MultiCellDS: a standard and a community for sharing multicellular data

Abstract

Abstract:Cell biology is increasingly focused on cellular heterogeneity and multicellular systems. To make the fullest use of experimental, clinical, and computational efforts, we need standardized data formats, community-curated "public data libraries", and tools to combine and analyze shared data. To address these needs, our multidisciplinary community created MultiCellDS (MultiCellular Data Standard): an extensible standard, a library of digital cell lines and tissue snapshots, and support software. With the help of experimentalists, clinicians, modelers, and data and library scientists, we can grow this seed into a community-owned ecosystem of shared data and tools, to the benefit of basic science, engineering, and human health. Unmet needs for collecting and curating multicellular dataBiology is increasingly focused on studying cellular heterogeneity and multicellular systems. Novel experiments, clinical trials, and simulation studies are generating incredible amounts of data on cell behavior, cell-cell and cell-matrix interactions, and cellular microenvironmental conditions. These advances are creating exciting new opportunities to formulate and test hypotheses, while synthesizing these disparate data sources to gain a deeper tissue-level understanding of health and disease.However, the deluge of data has pushed existing data sharing and analysis paradigms to their limits. Key insights are effectively hidden in plain sight: tucked away in images, graphs, and tables; divorced from context; and inaccessible to computer analysis without significant manual work. While some data are online, much more are trapped offline on researchers' flash drives, manually traded in emails, or inaccessible in private cloud storage. This severely limits data sharing, collaboration, and post-publication analyses that can offer new and unexpected insights.There have been significant efforts to address these issues, but so far they have focused on describing genomic and molecular data (e.g., the Gene Ontology [1] for genetic data) or mathematical models (e.g., the Systems Biology Markup Language [2] for cell signaling models). None of these efforts have created a fixed data format for interchanging multicellular data or collected cell phenotype insights from many labs into shared, community-curated libraries with a uniform format. And while vast troves of experimental and clinical image . CC-BY 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/090696 doi: bioRxiv preprint first posted online Dec. 9, 2016; data are available online to drive machine learning, we lack a standardized way to record extracted features, such as cell positions, sizes, shapes, and immunohistochemical stain statuses. Moreover, our lack of standardized data prevents us from directly linking between experimental and computational model systems, while also hindering our efforts to reconcile experimental and simulation results against...

Open source tools and standardized data in cancer systems biology

“…More recently, the Cell Behavior Ontology (CBO) [7] emerged to annotate key cellular processes (e.g., division). See [3] for more discussion.…”

“…More recently, MultiCellDS (multicellular data standard) emerged to describe multicellular data, centered around digital cell lines that annotate microenvironment-dependent phenotype, and digital snapshots that record the single-time state of a multicellular system. It is currently working to embed existing standards for molecular-scale processes, with CBO to describe interactions between cells and their phenotypes [3].…”

“…We need to represent, collect and share not just measurements (data models), but also observations and hypotheses using semantic annotations such as CBO and mathematical model languages such as SBML. In [3], we observed the particular need to record cell phenotypic interaction networks and differentiation networks involving multiple cell types. Here, too, we could link data models like MultiCellDS with mathematical model descriptors and semantic annotations.…”

“…1 (top). Moreover, to ensure that scientific insights gleaned from such simulation frameworks are repeatable, reproducible, and extensible, the underlying software must not only be available in the long term, but also as open source, providing the modeling community with the full source code to examine and modify, with unambiguous reuse rights via open source licenses [2,3]. Historically, the trend in computational oncology has been the opposite: singleuse computational models written with ad hoc, non-standardized data, in-house visualization and data analysis scripts, and limited or no availability of source code.…”

Harmonizing semantic annotations for computational models in biology

“…Our perspective on semantic annotation is shaped by our participation in the COmputational Modeling in BIology NEtwork (COMBINE), a community of researchers developing standards for modeling in computational biology 7 . Collectively, we represent teams developing SBML, CellML, the Simulation Experiment Description Markup Language (SED-ML) 8 , the Systems Biology Graphical Notation (SBGN) 9 , the Synthetic Biology Open Language (SBOL) 10 , NeuroML 11,12 , rule-based modeling languages 13,14 , MultiCellDS 15 , the COMBINE archive 16 , the Semantic Simulation (SemSim) architecture 17 , FAIRDOMHub 18 , the SABIO-RK database 19 , as well as BioModels 20 and the Physiome Model Repository 21,22 . Here, we present a set of agreed-upon recommendations for harmonizing semantic annotations for computational models across these development efforts.…”