Quantitative virtual infection modeling in sepsis

CONTACT [email protected] Research Group Systems Biology / Bioinformatics Acknowledgments: This work was supported by the Deutsche Forschungsgemeinschaft (DFG) in the Collaborative Research Centre/Transregio 124 FungiNet (subprojects B3, INF, C3) as well as German Ministry for Education and Science in the program Unternehmen Region (BMBF 03Z2JN21). Headline 1 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 Acknowledgments: This work was supported by the Deutsche Forschungsgemeinschaft (DFG) in the Collaborative Research Centre/Transregio 124 FungiNet (subprojects B4). References: [1] Hünniger K, Lehnert T, Bieber K, Martin R, Figge MT, et al. (2014), PLoS Comput. Biol. [2] Lehnert T, Timme S, Pollmächer J, Hünniger K, Kurzai O and Figge MT (2015), Front. Microbiol. [3] Medyukhina, A., Timme, S., Mokhtari, Z. and Figge, M. T. (2015), Cytometry C. albicans associated to monocytes extracellular C. albicans cells C. albicans associated to neutrophils Association assays [1] C. albicans cells Laboratory Experiment Computer Simulations of derived Models Quantitative Characterization of Processes Aquisition and Analysis of Data Systems Biology Laboratory experiment Quantitative mathematical modeling Aquisition and analysis of data Systems Biology Predictions based on computer simulations Laboratory experiments Aquisition and analysis of data Predictions and new hypothesis Quantitative characterization of processes agent-based model M0,0 M1,0 M2,0 M1,1 M0,1 M0,2 G0,0 G0,1 G0,2 G1,0 G1,1 G2,0 CAE CAR CKE CKR ... ... ... ... ... ... ... ... CAE free, alive fungal cell CKE free, killed fungal cell CAR alive and evading fungal cell CKR killed and evading fungal cell Gi,j neutrophil/PMN with i alive fungal cells j killed fungal cells Mi,j monocyte with i alive fungal cells, j killed fungal cells Intracellular Killing Immune Evasion Gi,j Gi-1,j+1 κG Mi,j Mi-1,j+1 κM Gi,j CAE Gi+1,j φG CKE Gi,j Gi,j+1 φG Mi,j CAE Mi+1,j φM Mi,j CKE Mi,j+1 φM CAE CKE κext(t) CAR CAE ρ CKR CKE ρ Phagocytosis by Immune Cells Extracellular Killing state-based model ordinary differential equations (ODE) partial differential equations (PDE) agent-based model (ABM) state-based model (SBM) space resolution resolution of individual objects Zeit t=0min humanes Blut C. albicans Zeit t=10, 30, ... 240 min monocytes neutrophils Whole blood infection assay [1] FACS analysis: association assays Experimental time-series data: Spatio-temporal experimental data: Automated image analysis (segmentation, classification, tracking) Bottom-up approach Strategy for parameter estimation with different mathematical models [2] Using the mathematical model, new in silico experiments can be done Models with increasing level of complexity build on one another The output on one level is used for the calibration of the models at a higher level Virtual infection model State-based model: (no spatial resolution) [1,2] C. albicans and immune cells exist in different states The state of one cell is updated according to the transitions of the model via a stochastic process Transition rates are determined by the global optimization method simulated annealing Agent-based model: (spatial resolution) [2] Adding spatial information (morphology and migration) Transition rates were adopted from the SBM model due to high computational costs of ABM Estimation of diffusion coefficients of immune cells Quantification and extrapolation of the immune response in dysregulated immune systems with decreasing number of immune cells [2], e.g. neutropenia Produce useful predictions or extrapolations that match experimental results or suggest new experiments Permit data to be generated that is beyond present-day experimental capabilities Situation-dependent mathematical modeling Virtual patient Yield non-intuitive insights into how a system or process works Identify missing processes or components in a system Understand and visualize complex processes Components of the model are described by parameters which define the dynamics and morphology of the components Choice of appropriate modeling approach depends on the underlying data and the hypothesis to be tested With increasing resolution of space and individual objects the computational costs are increasing Parameters are estimated by fitting experimental data to the model Quantitative virtual infection modeling of sepsis Alexander Tille1,2, Teresa Lehnert1, Sandra Timme1,2, Maria Prauße1,2 and Marc Thilo Figge1,2 1 Applied Systems Biology, Leibniz Institute for Natural Product Research and Infection Biology – Hans-Knöll-Institute, Jena, Germany 2 Friedrich-Schiller-University Jena, Germany