Computational Biology

Brief Introduction

Our research focuses on the analysis of cancer progression using evolutionary models, game theory, and machine learning approaches. Each year, cancer is responsible for 13% of all deaths worldwide. Despite significant advances in the fight against cancer, these statistics make clear the need for additional research into new treatments. There has been growing interest in the use of computational models as an effective approach to aid cancer researchers.

Cancer is a very complicated topic and can be studied on many different levels. It involves multiple interactions between molecules, cells, cellular and subcellular events, and their environment. As in immunology, I expect that mathematical, statistical and computational models are essential to complement experimental work in order to obtain a satisfactory understanding of this complex biological system.  We are exploring the idea of viewing tumor (a population of cells in a cooperative state, with a working signaling system in place) as evolutionary signaling systems in the evolutionary game-theoretic, public goods game and multi-agent approaches. 

 

Finished projects:

1. Modeling of Cancer Cell’s Metabolic Behavior in Reverse Warburg Effect
During the reverse Warburg effect, cancer cells induce Warburg effect on their neighboring stromal Fibroblasts which results in secretion of lactic acid. Secreted lactates then could be consumed by cancer cells during respiratory cycle and produce ATP needed for cell growth. As we believe that cells are evolutionary intelligent agents, multiagent approaches seem to be appropriate methods for better understanding such a behavior.
2.A Game Theoretical Model for Switching Dormant to Proliferative Phenotypes in Cancer Tumors

 The role of the immune system in the tumor development covers the concept of cancer immunoediting, which includes three phases: elimination, equilibrium, and escape. The first step in controlling and treatment of a disease is cognition the disease and its behavior. It must be determined precisely the progression of the disease model. Game theory is a strong tool for modeling situations in which the payoffs of different entities with different strategies depend on the action chosen by the others. Evolutionary stable strategy (ESS) is a known topic in game theory which provides a suitable tool for modeling evolution of cancer as well as the invasion of the immune system. We propose and analyze three models of cancer-immune cells interactions by using this class of game theory, ESS, as follows: 1) An evolutionary game model of cancer-immune interaction. 2) An evolutionary game model of proliferating, quiescent and immune cells interactions. 3) An evolutionary game model of proliferating and quiescent cells interactions. In these models, population dynamics and interactions between the immune system and cancer cells have been investigated and equilibrium points are determined. 

3. Modeling Dynamics of HIV-1Quasi-Speices Population

HIV (Human Immunodeficiency Virus) is the virus that can lead to AIDS. HIV is a retrovirus that targets CD4+T cells. The body attempts to replace lost CD4 cells, but over the course of many years, the body is unable to keep the count at a safe level. Destruction of large numbers of CD4 cause symptoms of HIV to appear with increased susceptibility to opportunistic infections, disease, and malignancy.
The recent AIDS treatment mainly depends on the multidrug therapy called cART, which composed of more than 3 different anti-HIV drugs but bears some problems. The model of Harada aimed to establish fundamental theories for an alternative AIDS treatment for cART based on the use of multiple anti-HIV drugs.
We make use of this well-established methodology and intend to study this subject from the perspective of game Theory. We intend to determine the domain of values which cause creating Mix ESS. Our goal is modeling the evolution of population and interaction between exiting phenotypes and thereby determining the domain of effective parameters for payoff values. So this prevents dominating wild type to population and increase latency period.

4. MMP-TIMP Interactions in Cancer Invasion: An Evolutionary Game-Theoretical Framework

One of the main steps in solid cancers to invade surrounding tissues is degradation of tissue barriers in the extracellular matrix. This operation that leads to initiate, angiogenesis and metastasis to other organs, is essentially a consequence of collapsing dynamic balance between matrix metalloproteinases (MMP) and tissue inhibitors of metalloproteinases (TIMP). In this work, we model the MMP-TIMP interaction in both normal tissue and invasive cancer using evolutionary game theory. Our model explains how invasive cancer cells get the upper hand in MMP-TIMP imbalance scenarios.

5. Evolutionary of Impure Multiple Public Goods Games

Public good game investigates the problem of cooperation and defection. In this game, cooperators are players that donate to a public good and pay a given cost. While defectors are players that consume the public good but do not pay any cost. Therefore, defectors dominate the cooperators. In many situations, such as biological evidence, there exist several public goods in which players pay various costs for cooperation and have various preferences for participations. These scenarios require a more general form of public good games, in which players have different preferences for contributing or defecting in several public goods. In this work, we propose multiple impure public goods game (MIPGG) for three public go o ds that can simply be extended to include n public goods.

 

Figure 1: The individuals of a population invest in different public go o ds. Some public goods might be invested by many individuals, while others are less or empty. In this example, there are 3 strategies (indicated by the red circle, blue triangle and green rectangular), M = 3 public goods, and the population size is n = 30. Individuals interact with each other to contribute to a public good. In an evolutionary view, these interactions result in a payoff from evolutionary games. Three strategies (phenotypes in the game) and a number of contributions to three public goods are updated proportionally to their payoffs. Public good containing successful individuals attract more investors.