In Table 1, we compiled interaction and constraint types.
This was (still is) a progressive stage that follows other post-genome sequencing initiatives such as the HapMap consortium project describing human genome diversity [ 8] - further supplemented by the 1,000 genomes Project [ 9] - or the Human Microbiome Project [ 10] that compiles microbe interactions in either healthy or diseased humans.
For the signaling network, we compiled signaling interactions from KEGG, BioCarta (http://pid.nci.nih.gov/) and TRANSPATH [ 36], as well as through manual curation of some undirected protein-protein interactions.
Taken together, optimal use of all the databases compiling the interactions obtained by different methods will reduce the time and expense of finding a specific PDZ-binding partner for further studies at a genome-wide level, and will also aid its functional characterization [ 24, 73, 111].
Examining this topic further, we validated the most distinctly deregulated over- and under expressed miRNAs by RT-qPCR, utilized traverse analysis of altered miRNAs between the sample groups, characterized the reorganized miRNAs by compiling their interactions in ingenuity pathway analysis, expression in multiple tumor systems, and their functional role in metastasis and tumor progression.
The STITCH database compiles chemical-protein interactions curated by text-mining of the literature and provides confidence scores that reflect the level of significance and certainty of interaction between small molecules and targets.
In generating the human interactome network, a previously compiled data set was used, mainly containing experimentally demonstrated interactions compiled through a literature-curation process, combined with data from different types of experimental and computational evidence [ 12].
Using previously identified genetic and physical interactions compiled at SGD, we determined the interaction network for the 9 genes (BEM1, CTF4, CTK1, HFI1, MFT1, NUP133, THO2, THP1 and THP2) which demonstrated diploid-specific sensitivity to DOX.
In the first phase, edges – compiled from mammalian interactions in BioGRID [27] and HPRD [28] – were pruned from the network if they did not resemble likely interactions (as defined by a logistic regression model), with the goal of reducing false positives among the reported interactions.
Using GO annotations, a curated list of fission yeast complexes can be compiled (GO complex interactions).
The network contains 80,922 interactions compiled from a total of 21 different human PPI databases (see table 1 of Bossi and Lehner 2009 for details).
