display the 3D image data. (D) Users can also view data from embryo-specific procedures on the same page, such as viability and lacZ expression (http://www.mousephenotype.org/data/genes/MGI:1890602#heatmap).

Figure 3

HARP is a desktop application for Windows and Linux. (A) The main application window. In the left panel, users can configure output directories and metadata. In the right panel, options for cropping, resizing and compression are provided. (B) A maximum intensity projection of a micro-CT image before (left) and after (right) automatic cropping. Here, the cropping process decreases the total file size from 56.8 to 4.5 GB (uncompressed). Image compression further reduces the file size to 3.6 GB, drastically reducing storage requirements.

Figure 4

A mouse embryo atlas at E14.5 dpc. This atlas is an average of 16 micro-CT images of C57Bl/6N (wild-type) embryos produced using in-house software. The average was segmented in 3D Slicer using our WASP plug-in, taking ∼8 h in total. The organs shown in this figure include the liver (LI), lungs (LU), dorsal root ganglia (D), trigeminal ganglion (T), humerus primordium (H), kidney (K), adrenal gland (A), stomach lumen (S) lateral brain ventricle (LV) and mesencephalic vesicle (M).

Figure 5

VPV is a desktop application for visualizing the results of 3D image registration. Designed to accompany the automated analysis pipeline, VPV brings together many of the most desirable features found across existing visualization tools into a simple, intuitive graphical user interface. The screenshot shows the software being used to view the results of analysing Cbx2 knockout embryos imaged by micro-CT. Users can simply drag and drop their images and analyse data to instantly view regions of significant dysmorphology as heatmaps, shown in red/blue overlays. Deformation (vector) fields can also be overlaid to observe how the registration has behaved, as shown in green.

Figure 6

IEV is a web-based tool that allows users to directly compare wild-type and mutant embryos in a dynamic, intuitive manner. Users can navigate the volumes in synchrony by using the sliders, mouse wheel or crosshair tool (shift key while moving the mouse). The settings panel provides options to toggle viewports on/off, configure the scale bar, zoom in/out, create bookmarks and switch modality. Each viewer has its own drop-down box listing the embryos available to view, an image contrast slider, and the ability to link/unlink individual viewports. Other features include fully interactive 3D renderings, ability to bookmark viewer state and share them and full resolution downloads. Automated analysis results can also be displayed where available, as shown here in the top viewer for the gene Cbx4 (http://www.mousephenotype.org/embryoviewer/?mgi=MGI:1195985&wn=Average). Images and analysis results were produced by The Centre for Phenogenomics, Toronto. A user guide can be viewed and downloaded by clicking the Help tab.