Project Details
Projekt
Quantitative Data Analysis and Computational Modelling
Applicants
Dr. Daniel Baum; Professor Dr. Max von Kleist
Subject Area
Experimental and Theoretical Network Neuroscience
Developmental Neurobiology
Developmental Neurobiology
Term
since 2022
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 453877723
The goal of RobustCircuit is to understand common principles underlying the requirements of imprecise processes at lower scales (from molecules to cells) to yield robust outcomes at higher scales (from cells to behavior) in neural circuit assembly. Live imaging is the principle means to obtain quantitative data on imprecise and robust processes at the levels of molecular, subcellular and cellular dynamic processes. Seven of the eight projects utilize intravital and ex vivo live imaging to obtain large amounts of raw visual data on (sub-)cellular dynamics during brain development. Foremost, this raw data is computationally processed, yielding statistically powerful numerical data enabling to detect and quantify imprecisions, including noise, in the developmental process. Based on such quantitative data, computational modelling allows to make predictions regarding the roles that imprecision plays in creating robust developmental programmes and functional neural wiring. Based on the experiences and successful application of this pipeline in the first funding period, the Z1 project is designed to continue to tackle two core challenges for the consortium effort: Raw data analysis processing and analysis should be automated as far as possible to reduce subjectivity in the outcomes and to facilitate fast processing of large amounts of data. Moreover, during the first funding period, connectomics analyses have become a key new approach to quantitatively assess variability in the outcome. Objective 1 is devised to continue to provide automated image processing that turns images into quantitatively comparable numerical data, enabling powerful statistical analysis and downstream modelling of developmental processes. In addition, connectomics has become an additional focus of some projects for the second funding period. Z1 will devise new methods of deep connectome analysis to support respective RobustCircuit projects. Statistically pinpointing if, where, and how ‘noise’ turns into ‘robustness’ is an inherent challenge to all projects, that requires both robust statistical methods and mechanistic modelling. In addition, probing robustness by manipulating imprecise parameters without collateral effects is challenging in biological experiments. Computational modelling offers a means to specifically conduct these types of experiments and provide testable hypotheses for the generation of robust outcomes. Objective 2 is devised to continue to develop solid statistical tests that enable detecting ‘where’ robustness emerges. Moreover, stochastic modelling approaches, which were successfully developed during the first RobustCircuit funding period, will be employed to explain how robustness in neural circuit assembly emerges from intrinsically stochastic subcellular dynamics across RobustCircuit projects.
DFG Programme
Research Units
