Biologically-reliable computer simulation of the nervous system is the cross-disciplinary area of research which is quite novel and mostly related to neurobiology, neurocybernetics and artificial intelligence. The main idea is to study the mechanisms of brain functioning by means of investigation of its work in details at neuronal and subneuronal level (membranes, neurites, neuromediators, ion channels etc.) and implementation of this data and knowledge within a dynamic model. It’s quite obvious to start from the most simple organism to test the overall conception, and continue with more complex ones in case of success. For this purpose C. elegans nematode is currently the only animal with almost known neural network architecture (connectome). Moreover, its structure, including neurons positions, morphological features and interneuron connections is invariant for all animals of one sex. Relative simplicity but still quite complex behavioral patterns, as well as abilities for learning and memory (Rankin, 2004), provide incomparable perspectives for a wide spectrum of neurobiological studies. Although C. elegans connectome, to a first approximation, was determined about 25 years ago (White et al., 1986), and the first attempt of creation of virtual organism based on C. Elegans was performed in 2005 (Suzuki et al.), still there is no valuable virtual C. elegans.
During last decades C. elegans was extensively studied both with experimental methods and via computer simulations, providing comprehensive data about it. Our research group’s work which started in 2007 is aimed to consolidate these data in a functional form of 3D dynamic model including at least nervous, sensory and muscular systems interacting with each other in simulated physical environment (to enclose feedback loop so that worm’s nervous-system-induced movement causing changes in its local environment can provide updated sensory data back to ‘brain’ ). The prototype of worm body, muscular system and a small fragment of nervous system responsible for locomotion were successfully simulated (Palyanov et. al., 2011). In the beginning of 2011 we joined the international project ‘OpenWorm’ sharing the same goals and objectives. OpenWorm consolidates specialists in a number of related fields, like biophysics, theoretical and experimental neuroscience, numerical methods for simulations, programming etc. for faster and more efficient work. Current progress will be reported.
Certainly in this area of research we should also mention the Blue Brain project which represents an essential first step toward achieving a complete virtual human brain. The researchers have demonstrated the validity of their method by developing a realistic model of a rat cortical column, consisting of about 10,000 neurons. Eventually, of course, their goal is to simulate systems of hundreds of millions of neurons. But currently, when the simulated fragment is a very small part of the whole brain it’s hard to check the validity of its functioning. Our objectives are significantly less ambitious but more feasible and sequential, which is important because even simulation of the worm’s nervous system can, for example, reveal disagreement between current neuroscientific state-of-the-art and reality and, with high probability, only step-by-step movement with approval at each milestone can lead researchers to the success in understanding of fundamental processes underlying the brain functioning.
