Simple models are often used in physics to study the universe as well as the inner spaces of the atom. However, living world seems to be different and more challenging. This made Stephens and his colleagues wonder whether it was possible to build a simple model with behaviour that is virtually indistinguishable from a real living system. Furthermore, if they would such a model, could they then use it to disentangle the dynamics that may lead to more complexity?
Predicting worm behaviour
The researchers used the roundworm C. elegans, a commonly studied animal in biology, as an example. This worm has thousands of genes, many different cell types, and a deeply connected network of neurons for processing sensory information and guiding adaptive behavior. Stephens: “If we could capture the effect of all of that biological detail in an interpretable computational model, then we could use the model to deduce precisely which biological details are connected to which aspects of behavior.”
Modeling the behavior or movement of any animal at high precision is incredibly hard. The researchers, however, realized that movements are not random from moment to moment, but instead occur in sequences that endow context to each movement. When the researchers paid close attention to recent movements, it became easier and simpler to build a more accurate model. By including movements close to each other in time, they could tap into a well-studied class of computational models known technically as Markov models.
Health sector
According to Stephens, the ideas he and his colleagues developed aren’t just restricted to worms, but are generalizable to other animals, including humans. This could be incredibly helpful, for example, in understanding diseases with locomotor symptoms like Parkinson’s. The researchers are currently in discussions with companies in the health sector who are interested in their techniques.
Stephens: “Effectively, our approach provides a new “computational microscope” with which to understand complex, living movement and how it emerges from a complex interplay between internal biological mechanisms and continuous interactions with the environment.” The study was published in Proceedings of the National Academy of Sciences (PNAS).