MultiCellXML : An open XML data standard for multicell agent models

MultiCellXML Project Home

Update: March 2014

In April 2013, I gave a talk at the National Cancer Institute's Fourth Annual Physical Sciences-Oncology Centers (PS-OCs) Network Investigators' Meeting on the need for greater computational cancer model standardizations, including digital cell lines and related data structures proposed in ongoing development of MultiCellXML. (See this interview with the NCI, immediately following the talk.) Since that time, with funding in part from the Breast Cancer Research Foundation, I have been developing digital cell lines: a representation of a cell line's biophysical and phenotypic properties, along with appropriate microenvironmental context and experimental metadata. My colleagues and I envision a repository of standardized digital cell lines, for use in compatible simulators, and for eventual larger-scale, comparative analyses.

In the coming months, an updated specification of MultiCellXML is anticipated, with the first definition of digital cell lines, digital snapshots (a readout of cells and their phenotypic states in a simulation or experiment at a fixed time, along with environmental conditions), and digital replicates (a bundling of snapshots from a single simulation run or experimental replicate). Early progress was reported at the 2014 Annual Meeting of the American Association for the Advancement of Science (AAAS). (Abstract).

Thanks very much for your interest in MultiCellXML and digital cell lines. I'll look forward to sharing further progress with you as it further develops! Best -- Paul Macklin (March 2014)

Introduction

In recognition of the growing need for open, easily-parsed multicellular data formats to facilitate collaboration, we present and fully document MultiCellXML: an open, XML-based multicell agent model data format. We include several benchmark datasets and open source postprocessing and visualisation code.

This work dates back to 2007, when Paul Macklin et al. began developing an agent-based model of ductal carcinoma in situ--a type of breast cancer whose growth is confined to the duct lumen by a basement membrane. Macklin selected an XML-based data format to simplify code development, with the eventual intention to integrate with other markup-based specifications (e.g., SBML). However, he found those languages to be better suited to single-cell simulations, rather than to large systems of interacting cells in potentially complex geometries.

We therefore put forth MultiCellXML as a potential standard for sharing computational and experimental data, particularly in the context of cancer cell biology. It is our hope that we can: