|
I've been learning about the gene regulatory network (GRN) and protein-protein interaction network (PPI) recently. I've found a huge amount of extremely interesting papers about how biological network data is accumulated and how it can be studied. However, what I haven't found is any examples where the network data was used to make a novel prediction which was later experimentally shown to actually exist. What are some of the cases where this has happened? |
|||
|
|
|
There is very interesting work in this regard done by Ed Marcotte. An overview of the "machine" and its conservation across species is here. The point here is the machine or network is conserved, but used in different ways in different organisms. Hence, the phenotypes of mutants might be different, but they'd be consistent across many variants in the components of the machine. One paper of his describes a tool, which uses well developed guilt-by-association principles upon a human gene network to identify associations of gene sets. This is available at www.functionalnet.org/RIDDLE. Look at their approach to prioritizing candidate disease genes by network-based boosting of genome-wide association data. Some specific examples of the conserved "machine" are given in this elegant PNAS paper. It is nice how they tested the genes in Xenopus embryos. |
|||
|
|
