Principles of human tissues 

Organs are made of cells that work together and communicate with each other in order to achieve joint functions. In order to make sense of their complexity requires principles that can guide our understanding of tissue biology. We ask questions about the general design principles of organs and tissues:

How do organs maintain a proper size despite the fact that their cells constantly divide and die?
How do organs maintain proper ratios of their different cell types?
How do organs adjust their function to match variation in distant tissues with which they communicate? How do tissues resist takeover by mutants that missense feedback signals?

Remarkably, there exist principles for tissue-level circuits that can address these challenges. These principles unify our understanding of very different tissues. The resulting feedback circuits are essential for organ function but have specific fragilities that lead to disease. Understanding the origins of diseases in this way can offer fresh perspectives on prevention and therapy.

A feedback loop in which glucose controls pancreatic beta cell proliferation and death allows organ size control and compensation for insulin resistance:
Dynamical compensation in physiological circuits

A biphasic mechanism in which glucose kills beta cells at high concentration is essential to resist mutant takeover but has a fragility that leads to type-2 diabetes.

Biphasic response as a mechanism against mutant takeover in tissue homeostasis circuits

Principles of cell circuits for tissue homeostasis:

Continuum of Gene-Expression Profiles Provides Spatial Division of Labor within a Differentiated Cell Type

Circuit Design Features of a Stable Two-Cell System

Hormone circuits

Hormone circuits in textbooks and math models are expected to work on the timescale of minutes to hours - the lifetime of the hormones, and on the timescale of a day due to circadian rhythm. We added to this picture interactions that provide a timescale of weeks-months. These are well characterized changes in the functional mass of the hormone glands (eg growth of the thyroid or adrenal gland). This growth is due to the growth-factor effects of the hormones in each pathway. Although characterized, these effects have not been considered on the system level. 

Our gland-mass models of the human stress pathway (HPA - accompanied by longitudinal measurements of hair cortisol), thyroid axis, beta cells and ovaries explain phenomena on the timescale of weeks-months. These phenomena include addiction, depression, bipolar disorder mood episodes, prediabetes, subclinical thyroid diseases, ovarian dynamics and PCOS, and hormone seasonality. The growth and shrinkage of glands also provides systems level functions such as dynamic compensation (strict homeostasis of key variables and strict robustness of their dynamic response curves to a given stimulus) in the face of variation in physiological parameters such as insulin resistance, blood volume and metabolic state of the cells.

A New Model For the Hpa Axis Explains Dysregulation of Stress Hormones on the Timescale of Weeks

Hormone seasonality in medical records suggests circannual endocrine circuits

An opponent process for alcohol addiction based on changes in endocrine gland mass

Dynamics of Thyroid Diseases and Thyroid-Axis Gland Masses

Immune circuits
We study immune circuits mathematically. This includes a theory for endocrine autoimmune disease based on the hypothesis that T cells weed out hypersecreting secretory cells. We also address the question of autoimmune flares by modeelign them as an exciatbel system. We are currently working on the systems understanding of cancer immunotherapy and the pathogen response of the immune system.
Simplifying inflammation and Fibrosi​​​​​​s​
Fibrosis -excess scarring- is a condition that cuts across medicine, causes many diseases, and has no cure. It is a complex process with many cell types and molecules. To understand fibrosis, we developed a simple mathematical model of the relevant cell populations (macrophges, myofibroblasts). The model provides concepts that explain the basic features of fibrosis, and predict several types of fibrosis such as hot fibrosis (with macrophages) and cold fibrosis (without them). The model also suggests drug targets to reduce fibrosis, such as inhibiting the myofibroblast autocrine loop. These concepts led to experiments with these drug candidates that reduce fibrosis in heart attacks and liver cirrhosis in mice.