When David Harel started the experiment, the petri dish of mouse cells looked just like any other. Genes were being expressed, proteins were being made, and the tissue was being perfused with oxygen-rich blood.

But then things started to change.First one cell changed position and moved across the plate, followed quickly by another. Eventually, through migration and other changes in cell functionality and signaling, the cells had differentiated, with the lucky ones becoming fully-fledged thymus gland T cells. And it all happened in a fraction of the time that biologists would have expected based on several decades of physiological and development studies; after all, this experiment was happening inside a computer, in virtual organs modeled by complicated diagrams, simulating their real-world counterparts.

Harel, a Professor of Computer Science at Israels Weizmann Institute, sees his teams work at the leading edge of a dramatic shift in scientific thinking. Biological research is ready for an extremely significant transition, he writes, from analysis (reducing experimental observation to elementary building blocks) to synthesis (integrating the parts into a comprehensive whole).

Harel presented his views on the changing landscape of biological research during a talk at the Falling Walls Conference in Berlin, where curious attendees filled a re-purposed water pumping station to hear about coming scientific breakthroughs. My approach is to get a holistic view on an entire system, he tells me after his talk. I think the excitement comes not from breaking it down to the last detail, but from the fact that you can build a model to understand something and then you get emergent properties; those complex, not always predictable, responses that come from millions of tiny interactions.

One of the most prominent trends in biology during the era of modern science has been the reductionist pursuit of ever-smaller biological components. In an attempt to understand the flora and fauna around us, we looked closer and found cells; cells led to DNA and genes and proteins and metabolites a dizzying constellation of interacting small molecules that, together, make life possible. Harels approach suggests a complementary path: the computational re-construction of biological systems (or reverse-engineered biology as he describes it) to run experiments that would be impractical or unwieldy in the lab. Its a way, in essence, to test the untestable.

A model is only as good as the data you put into it, and Harels models of worms or organs, and his recent project on modeling a cancerous tumor, depend upon thousands of previous studies of gene behavior of enzyme kinetics. If you put everything thats known into the model in a consistent manner, he says, you can carry out intricate runs of the model under a variety of circumstances, and begin to make inferences. That makes our models realistic, interactive, and modifiable as new data becomes available.

But what if the unpredictable properties that emerge run into practical, un-modeled limits of the real world, like nutrient limitation or faulty gene circuits? To Harel, such a disconnect is actually an opportunity, an attitude that reflects his thinly veiled penchant for pressing biologists buttons. You have to put things in your model that make sense, he warns, but not always. If you make the biologists angry enough, they will go and do an experiment to prove you wrong, and thats what makes for exciting new discoveries.

By way of an example, Harel describes how once, when no one knew the reason for a stage of worm development, he added an unreasonable artifact something with no basis in reality into his model in order to make it act like the real thing. My crazy insertion caused the model to behave properly, but was clearly wrong, he recalls with a mischievous grin. But it caused the biologists to get to work and discover the real answer.

By playing provocateur, Harel believes he can push our knowledge of living systems forward. Why model things? Harel asks philosophically, overlooking the Spree River. Because Id like to truly understand life, to uncover gaps, correct errors, and form theories. The magnitude of the things you can do with such models of biology is mind-boggling. The skys the limit.

Follow this link:
Why the Frontiers of Biology Might Be Inside a Computer Chip