OpenWorm is an international open science project to simulate the roundworm Caenorhabditis elegans at the cellular level in silico.  Although the long term goal is to model all 959 cells of the C. elegans, the first stage is to model the worm's locomotion by simulating the 302 neurons and 95 muscle cells. This bottom up simulation is being pursued by the OpenWorm community. So far the physics engine sibernetic has been built and models of the neural connectome and a muscle cell have been created in the NeuroML format. A 3D model of the worm anatomy can be accessed through the web via the OpenWorm browser. The OpenWorm project is also contributing to develop the geppetto simulation framework, a web-based multi-algorithm, multi-scale simulation platform engineered to support the simulation of the whole organism.
Background: C. elegans
The roundworm C. elegans has one of the simplest brains of any organism, with only 302 neurons. Furthermore, the structural connectome of these neurons is fully worked out. There are fewer than one thousand cells in the whole body of a C. elegans worm, each with a unique identifier and comprehensive supporting literature because C. elegans is a model organism. Being a model organism, the genome is fully known, along with many well characterized mutants readily available, a comprehensive literature of behavioural studies, etc. With so few neurons and new calcium 2 photon microscopy techniques it should soon be possible to record the complete neural activity of a living organism. By manipulating the neurons through optogenetic techniques, combined with the above recording capacities the project is in an unprecedented position to be able to fully characterize the neural dynamics of an entire organism.
In trying to build an "in silico" model of a relatively simple organism like C. elegans, new tools are being developed which will make it easier to model more complex organisms.
Project Nemaload  is a current research program which is trying to empirically establish the relevant biological facts which are necessary for a true bottom up simulation. The project founder, David Dalrymple, is a collaborator on the OpenWorm project.
Although the ultimate goal is to simulate all features of C. elegans behaviour, the project is new and the first behaviour the Open Worm community decided to simulate is a simple motor response: teaching the worm to crawl. To do so, the virtual worm must be placed in a virtual environment. A full feedback loop must be established from: Environmental Stimulus > Sensory Transduction > Interneuron Firing > Motor Neuron Firing > Motor Output > Environmental Change > Sensory Transduction ...
There are two main technical challenges here: modelling the neural/electrical properties of the brain as it processes the information and then modelling the mechanical properties of the body as it moves. The neural properties are being modeled by the Hodgkin Huxley equations, and the mechanical properties are being modeled by a Smoothed Particle Hydrodynamic algorithm.
The OpenWorm team built an engine called Geppetto which could integrate these algorithms and due to its modularity will be able to model other biological systems (like digestion) which the team will tackle at a later time.
The team also built an environment called NeuroConstruct which is able to output neural structures in NeuroML. Using NeuroConstruct the team reconstructed the full connectome of C. elegans.
Using NeuroML the team has also built a model of a muscle cell. Note that these models currently only model the relevant properties for the simple motor response: the neural/electrical and the mechanical properties discussed above.
The next step is to connect this muscle cell to the six neurons which synapse on it and approximate their effect.
The rough plan is to then both:
- Approximate the synapses which synapse on those neurons
- Repeat the process for other muscle cells
In 2005 a Texas researcher described a simplified C. elegans simulator based on a 1-wire network incorporating a digital Parallax Basic Stamp processor, sensory inputs and motor outputs. Inputs employed 16-bit A/D converters attached to operational amplifier simulated neurons and a 1-wire temperature sensor. Motor outputs were controlled by 256-position digital potentiometers and 8-bit digital ports. Artificial muscle action was based on Nitinol actuators. It used a "sense-process-react" operating loop which recreated several instinctual behaviors.
These early attempts of simulation have been criticized for not being biologically realistic. Although we have the complete structural connectome, we do not know the synaptic weights at each of the known synapses. We do not even know whether the synapses are inhibitory or excitatory. To compensate for this the Hiroshima group used machine learning to find the weights of the synapses which would generate the desired behaviour. It is therefore no surprise that the model displayed the behaviour, and it may not represent true understanding of the system.
The Open Worm community is committed to the ideals of open science. Generally this means that the team will try to publish in open access journals and include all data gathered (to avoid the file drawer problem). Indeed all the biological data the team has gathered is publicly available, and the five publications the group has made so far are available for free on their website. All the software that OpenWorm has produced is completely free and open source.
Open Worm is also trying a radically open model of scientific collaboration. The team consists of anyone who wishes to be a part of it. There are over one hundred "members" who are signed up for the high volume technical mailing list. Of the most active members who are named on a publication there are collaborators from Russia, Brazil, England, Scotland, Ireland and the United States. To coordinate this international effort, the team uses "virtual lab meetings" and other online tools that are detailed in the resources section.
- nematode fanciers open their worm to a kickstarter
- Towards a virtual C. elegans: A framework for simulation and visualization of the neuromuscular system in a 3D physical environment. Palyanov A., Khayrulin S., Larson S.D. and Dibert A. (2012) In Silico Biology 11(3): 137-147
- Current practice in software development for computational neuroscience and how to improve it. Gewaltig, M.-O., & Cannon, R. (2014). PLoS Computational Biology, 10(1), e1003376. doi:10.1371/journal.pcbi.1003376
- Openworm is going to be a digital organism in your browser
- The Perfect C. ELEGANS Project: An Initial Report
- Whole brain emulation and nematodes
- A dynamic body model of the Nematode C. elegans with neural oscillators
- A model of motor control of the nematode C. elegans with neuronal circuits
- P. Frenger, “Simple C. elegans Nervous System Emulator”, Houston Conf Biomed Engr Research, 2005, pg.192.