MISA++: a standardized interface for automated high-performance big volume image analysis

CONTACT [email protected] Research Group Systems Biology / Bioinformatics References: [1] Surname, N., Surname, N. N., and Surname, N. N. (2008). Nature Reviews Immunology Picture of you CONTACT [email protected] Research Group Applied Systems Biology References: [1] Anika Klingberg et al., “Fully Automated Evaluation of Total Glomerular Number and Capillary Tuft Size in Nephritic Kidneys Using Lightsheet Microscopy,” Journal of the American Society of Nephrology: JASN 28, no. 2 (February 2017): 452–59, https://doi.org/10.1681/ASN.2016020232. MISA++: a standardized interface for automated high-performance big volume image analysis Ruman Gerst1,2, Anna Medyukhina1, and Marc Thilo Figge1,3 Light-sheet fluorescence microscopy (LSFM) allows quantitative threedimensional analysis of whole organs. This includes the evaluation of structural changes such as a reduced number of glomeruli in kidneys [1]. 4. Standardized data management 1. Implementing a big volume image analysis 1 Applied Systems Biology, Leibniz Institute for Natural Product Research and Infection Biology – Hans-Knöll-Institute, Jena, Germany 2 Faculty of Biological Sciences, Friedrich-Schiller-University Jena, Germany 3 Institute of Microbiology, Faculty of Biological Sciences, Friedrich-Schiller-University Jena, Germany Whole murine kidneys 23 kidneys 110 GB x z yRegions of interest e.g. Glomeruli Quantification results e.g. Volume 41 min 3 days 4 days Big volume image data analysis is hardware-intensive and requires highperformance implementations in efficient languages such as C++. 2. MISA++ - A platform for custom C++ analysis tools Your software 1. C++ framework 2. JSON interface Parallelization Modularization CLI Memory management Parameters Data organization Data types Quantification results Easy development Easy integration Easily extendable Ready-to-use components for common tasks Standardized JSON interface for data & parameters Integration of custom data types & more Comes with: 3. Parallelization and modularization 6. Standardized parameters & documentation Software library Task Task Task Task Your app Task Task Task Task Root Task Workload is organized in a directed acyclic graph (DAG) where nodes represent tasks and edges represent dependencies between two tasks. MISA++ automatically runs tasks with satisfied dependencies in parallel until the work is done. Easy modularization A set of tasks can be separated into a software library to be developed independently. The DAG of any MISA++ application can be re-used by another application via a standardized modularization interface. Easy graph construction MISA++ comes with optional functions to simplify DAG creation and enhance code readability. Task in DAG Thread-safe Read/Write Data cache C++ representation Metadata Folder I/O MISA++ wraps custom/third-party data types within structures termed caches. Automated I/O A cache is associated to a location on the hard drive that is used to store currently unused data. On accessing a cache, it automatically loads the data from the hard drive location. Thread-safe access Tasks can access cached data via thread-safe functions. Attaching metadata Any cache provides functions to attach user-defined metadata, such as quantification results. 5. Standardized data locations VFS root Data cache Data cache /data/imgs /results Caches are associated to a location in a virtual file system (VFS) defined in C++ code. VFS locations are linked to hard drive locations by the user. Flexible data locations Users can either follow the VFS structure or redirect cache locations based on the current machine's file system. Input JSON schema Parameter file structure Your software Parameter file VFS links Samples Algorithm settings Generates MISA++ applications read all application settings from a standardized parameter file that links filesystem locations to caches in the VFS, defines samples and their parameters, and allows to change algorithm settings. Any MISA++ application can automatically generate a file in JSON schema format that describes all available parameter file properties. Easy parameter definition Parameters and optional documentation are created in C++ code and automatically assigned to a unique location. Automated documentation The JSON schema describes all parameters, as well as the VFS structure, including input and output data. It can be read by humans and also parsed by other software. 6. Standardized graphical user interface Your software MISA++ interface ImageJ/Fiji GUI Via ImageJ plugin We developed a graphical user interface for any MISA++ application as ImageJ plugin. It extracts all necessary information from the automatically generated JSON schema to generate user interfaces for setting up analyses, running them, analyzing results, and creating pipelines. User-defined link Gerst, R., Medyukhina, A., & Figge, M. T. (2020). MISA++: A standardized interface for automated bioimage analysis. SoftwareX, 11, 100405.