Cell-Cell variability
Through the development of a wide range of single-cell characterization techniques, it has become clear that mammalian cells of the same phenotype that are grown under the same conditions, often exhibit significant cell-to-cell variability. Cellular variability is often connected to understanding cancer recurrence and stem cell differentiation. We are currently working on the development of novel micropatterning technology that will allow us to observe clonal colonies that derive from a single cell while allowing long-term repetitive observation of each individual cell. If successful, our technologies enable measurements that will give us a unique insight into the cell’s circuit architecture. From our experiments, we anticipate discovering how tightly cellular rate constants are regulated and to determine the relaxation times that describe how fast fluctuations around the homeostatic value decay back to equilibrium.
Cells on patterned surface
Cancer-on-a-chip
In recent years, the role of metabolism in cancer has become an area of intense interest. In the 1920s, Otto Warburg suggested that cancer is a metabolic disease caused by dysfunctional mitochondria after he found that tumors consume much more glucose than other tissues and doing so by using the less efficient metabolic pathway of glycolysis to produce ATP. Even though his original interpretation has been disproved, the Warburg effect is still actively and intensely studied because it is linked to the emergence of invasive and metastatic cancer. Most, but not all, invasive cancers exhibit a phenotype that metabolizes glucose at a higher rate than non-invasive cells. At the same time, most invasive cell lines exhibit an increased utilization of the glycolysis even if exposed to oxygen (aerobic glycolysis or Warburg effect). This is in contrast to “normal” cells that only utilize glycolysis when exposed to hypoxia (oxygen deficiency), which is also called the Pasteur effect. The Warburg effect has led to several hypotheses to explain how aerobic glycolysis is advantageous to the fitness and survival of cancer cells. We are currently working to create tumor-on-a-chip technology to grow well-defined 3D tumor microenvironments that can be exposed to defined nutrient stresses such as differing oxygen and nutrient gradients. In addition, we will be able to determine the metabolic phenotype of cancer cells with differing oncogenic changes and degrees of aggressiveness. Our technology should be able to distinguish between competing and synergistic hypotheses to address the question of the role of the TME on metabolic reprogramming and in particular glycolysis.
Bayesian Analysis of time-series and Network discovery
Over the last few years, my group has developed several data analysis algorithms for time-series data. In particular, we have developed a comprehensive theory for the Ornstein-Uhlenbeck process, which describes a stochastic process that has a stationary solution (mean) and fluctuates around a mean through a Wiener process (random walk). Our maximum-likelihood solution results in the probability distributions of the parameter of the model (amplitude, mean, and relaxation time) from a given experimental time-series (PRE, 2019,110, 062142). We are currently working on applying these techniques to brain functional magnetic resonance measurements to discover and characterize functional brain circuits.
