Systems Biology – Frank Doyle

The success of the human genome project has ushered in a new era that emphasizes a systemic or integrated approach to ascertain the cellular behavior arising from complex cellular networks. Scientists are now embarking on a quest to elucidate the organization and control of cellular networks that underlie the phenotypic behavior of a cell; these are the so-called “omics” such as genomics, transcriptomics, proteomics, metabolomics. Fueled by recent advances in molecular biology providing high-throughput and in-depth data of gene and protein interactions, it is increasingly clear that cell behaviors arise from complex interactions among the genes and proteins through crossover and cascade regulations and signal transductions, and thus can be explained only through a system-level understanding of these interactions. This is the goal of systems biology, which involves application of systems theoretic approaches and integration of experimental and computational research. Two recent projects in the lab have focused on post-traumatic stress disorder (PTSD) and circadian rhythms.

Post-Traumatic Stress Disorder

PTSD is an anxiety disorder that occurs among people exposed to a traumatic event involving life threat and injury. Understanding the disorder from a systems level perspective requires probing the problem from different scales. We have built statistical and ordinary differential equation models to gain insights into the mechanisms of the disorder.

A central challenge in high-throughput data analysis is the identification of informative molecular biomarkers from among tens of thousands of candidates. We have developed robust gene expression analysis tools which we have applied to human PTSD (and mouse social defeat ) transcriptomic data. In particular, we have developed COMBINER, a computational framework which efficiently and robustly identifies disease core modules by integrating data collected from different cohorts and across different tissue samples.

Neuroendocrine studies identified the hypothalamus-pituitary-adrenal (HPA) axis as the site of action that brings about biochemical changes in response to severe stress. In particular, the stress responsive HPA axis regulates cortisol (an informative endocrine biomarker that can distinguish PTSD from other co-morbid disorders such as major depression) through feedforward and feedback loops. One of the molecular networks that is initiated during stress involves the heat-shock protein 90 (HSP90) that plays a central role in signaling various downstream regulators including the transport of cortisol to the nucleus.

Our group has constructed a mathematical model for the cortisol dynamics in the HPA axis that supports the hypothesis that high stress intensity and strong negative feedback loop may cause hypersensitive neuro-endocrine axis that results in hypocortisolemia in PTSD. Moreover, we modeled the effect of acute and chronic stress on the dynamics of the HSP90 network and extended this to the human model to make predictions about the occurrence of depression, PTSD and the comorbidity.

Circadian Rhythms

Circadian rhythms are near 24-hour endogenous oscillations in physiological processes found in many organisms, coordinated through transcription-translation networks with inherent time-delayed negative feedback. These networks serve as an excellent example of a functional genetic circuit, able to process subtle environmental cues while remaining robust to thermal and evolutionary disturbances. To better understand how these characteristics are obtained, mathematical models have been constructed to demonstrate how features such as limit cycle oscillations, robustness, and temperature compensation can be achieved in genetic regulatory networks.

Key to understanding how these genetic circuits operate is connecting how inputs (i.e. light, energy levels, activity) are processed to give the appropriate outputs (phase shift, amplitude change). Thus, an accurate model must capture not only the correct constant-dark time-dependent dynamics, but also the correct input-output response. For circadian rhythms, knockdown experiments using RNA interference technology (siRNA) and small molecule modulators has resulted in a wealth of data on the dynamic responses to changes in key rates. Additionally, high-throughput microarrays have provided high-resolution time-series data of gene expression levels. This data, together with qualitative knowledge of the underlying network structure, enables mathematical models to reveal new insights into the design principles of the circadian clock.

In our group, we have used mathematical models of circadian rhythms to both interpret complicated feedback circuits and guide further experimental inquiry. Specifically, we are interested in connections between the clock and metabolism as well as the potential development of circadian-related pharmaceuticals.