Note: I don't update this page. Please use this page instead.

I am a theorist and computational scientist working on problems in systems biology.

What is systems biology? The field of systems biology emerged in response to the swelling bounty of high-throughput data about molecular aspects of cell biology. These data intensified the need for understanding how coherent system behavior arises (or is lost) from a staggering diversity of molecular assemblies that interact with one another asynchronously and autonomously. Systems biology, thus, aims at developing and integrating new experimental and mathematical techniques in the pursuit of principles that would make the nature of cellular phenotypes more intelligible and their control more deliberate. This pursuit is driven by the practical need to cure disease. However, it also reflects a want for a theoretical perspective needed to think the complexity of the cell and the organism, and, by extension, evolution itself.

In an earlier phase of my research, I made practical contributions to computational RNA folding. This started the Vienna RNA package, an open-source software tool kit for the computational study of RNA secondary structure. I studied the relation between plasticity and evolvability in the context of RNA folding, which I viewed as a proxy for a genotype-phenotype relation. With my colleagues, I established the concept of a neutral network: a mutationally connected network of sequences that share the same phenotype. The concept found experimental corroboration, and spilled over into other areas of biological research. It can be used to formalize a notion of mutational accessibility between phenotypes, giving rise to a topological concept of neighborhood in phenotype space, sufficient for a sound definition of continuity and discontinuity in evolution. I took this as a basis for reasoning about evolutionary innovation.

Another thread of early work aimed at establishing a mathematical notion of (functional) organization. Biologist Leo Buss and I defined a model in which "molecules" were programs that act on each other (one program using the other as input data) yielding new programs according to a specific formal language known as λ-calculus. A flow-reactor of many such logical molecules evolves towards steady-states that are kinetically self-maintaining algebras, which we took as a definition of organization. This journey connected me to the theoretical foundations of computer science.

It was Étienne Bonnot de Condillac (1714-1780) who observed that "languages are true analytical methods". Indeed, the transition from alchemy to chemistry was facilitated by the emergence of a symbolic language that formalized the compositional character of chemical substances. From today’s perspective, de Condillac’s phrase appears as an apt characterization of computer science some 200 years before it existed. It is completely natural, thus, that computer science joins physics and chemistry as a conceptual and formal foundation of biology.

Reflections of this sort inspired my current computational work , which is focused on developing and applying novel methods for making abstract mechanistic models of processes based on protein-protein interactions as they occur in signaling and assembly. These efforts produced a rule-based modeling language with a graph-rewrite semantics, known as Kappa. Kappa is best viewed as a symbolic language like chemistry but adapted to molecular biology. The techniques we deploy are guided by formal methods in computer science, especially concurrency theory. Our research is motivated by the belief that model-based reasoning must become an integrated aspect of biological informatics and the belief that capturing the combinatorial complexity of signaling systems is necessary for understanding their plasticity and evolvability in phylogenesis, ontogenesis, and disease.

Colleagues in chemistry and computer science independently developed a graph-rewriting approach to organic chemistry: the Mød platform. I'm excited to join forces with these researchers in viewing chemistry from the rule-based perspective.

When joining the Department of Systems Biology at Harvard Medical School in 2004, I developed an interest in experimental research. I chose to study the phenomenon of aging in the model organism C. elegans. Some of our work aimed at longitudinal measurements of molecular physiology, such as redox potential, using fluorescent sensors in single cells of individual organisms throughout their lifespan. Our main approach to aging was based on endpoint statistics, i.e. the change in the statistical properties of lifespan distributions under chemical, physical, and genetic perturbations. We discovered that the mortality statistics of C. elegans responds to an overwhelming number of physical, chemical and genetic interventions simply by a rescaling of time, leaving the shape of lifespan distributions invariant. This observation suggests the existence of an effective, or phenomenological, description of aging: a simple equation describing how a macroscopic state variable characterizing the resilience of an organism changes stochastically with time governed by a single rate constant. With a substantive milestone wrapped up, I decided in 2016 to phase out a 10-year journey into experimental research in order to simplify my life and re-focus on theoretical and computational approaches to biology.

My group brings together scientists with a broad diversity of backgrounds and skills including statistics, biophysics, genetics, computer science and molecular biology in an intellectually engaging environment.