X hits on this document

117 views

0 shares

0 downloads

0 comments

2 / 43

FEATURE

Linked: A Conversation with Albert-László Barabási

***

S. Bahar

The dynamical properties of networks have recently gained much attention, in guises ranging from small-world networks to the United States power grid. One of the most recent major contributions to the study of network dynamics comes from the research group of Albert-László Barabási. In a widely- cited series of papers in major journals such as Science and Nature, Barabási and his colleagues in the Physics Department at Notre Dame have added a powerful new dimension to our understanding of network phenomena – a dimension which is highly relevant to many of us working in biological physics.

The study of random networks began, as Barabasi explains in his recent popular book Linked: How Everything is Connected to Everything Else and What It Means for Business, Science, and Everyday Life (Plume, 2003), which recounts the history of the study of networks, as well as the story of his group’s contributions, with the work of Erdős and Rényi the late 1950s and early 1960s. Their work described randomly connected graphs, with N vertices each making P random connections. As Barabási explains, Erdős and Rényi “acknowledged for the first time that real graphs, from social networks to phone lines, are not nice and regular. They are hopelessly complicated. Humbled by their complexity, [Erdős and Rényi] assumed that these networks are random.” (Linked, p. 19)

The introduction of random network theory was a major advance, but it unwittingly provided a stumbling block as well. Researchers took it as a given that complexity was the same thing as randomness. This, it turns out, is very far from the case.

Imagine the network formed by airline routing maps in the United States: certain airports (Chicago’s O’Hare, LAX, JFK, etc.) act as hubs, while others are visited by only a few routes. The existence of hubs is also present in other networks, such as regulatory networks in the cell (consider the ubiquitous ATP, for example, or the regulatory protein p53, whose disturbance can cause a wide range of cancers). It was Barabási’s research group which first identified the predominance of hubs in many networks, and the theory they developed to explain the prevalence of hubs pushed network theory forward in a startling new way.

After growing up in Transylvania, in the Hungarian speaking region of Romania,

Albert-László undergraduate

Barabási studied inBucharest, and

as later

an in

Budapest with Tamás Vicsek, the well-known expert in fractal growth phenomena. Barabási then came to the United States to complete a doctorate in Physics at Boston University with Gene Stanley, thus receiving a strong background in materials science and statistical

2

Document info
Document views117
Page views117
Page last viewedFri Dec 02 20:28:31 UTC 2016
Pages43
Paragraphs915
Words20254

Comments