Most readers of this publication will know that “post-genomics” and “proteomics” are phrases that mean little that is specific but herald an encyclopaedic era of information about the way biological cells and their genes and proteins behave. But how best to make sense of it all? It is, at last, possible to anticipate mathematics becoming useful in the modelling of the systems. And in that spirit it would be hard to be more ambitious than the efforts of Leroy Hood's Institute for Systems Biology, in Seattle, and Al Gilman's Alliance for Cellular Signalling, based in Dallas.

Hood aims to synthesize gene expression, protein expression and protein interaction into models of specific cell processes, such as immune responses (see page 828). Gilman's consortium aims to characterize every protein–protein interaction in two cell types, then turn the data into a cellular wiring diagram (see Nature 407, 7; 2000).

The projects differ both in dimension and scope. Hood's approach, for now at least, straddles several levels of data in a smaller system, whereas Gilman's effort sticks to one level of data in a larger system. These differing emphases will no doubt determine the kinds of models that will emerge from each effort. Gilman's “virtual cell” would seem best represented by a series of differential equations, as protein–protein interactions are characterized in a linear fashion. Hood's approach appears to demand different kinds of maths — perhaps forms that capture the probabilities of different cellular elements being simultaneously engaged, or algorithms that account for constantly fluctuating cellular conditions.

But such programmes don't come cheap. Gilman's centre has already received $25 million over five years from the US National Institutes of Health and is seeking an additional $25 million from private sources. Hood is amassing a $200 million endowment for his institute. The centres' different approaches complement each other well. But such work is embryonic, and the two centres could well need to exploit each other's developments as each evolves. They could also need to alter their emphasis as more protein and gene information becomes available in the next few years. But provided they and others like them maintain their flexibility to adapt, they should achieve a long-awaited development when mathematics does what it has so often done in other disciplines: provide a basis for prediction and thereby lead rather than lag behind the experiments.