Scientists always say you can't understand something until you can create a model of it. This rule is the driving force behind the rise of computational biology, the effort to replicate the inner workings of a biological organism in software. Efforts to do this have been pretty limited and basic — until now.
A new project has created the first complete computational model of an actual organism. And this breakthrough represents a significant step forward in the field of artificial life — and the promise of developing entirely new organisms.
Engineers and architects have been using computer aided design (CAD) for decades, and biologists are finally getting in on the action. Biological processes happen to be very computational in nature — often unfolding in the same way that a software program does. But the biggest challenge is documenting and understanding all the various "algorithms" and processes that a living organism employs.
Converting academic papers into algorithms
Thankfully, a lot of this work has already been done, but no one has really taken this information to the next level — at least not until now. A Stanford research team, led by Markus Covert, took the data from over 900 scientific papers to account for every molecular interaction that takes place in the life cycle of Mycoplasma genitalia, a sexually transmitted disease.
This pathogen happens to be the world's smallest free-living bacteria, making it an excellent candidate for experiments such as these. Because it has only 525 genes (compared to E. coli's 4,288), M. genitalia is serving as an excellent starting block for computational biologists. In fact, the pathogen played an important role in J. Craig Venter Institute's 2008 synthesis of the first artificial chromosome.
But just because it's one of the world's most simplest organisms doesn't mean it still isn't complex. Covert's team had to isolate more than 1,900 experimentally determined parameters. And in order to create the computational equivalents, they developed 28 separate "modules," each governed by its own algorithm. In turn, these modules had to be configured such that they could properly interact with each other — which they did, thus creating an exceptionally accurate virtual model of M. genitalia. And in fact, their model was able to predict the phenotype (what the organism looks like) by using the genotype alone.
The end result was a computer simulation of a pre-existing organism that included all of its molecular components and interactions.
Fascinatingly, the construction of a virtual organism yielded insights not previously known by biologists. For example, the researchers discovered that the length of individual stages in the cell cycle varied from cell to cell, while the length of the overall cycle was much more consistent. By referring to their computer model, the scientists concluded that the overall cell cycle's lack of variation must be the result of a built-in negative feedback mechanism. What's interesting about this is that other biologists and experts on M. genitalia never knew this — but now they can go back to the real thing and test this hypothesis.
Clearly, breakthroughs such at this one are allowing computational biologists to address questions and glean insights that wouldn't otherwise be available. Their work is yet another example showing the tremendous potential for artificial life and the use of CAD in bioengineering and medicine. Through future work, scientists may be able to develop new approaches for the diagnosis and treatment of disease — including the creation of yeast or bacteria designed to mass-produce pharmaceuticals — and to create personalized medicine.
Check out the entire study which recently appeared in the journal Cell.
Top image via Telegraph. Inset image via inntermostsecrets.