Using cellular fluctuations and computational analyses to probe biological mechanisms

点击翻译按钮获取中文摘要

Project Summary New experimental approaches in single‐cell imaging and sequencing are producing an unprecedented amount of data to quantify the intricate dynamics of biomedical processes. These processes are subject to the intertwined issues of complexity and randomness, and it can be difficult for medical professionals to interpret, understand or act on this data. In particular, spatial, temporal and stochastic fluctuations in cellular processes introduce huge uncertainties that compromise responses, complicate modeling, and make predictive understanding seemingly impossible. We hypothesize that the fluctuations and heterogeneities of single‐cell dynamics can contain powerful information resources that can be unlocked with improved computational methods and integrated experiment designs. This project will create these tools and use them to study the dynamics of Mitogen–Activate Protein Kinase signaling and downstream regulation for multiple genes in multiple organisms. We will integrate state‐of‐the‐art single‐cell‐single‐molecule super‐resolution microscopy experiments with novel discrete stochastic analysis methods and seek to unlock the mysteries of (1) How do MAPK signals and transcription factors interact in space and time to differentially control expression of multiple genes in response to different external stresses and (2) How do mRNA sequences, protein regulators, and ribosomes interact to affect the natural and aberrant dynamics of translation activation, initiation, elongation and termination? We will also create a set of advanced computational tools and build them into a user‐friendly software package (the Stochastic System Identification Toolkit, SSIT), which will enable the systematic integration of discrete stochastic modeling approaches with single‐cell experiment techniques. We will build the SSIT to accomplish crucial tasks in the design, interpretation, prediction, and control of single‐cell experiments. To guarantee the broadest possible impact, the SSIT will be validated in direct collaboration with at least four of the nation’s top single‐cell experimental groups in bacteria, yeast, insect, and human research. Once validated, all SSIT tools will be made publically available, and the theory, algorithms and techniques will be taught to scores of graduate students, postdocs, and other young biomedical researchers at Colorado State University, Vanderbilt University, UC Berkeley, and Los Alamos National laboratory as well as at the NIGMS‐funded q‐bio Summer School, an internationally recognized program organized by the PI and held annually at the CSU. Our long‐term goal is to make systematic and rigorous computational modeling an accessible and standard practice for biological and biomedical research laboratories around the world. Successful completion of our goal will broadly support NIH mission areas to seek predictive knowledge about the nature and behavior of living systems; to enable more rapid and cost effective discoveries in health‐related fields; and to develop the human, physical and computational resources necessary to enhance the nation's economic well‐being and ability to prevent disease.