An experimentally derived confidence score for binary protein-protein interactions

Abstract

Information on protein-protein interactions is of central importance for many areas of biomedical research. Currently no method exists to systematically and experimentally assess the quality of individual interactions reported in interaction mapping experiments. To provide a standardized confidence-scoring method that can be applied to tens of thousands of protein interactions we have developed an interaction tool-kit consisting of four complementary high-throughput (HT) protein interaction assays. These assays were benchmarked against positive and random reference sets (PRS and RRS) consisting of well documented human interaction pairs and randomly chosen protein pairs, respectively. A logistic regression model was trained using the PRS/RRS data to combine the assay outputs and calculate the probability that any novel interaction pair is a true biophysical interaction once it has been tested in the tool-kit. This general approach will allow a systematic and empirical assignment of confidence scores to all individual protein-protein interactions in interactome networks.

Mapping transcription factor interactome networks using HaloTag protein arrays

“…A critical question for any new technology regards the quality of the obtained data. To address this, we followed an approach we previously described in several publications (16,23,25,26) in which a subset of interactions from a new dataset is systematically validated in a second interaction assay, in our case a pull-down assay. The challenge of this approach is that no assay can detect all interactions and each assay has a different interactiondetection profile (25,27).…”

“…To address this, we followed an approach we previously described in several publications (16,23,25,26) in which a subset of interactions from a new dataset is systematically validated in a second interaction assay, in our case a pull-down assay. The challenge of this approach is that no assay can detect all interactions and each assay has a different interactiondetection profile (25,27). To estimate the false-discovery rate of a new dataset, it is therefore important to measure the sensitivity and background of the validation assay by benchmarking this against sets of control pairs: a positive reference set (PRS) of well-documented interactions and a random reference set (RRS) composed of random protein pairs (27).…”

“…Subsequently, the retest rate of the new dataset in the validation assay can be interpreted in light of these benchmarking data. To measure the sensitivity and background of our pull-down assay, we benchmarked it against (i) a set of 49 well-documented interactions from the literature (PRS; AtPRS_v1s) and (ii) a set of 69 random protein pairs for which there was no evidence of interaction in the literature (RRS) (25). The 49 AtPRS_v1s pairs are a random subset of our original set of 118 PRSs (AtPRS_v1) (16).…”

Mapping, modeling, and characterization of protein–protein interactions on a proteomic scale

“…This screening method has therefore proven to be a useful tool, enumerating binary interactions not only for human, but for a number of model organisms as well, including Saccharomyces cerevisiae [9][10][11], Schizosaccharomyces pombe [12], Escherichia coli [13], Caenorhabditis elegans [14,15], and Arabidopsis thaliana [16]. While this method is easily scaled and relatively inexpensive, it may fail to capture interactions between proteins which rely on intermediary or scaffold proteins (such as those between protein complex subunits), those involving proteins from specific subcellular compartments (such as membrane proteins), or those which require post-translational modifications [17]. Moreover, this assay requires proteins to be expressed at nonendogenous levels in the yeast nucleus.…”

“…However, this can be put in perspective when considering the many different technical parameters that influence the detection of PPIs. The genes queried, the strains or cell types used, and the presence and orientation of proteins tags are all examples of the many variables that impact the detection of PPIs [7,17]. Ultimately, the combination of maps generated with different methods provides a more complete view of interactome networks, since each method highlights a different subset of the interactome.…”