New Model Predicts Molecular Response of Living Cells

Scientists at the Institute for Systems Biology in Seattle, Wash., in collaboration with researchers from New York University, have developed a model that rapidly characterizes and accurately predicts the molecular-level, mechanistic response of a free-living cell to genetic and environmental changes. The paper describing the model was published Dec. 27 in the online edition of the journal Cell.

The Environmental and Gene Regulatory Influence (EGRIN) model provides information that helps researchers understand how complex biological systems work. Scientists hope that knowledge will open the door to more complex genetic engineering that produces fewer unintended consequences.

"Unraveling complex biological networks is why I came to ISB," said Nitin Baliga, Ph.D, an associate professor at the institute. "The systems approach to biology, of which the founders of ISB were early champions, has proven to be a spectacular success in achieving a molecular level understanding of complex biology, which is necessary if we are to engineer cells back to health or reengineer organisms to improve bioenergy production or bioremediation, for example," Baliga said.

The EGRIN model linked biological processes with previously unknown molecular relationships and accurately predicted both new regulation of known biological processes and the transcriptional responses of more than 1,900 genes to completely novel genetic and environmental experiments.

Baliga and colleagues used Halobacterium salinarum NRC-1, a member of the Archaea family of organisms, because it has been the subject of relatively little scientific study. Archael organisms have evolved to thrive in harsh environments that would be lethal to most other organisms. As a result, their unique biology could provide new solutions to challenges in environmental contamination, energy production and healthcare.

Working with an organism about which relatively little is known allowed the Baliga lab to demonstrate the value of taking a systems approach, which can lead to the rapid discovery of structure and function in unstudied biological networks.

"The ability to gather this level of information regarding a poorly characterized organism from a single study is significant and unprecedented," Baliga said. "In addition, the nature of the EGRIN model is such that it's applicable to many complex biological networks."

The process of discovery involved perturbing cells (for example, altering, individually and in combination, 10 environmental factors and 32 genes), characterizing growth and/or survival phenotype, quantitatively measuring steady state and dynamic changes in mRNA, assimilating the changes into a network model able to repeat the observations and experimentally validating hypotheses formulated through the model. More than 230 out of 413 microarray experiments used were collected and/or conducted specifically for this study. In addition, researchers used data from genome-wide binding location analysis for eight transcription factors, mass spectrometry-based proteomic analysis, protein structure predictions, computational analysis of genome structure and protein evolution as well as data from public sources.

The vast array of approaches to data gathering and validation required a systems biology approach, in which scientists of varied disciplines (for example, biochemistry, physics, mathematics, computation, statistics, genetics and more) collaborate and contribute their skill sets to the achievement of a single scientific objective.

The researchers' next steps involve applying the EGRIN model to more complicated organisms and/or networks, and actually reengineering organisms based on knowledge obtained through the EGRIN model.

"It will take a lot more effort before the EGRIN model can be applied in a practical fashion," Baliga said. "At this point we've basically proven that we can develop a comprehensive understanding of how complex biological systems work, which has been an open question to this point."

Featured Webinar