Flow of information in biological systems
In the biological world, information is rarely broadcasted uniformly, nor is it limited to pairwise interactions. Rather, often information percolates and is propelled through networks that change dynamically, partly in response to the information they themselves convey. In our research group, we therefore develop computational tools and mathematical models to understand how information moves through living systems and how the movement of information controls their behavior.
All of our work focuses on the interplay between information, communication, and structure in living systems and three questions that we find particularly interesting are: How do incentives to lie and deceive change the flow of information through living systems? Faced with a barrage of data, how should we comprehend the ﬂow of information through biological and social systems? Can the information generated by natural selection over generations be quantified by the apparatus of information theory?
Networks. The whole is greater than the sum of its parts — it is the sum of its parts plus the local interactions. These local interactions induce a system-wide flow of information that is fundamental for maintaining the overall cooperation between distant parts of a complex system. In biological and ecological systems, information is seldom globally accessible but instead often channeled through specific links between senders and receivers. This makes network modeling to a powerful tool to represent and to understand the constrained communication of the world.
Communication in biological and ecological systems offers plentiful opportunity to deceive, manipulate, and defraud. Consequently, to understand information flow in general, we must consider that information is constantly being manipulated. In our research, we extend dynamic network models beyond transmission of honest information to better understand how strategic communication and the pattern of interactions co-evolve.
Maps. Networks make it possible to characterize the complex systems of our world, but even as abstractions they remain highly complex. It is therefore often helpful to decompose the myriad nodes and links into connected modules that represent the network. Good representations both simplify and highlight the underlying structures and the relationships that they depict; they are maps.
The best maps convey a great deal of information but require minimal bandwidth: the best maps are also good compressions. This is an optimal coding problem and can be resolved by the minimum description length (MDL) principle. In our research, we take advantage of this principle and develop powerful mapping tools to reveal important structure in large networks.
Evolution. Over the past three and a half billion years, living organisms have evolved to acquire, process, store, and transmit information. Yet, for all the attention that is directed toward the information revolution in human society, remarkably little is known about the broad role of information in biological systems. To date, information is little more than a metaphor in evolutionary biology. How can we take information beyond the level of metaphor?
By exploring evolution in an information-theoretic framework, we highlight the way natural selection generates correlation between the sequence properties of the world and the sequence properties of life. In our research, we formalize ideas about the role of information in evolutionary biology to take information in biology beyond the level of metaphor.