In year 2012 Dr Shinya Yamanaka received the Nobel Prize in Physiology or Medicine for his groundbreaking work of reprogramming differentiated cells into stem cells that can be further induced to transform into other cell types. This is an example of cell phenotypic transition (CPT). CPT is ubiquitous in biology. Mammalian cells that share the same genome exist in notably distinct phenotypes, characterized by different morphology, gene expression patterns, and epigenetic chromatin statuses. The human cell atlas project aims to characterize all human cell types. With all this increasing knowledge, understanding how a CPT progress is regulated and how it proceeds will provide guidance for interfering and manipulating the transition process, either suppressing undesirable transitions such as in cancer metastasis and fibrosis, or directing transitions to the desired direction such as specific differentiation routes of stem cells in regenerative medicine. Consequently, quantitative study on CPTs emerges as one of the most exciting frontiers for basic science and biomedical research.
Two grand technical challenges, however, impede further development of the field. Fixed cell-based approaches can provide snapshots of high-dimensional expression profiles but have fundamental limits on revealing temporal information, and fluorescence-based live cell imaging approaches provide temporal information but are technically challenging for multiplex long- term imaging. The Xing lab currently focuses on developing and applying integrated experimental and computational approaches to tackle these challenges, and study CPT dynamics.
1) Gene regulatory networks (GRN), consisting of interlocking feedback loops, govern cellular dynamics and cell fate decisions. A paramount task in systems biology is to learn the governing equations defined by the network structure. Decades of research on GRN have demonstrated that this is an unfathomable challenge. In collaboration with the Ivet Bahar lab and the Jonathan Weissman lab at MIT, we are developing an experimental/computational procedure that can map out the governing vector field of a CPT process, and applying it to study cell differentiation. We built a comprehensive framework, dynamo, and with it we show that genome-wide vector field function can be accurately and efficiently reconstructed with the improved velocity. Our method of single cell vector field reconstruction thus contributes as a significant step towards the holy grail of learning the governing equations of any cellular dynamic processes. We are applying techniques from multiple disciplines such as molecular physics, nonlinear dynamics, and differential geometry to analyze the topological characters of the obtained multi-dimensional vector fields and extract information on the dynamics of corresponding cellular processes.
2) Reconstructing long-time dynamics of a CPT process requires live cell imaging. We first developed a live-cell imaging platform that tracks cellular status change through combining endogenous fluorescent labeling that minimizes perturbation to cell physiology, and/or live cell imaging in a high-dimensional cell feature space (Sci Adv 2020), facilitated by deep-learning based computational image analyses (Comp Biol Medicine 2019). With our platform, we recorded live cell trajectories reveal parallel paths of epithelial-to-mesenchymal transition (EMT) missing from snapshot data due to cell-cell dynamic heterogeneity. Recognizing that CPTs are examples of rate processes, we introduced transition path analyses and the concept of reaction coordinate from the well-established rate theories into CPT studies, and applied on this EMT process. The study emphasizes the importance of live cell imaging in reconstructing cellular dynamics. Currently we are integrating the live cell studies in the cell feature space and the snapshot studies in the expression space.
3) During CPTs cells undergo global change of transcriptome, epigenome and chromosome structures. By analyzing different types of high-throughput data we aim at unraveling some basic principles of cell phenotype regulation (PLoS Comp Biol., 2019). Currently we are using CRISPR-based techniques to tag and track selected genomic loci in live cells, and genomic structure and cell fate in cancer and developmental models .
4) After acute kidney injury (AKI) a kidney repair system is activated. By combining mathematical modeling and mouse model studies, we unraveled some basic design principles of the repair system, and resolved puzzling roles of Wnt and EMT on the recovery process and progression to chronic kidney diseases (CKD) (iScience 2020). Now we are in the process of integrating the above-mentioned tools we developed into studying kidney injury and repair.
A brief introduction of our current projects can be found here, and my BIOKDD2020 talk. In my first lecture of our Cellular and Systems Modeling class I always give a philosophical presentation on why we need physics in biology.
Past Projects
Dynamics of protein motors
Assembly dynamics of microtubules
Mechanism of allosteric regulation