Global geometric affinity for revealing high fidelity protein interaction network

Abstract

Protein-protein interaction (PPI) network analysis presents an essential role in understanding the functional relationship among proteins in a living biological system. Despite the success of current approaches for understanding the PPI network, the large fraction of missing and spurious PPIs and a low coverage of complete PPI network are the sources of major concern. In this paper, based on the diffusion process, we propose a new concept of global geometric affinity and an accompanying computational scheme to filter the uncertain PPIs, namely, reduce the spurious PPIs and recover the missing PPIs in the network. The main concept defines a diffusion process in which all proteins simultaneously participate to define a similarity metric (global geometric affinity (GGA)) to robustly reflect the internal connectivity among proteins. The robustness of the GGA is attributed to propagating the local connectivity to a global representation of similarity among proteins in a diffusion process. The propagation process is extremely fast as only simple matrix products are required in this computation process and thus our method is geared toward applications in high-throughput PPI networks. Furthermore, we proposed two new approaches that determine the optimal geometric scale of the PPI network and the optimal threshold for assigning the PPI from the GGA matrix. Our approach is tested with three protein-protein interaction networks and performs well with significant random noises of deletions and insertions in true PPIs. Our approach has the potential to benefit biological experiments, to better characterize network data sets, and to drive new discoveries.

Figure 1. The illustration of the diffusion process based inference.

(A) displays a spurious PPI (in green color solid line) and a missing PPI (in red color dash line). (B) shows that the spurious PPI (in green color dash link) is removed and the missing PPI (in red color solid line) is recovered in the network.

Figure 2. Geometric Representation of PPI network.

(A) displays a PPI network. denotes the PPI and denotes NPPI. denotes the high dimensional feature vector of the point and denotes the high dimensional feature vector of the point (a blue point and a red point). The denotes the global geometric affinity between the point and point .

Figure 3. Determination of optimal propagation step.

This figure illustrates the process of determination of the optimal step. The horizontal axis denotes the propagation step, and the vertical axis denotes the AUC based probability value. Based on the observation, probability value increases initially and monotonically decreases after obtaining its maximum at step around .

Figure 4. Determination of the threshold for PPI assignment by OLF.

This figure illustrates the procedure of computing optimal threshold for PPI assignment. Figure (A) displays the histogram of global geometric affinity (GGA). Figure (B) displays the value of match function at different threshold. The optimal threshold for GGA corresponds the one with the maxima of match function.

Figure 5. Comparison GGA-method and MDS-method by ROC curve.

This figure plots the ROC curve for the comparison between GGA-method and MDS-method using the CS PPI network data. The vertical axis denotes the sensitivity, and the horizontal axis denotes 1-specificity.

Figure 6. Comparison GGA-method and MDS-method by PR curve.

This Figure plots the PR curve for the comparison between GGA-method and MDS-method using the CS PPI network data. The vertical axis denotes the precision, and the horizontal axis denotes recall.