Communicating Complexity and Transforming Interdisciplinary Research
When a loved one is diagnosed with cancer, the first questions that often come to mind are, “why did this happen to them,” and “what can I do.” Faced with this very situation, Dr. Ivan Ivanov, clinical associate professor in bioinformatics at the Texas A&M; College of Veterinary Medicine & Biomedical Sciences (CVM), began a journey helping advance research that one day may be able to answer those questions.
To create new knowledge in this multidisciplinary environment requires the ability to ask questions and design
studies that are based on scientific principles, and communicating that complexity requires a common foundation-a foundation Ivanov is helping to build. However, it was more than an interest in collaboration and complex systems that led Ivanov to his research in cancer biology; it was a series of life events.
The Journey Begins
When Ivanov was a middle school student in the former communist country of Bulgaria, his teacher suggested he apply to a national high school that specialized in mathematics. After speaking with his mother, he decided to apply and was accepted after taking two required entrance exams.
“From there, I went to the university, and I was successful because the teachers I had at the high school were not ordinary teachers,” Ivanov said. “They were Sofia University professors. They ignited one’s curiosity, and that was very exciting.”
While still a mathematics student at the university, Ivanov’s former father-in-law was diagnosed with cancer and he deteriorated quickly. “I was thinking how he could be such a great man, a nonsmoker, a good father and husband, a hard worker-he does everything right, and then he’s hit by this disease,” said Ivanov. “He died after two months. I carried him around in my hands because he lost 50 percent of his weight. I wondered how this could be helped. I was just a mathematician, a mathematics student. I didn’t know anything about biology, and it was very complicated to me.”
Years passed after his former father-in-law’s death and Ivanov left Bulgaria, finished his Ph.D. in mathematics at the University of South Florida, and, after a one-year postdoctoral position at Syracuse University, arrived at Texas A&M; University for a postdoctoral fellowship in the mathematics department.
Once at Texas A&M;, Dr. Edward R. Dougherty, a distinguished professor in the Department of Electrical and Computer Engineering, asked Ivanov if he would be interested in doing some work in cancer biology. Still asking himself, “Why did this happen to my father-in-law, and what can I do about it?” Ivonov began his journey with Dougherty.
Dougherty, who also has a Ph.D. in mathematics, explained some of the problems that cancer investigators were working to solve. He asked Ivanov to write a brief paper involving the Probabilistic Boolean Network modeling of genomic regulation. As a result, Ivanov earned a postdoctoral position in the Training Program in Biostatistics, Bioinformatics, Nutrition, and Cancer led by Dr. Raymond J. Carroll-a distinguished professor in the statistics department at Texas A&M.;
For the next two and a half years, Ivanov began to learn more and more about biology and biological systems. It was during this time he was invited to work on a project with Dr. Robert Chapkin-a distinguished professor in the Department of Nutrition and Food Science. Through that collaborative effort, Ivanov was soon invited to consider a position at the CVM. After giving a talk to the faculty of the interdisciplinary program in toxicology, Ivanov spoke with Dr. Glen Laine, former head of the veterinary physiology and pharmacology department.
“Dr. Laine, who has a background in physics, understood the importance of the mathematics behind my presentation, and he offered me a position,” said Ivanov. “Totally by accident, or by fate, I ended up in the veterinary school, and then things started blossoming. I began collaborating with people from the college and other places. Now, in collaboration with the Fred Hutchinson Cancer Research Center in Seattle, Washington, we are taking part in a human trial about the potential benefit of lignan food supplementation in promoting colon health that may have direct application to cancer prevention. So, by coincidence, all of this is one big circle, supported by the foundation of mathematics.”
From Computers to Cancer
Engineers, according to Ivanov, have long examined mathematical models to help control complex systems like airplanes, cars, and computers. Following in the footsteps of other disciplines such as physics and chemistry, these advances underline the importance of the scientific approach: from experimental design, to developing a predictive mathematical model, validating the model with additional experiments, and ultimately controlling or influencing the system in question.
“It is now time when biology begins to evolve into a mathematically founded discipline,” said Ivanov. “Every time
you investigate areas like molecular biology and cancer, you begin to see complex systems. Gene (dis)regulation, for example, is a current focus in cancer research. In many aspects this process could be modeled after a computer’s architecture and logic. It [gene regulation] is a network. A computer is essentially what is known as a Turing machine, so things that are developed already by mathematicians and engineers-like logical gates and circuits-could be applied directly to biology, especially in cases where complex biological systems are faced with choices, and decisions are made by their regulatory elements.”
From Ivanov’s perspective, without having a systems approach to biology there would be less progress, because scientists would be missing a great part of the picture. Where many would see a wall between the sciences, Ivanov sees opportunity. Mathematical modeling, for him, has gone beyond just a discipline. It is becoming that common foundation, that common language that brings disciplines together. “If we don’t speak a common language, we will never do anything together,” said Ivanov. “Engineers will keep working on circuits and computers, and biologists will just do what Darwin used to do, which is categorizing all the different observed cases and trying to comprehend and explain huge degrees of variation. This is impossible for a human brain to do without a proper foundation.”
From Theory to Application
“The current understanding is that cancer is a molecular disease, which means it’s based on genes and their regulatory interactions within the cell. These genes are not alone, they communicate via different pathways with other genes and external to the cell stimuli,” explained Ivanov. “In this way, they form a communication network. Focusing on an individual gene might not lead to the desired result, because if you hit that gene with a drug, the cancer cell often has the capability to re-route its regulatory activity and still reach a proliferative state that causes the cancer to metastasize.”
“What is needed is to model the entire network or pathway of how the genes are communicating with each other,” said Ivanov. “Using these models, we can predict which genes are ‘turned on’ or highly expressed or ‘turned off’ or down regulated. You have to look beyond just the one gene and target the network if you want to control cancer.”
Ivanov explains that using mathematical models in biology has led to the concept of master-slave gene regulatory networks where one “master” gene controls the activity of a large number of “slave” genes. The thought is that if you can control the masters or some of the intermediate genes, you actually control the entire system. But this can be a tricky and difficult task. “You have to discover them [the master genes], and usually they are well-hidden because they are not usually highly up or down regulated,” said Ivanov. “What is highly up or down regulated are the slave genes, because they are controlled by the masters. You have to figure out which one of the entire section is the master gene and then develop a model-based strategy to control it.”
Ivanov is quick to admit that the potential of this approach is not going to immediately result in a cure for cancer, but it could lead to the ability to control the cancer and stop it from spreading in the body. In addition to his ongoing work in developing mathematical models for cancer biology, Ivanov also continues to collaborate with his colleagues in the food and nutrition sciences department. A recent study that used his mathematical models involved examining the microbiota that naturally live in the digestive tract and how they react with gut epithelial cells. The study found that babies fed a certain type of formula develop genetic signatures similar to babies born prematurely.
“This is an exciting finding and is important because people know if a baby is premature, it has a much higher risk of developing some kind of immunological problem in the future; therefore, babies who are fed that particular type of formula might have that same risk,” said Ivanov. “Babies who are breastfed showed the same kind of gene expression variations we would expect in a normal population.”
The research team determined that such gene expression signatures are strongly related to the composition of the gut microbiome, which suggests that there exists a certain epigenetic “programming” through the interactions of the nutrients, microbiota, and the epithelial cells in the digestive tract.
“These interactions represent a very complex system,” said Ivanov. “We have so many microbes naturally living in our digestive system, and many different cell types. We had to develop a way to model that interaction. That’s very exciting.” Ivanov’s expertise in abstract mathematics and mathematical modeling of complex systems enables him to serve as a bridge among diverse scientific fields. He aids leading edge research by developing theoretical approaches to controlling complex systems, finding applied methods for controlling cancer initiation and progression, and understanding how microbes in the digestive system influence human development from a very early stage of an infant’s life.
Ivanov engages investigators through his work in the Center for Translational Environmental Health Research (CTEHR). The CTEHR, a collaboration among Texas A&M; University, Baylor College of Medicine, and the University of Houston, has a mission to “improve human environmental health by integrating advances in basic, biomedical, and engineering research across translational boundaries from the laboratory to the clinic and to the community and back.” Here, Ivanov directs the CTEHR’s Quantitative Biology Core, which provides investigators with genomic, bioinformatics, and statistical and computational biological support services for their studies. This also includes helping to develop mathematical models.
Mentoring and Modeling
Even though Ivanov has played an integral role in bridging the gap between scientific disciplines, he says his greatest accomplishment is the success of the graduate students he mentors.
“One of my best students, Jason Knight, just successfully defended his doctoral dissertation, and I’m very proud of him,” said Ivanov. “His work was focused on developing modelbased frameworks for classification, finding gene signatures for any kind of condition-not just cancer. In mentoring students like Jason, I can bring in colleagues from many different fields, experts, to create a unique and meaningful graduate experience.”
In addition to his work with colleagues and mentoring graduate students, Ivanov is passionate about his work and the potential it has to impact cancer prevention and treatment, and perhaps other chronic conditions. “I am trying to find ways to reduce the complexity of these mathematical models of gene regulation,” said Ivanov. “We know that we can never model a thirty thousand gene network. It might be possible only with a supercomputer, but even then the computations that predict the dynamic behavior of the system would not be finished until after human beings are long gone. I’m trying to figure out how we can start with a large gene regulatory network model and reduce it to the most important twenty to fifty genes. Of course, we lose information, but maybe the larger regulatory system or a portion of it can still be controlled sufficiently well so that a new drug or other treatment can be developed to control cancer or some other chronic condition. In other words, the goal is to prevent or control a complex disease and keep it from progressing.”
Consequently, Ivanov views his work through the lens of complex systems. When asked the best part of his job, Ivanov replied, “It’s getting my hands on a model and finding a way to simplify it, it’s watching students develop that understanding of the complexity of the world around them, and it’s learning from my colleagues. It’s all of the above. Those things, in a way, form a network that induces my curiosity. I learn from my students and I learn from my colleagues because they have different perspectives. Then I go into my own world and I rethink all of those interactions and what I learn from them. What I take from them shows me how I should proceed. It’s great.”
