Berlin 2005 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Downloads | Hilfe
SYBN: Biological and Social Networks
SYBN 3: Biologische und Soziale Netzwerke, Postersitzung
SYBN 3.19: Poster
Montag, 7. März 2005, 14:00–15:30, Poster TU E
Model selection and sampling properties of molecular networks — •Michael Stumpf1, Piers Ingram1, Carsten Wiuf2, and Robert May3 — 1Department of Biological Sciences, Imperial College London, UK — 2Bioinformatics Research Center, University of Aarhus, 8000 Aarhus, Denmark — 3Department of Zoology, University of Oxford, UK
Most studies of molecular or cellular networks have only looked at small subsets of the true network. Applying formal statistical model selection methods to protein interaction and metabolic networks in 5 different organisms we find that simple scale-free network models do not adequately describe current data on such molecular networks. Only if the degree distributions of the network and randomly sampled subnets belong to the same family of probability distribution is it possible to extrapolate from subnet data to properties of the global network. We show that this is indeed the case for some important classes of networks, notably classical random graphs and exponential random graphs. For scale-free degree distributions, however, this is not the case.
We then apply the sampling formalism to several real data sets and discuss how it affects evolutionary inferences. Straightforward, and biologically realistic, extensions of scale-free models vastly improve qualitative and quantitative agreement between real data and theoretical models.