Physicist Albert-László Barabási’s network theory merges with a fundamental need of researchers and clinicians.
Since the mapping of the human genome, the amount and structure of the data we’re getting means we have to think differently about biological systems and disease pathologies.
Enter the human diseasome. This is all the diseases of an individual or group, viewed as a whole, with special focus on genetic features.
Like its cousins genome, proteome and metabolome, the diseasome (disease + ome) is a totality, a whole field of study and a new approach. And like 17th-century explorers circumnavigating the globe, we have only a partial map, which we revise as we sail.
Barabási stands among the first NIH-funded scientists to discover a new diagnostic model using scale-free networks, a model that explains their emergence in systems from the cell, to the cell phone to the diseasome.
His recent talk in Masur Auditorium—“Network Medicine: From Cellular Interactions to Human Diseases”—drew a large audience.
Barabási, who directs the Center for Complex Network Research at Northeastern University, holds appointments in physics, biology and computer/information sciences. He began with a simple analogy.
“A broken car with a smoking engine and dysfunctional lights has many similarities to human disease,” he said. “But there is one huge difference. It’s virtually guaranteed that if you take the car to the mechanic, he or she will be able to fix it. And that’s not something that we can say about many of our diseases. The mechanic has the spare parts…and that is about where we are in medicine. What the genome project really provided us with is the parts…the genes, the proteins, the metabolites.”
These parts in the human cell are so interdependent that disease is rarely caused by a single abnormal gene. Instead, disease reflects a disturbance in complex networks within the cells. Network medicine is a holistic approach for investigating these networks.
“The other thing the mechanic has,” Barabási continued, “that the medical doctor does not have, is the wiring diagram, the blueprint of the car. How are the different components wired together? This is what the secret of medicine has to be: to understand this wiring diagram.”
Yet Barabási and his colleagues were surprised to find that, in fact, each of these networks was not random, but rather a scale-free system of nodes—connecting points—and links.
These look like a map of U.S. airport connections, or a power grid where the nodes are generators, transformers and substations, while the links are transmission lines.
What makes such a network scale-free? Some nodes seem to have a limitless number of connections, as in the World Wide Web, or cellular metabolic networks.
“The popular nodes, called hubs, can have hundreds, thousands or even millions of links,” he said.
Such networks have other important properties as well. For example, they are robust against accidental failures—their numerous interconnections seem to compensate for them. Yet they are vulnerable to coordinated attacks.
“For the larger scale-free networks, you can remove 95 percent of the nodes and it will be okay,” said Barabási. “But if you attack big nodes, it breaks down.” If Chicago’s O’Hare International Airport, a major hub, went offline, the effect on air travel would be huge.
Furthermore, “rapid advances in network biology show that cellular networks are governed by universal laws,” Barabási said.
The laws of scale-free networks appear to apply equally to cells, computers and the World Wide Web, as well as human groups connected by email or by collaborations in science, art and business.
So Barabási’s “wiring diagram” would show the common genetic origin of many diseases. It would also reveal the interplay between the cell’s network organization and certain heritable diseases.
Identifying hub molecules involved in a given disease could lead to biomarkers and new drugs to target the hubs. Advances are also essential for finding new disease genes and understanding disease-associated mutations.
“Many diseases share genes,” Barabási said. “Is this meaningful? Yes. Connected diseases show significant comorbidity [co-occurrence]. All the data that we have access to indicate that disease genes are clustered in well-defined neighborhoods of the network.”
His kinetic slides of whirling networks inspire the same kick you feel at finding your own home with geo-mapping software. As you zoom to your block, then zoom out to planet Earth, you are jumping scale and can intuitively grasp how complex systems have structures that are repeated.
What Barabási does is offer convincing proof of that underlying architecture.
“It’s data-dependent. The diseasome map now is like a skeleton,” he said. “We could make it complete and it could have just as much power as the genome project has provided.”
Barabási explores both the molecular complexity within a given disease and among different phenotypes. These are individuals or groups with physical characteristics that are genetically and environmentally influenced, as in cases of asthma.
“Diseases correspond to a breakdown of a region of the network of the disease module,” said Barabási, “and we will not be able to map out those disease modules until we have a good understanding of the network as a whole.”
This is important because, for more than 40 years, scientists believed that complex networks were completely random.
Article originally appeared in the NIH Record.