A year after its publication in Genome Biology [BioInform 04-13-01], BioD, a visual language developed by Daniel Cook, is gaining traction as an approach for researchers looking for a way to graphically represent biological systems and the relationships of their components.
Earlier this year, Cook, a researcher affiliated with the University of Seattle, received a $150,000 grant from the NIH to further develop the semantic underpinnings of BioD. But he has also been working on a user-friendly Java-based editor called Chalkboard that will allow users to build their own models. He hopes this editor will be ready — with a limited vocabulary — within the next few months for beta testing.
BioD uses a visual vocabulary, consisting of objects that represent biological components and links that describe the actions between the objects. Underneath these connections, functional properties, both quantitative and qualitative, can be added over time, such as reaction rates. “You can start with a purely graphical description,” said Cook, a researcher affiliated with the University of Seattle. “But then as you get to know more and more about the system you are studying, you can start upgrading to quantitative analysis in the same environment.” Furthermore, complex systems can be divided into sub-models that are linked to each other, he said.
One group that has had early access to BioD is the pediatric epilepsy research center (PERC) at the University of Washington. Cook created a customized visual model describing the mechanisms of ketogenic diet as an anti-epileptic treatment. “It was fantastically helpful in terms of not only identifying the variables and the different levels of action for [them], but …to use this information to make hypotheses about what the mechanisms of action of this treatment might be,” said Philip Schwartzkroin, a professor at UC Davis and the former research director of PERC, which has since been dissolved. He was also pleased, he said, with the ability of BioD to integrate information from different biological levels, for example, molecular and cellular interactions.
Cook is planning similar projects with other researchers, including a group at the department of medicinal chemistry at the University of Washington to build models of drug detoxification mechanisms.
Once the BioD editor has been tried and tested, Cook said, he wants to make it more widely available, possibly as freeware. He also plans to build an interface that would make BioD compatible with the systems biology markup language (SBML) developed at Caltech. “That’s the sort of integration that I think needs to happen if we are going to solve these big-systems problems,” he said.