Prediction of ProteinProtein Interaction Pocket Using L‐Shaped PLS Approach and Its Visualizations by Generative Topographic Mapping

Abstract

Proteinprotein interaction (PPI) pockets in a hostguest protein system were predicted using an L-shaped partial least squares (LPLS) method. LPLS is an extension of standard PLS regression, where, in addition to response vector y and regressor matrix X, an extra data matrix Z is constructed which summarizes the background information on X. The regressor matrix X is a similarity matrix of Tanimoto coefficients of the paired fingerprints of pockets, while the background information Z constitutes eleven physico-chemical and geometrical parameters for describing a pocket. The Boolean response vector y specifies whether each pocket is PPI or non-PPI (indicated by 1 and 0, respectively). Constructing two LPLS models, we successfully predicted the PPI pockets of two protein clusters. Clusters 1 and 2 comprised the X-ray crystal structures of protein-peptide complexes and protein-protein complexes, respectively. From the loading plots derived from each model, we could speculate the geometrical constraints of the PPI pockets. These two models are exclusively unique and it was validated by the cross-prediction simulations. The PPI pockets of cluster 1 were projected onto 2D maps by generative topographic mapping (GTM) and the molecular lipophilic potentials (MLP). Among three examples, the MLP distributions were highly similar because the specimens shared the same p53 guest peptides. Contribution to the Autumn School of Chemoinformatics in Nara, Japan, November 27-28, 2013.

Nonlinear Dimensionality Reduction for Visualizing Toxicity Data: Distance‐Based Versus Topology‐Based Approaches

“…It can be applied for navigation in chemical space, analyzing of structural and property diversity of databases, and coverage of chemical space. [27,33] If visualization methods are applied to classification data, the basic assumption is that active and nonactive molecules form distinct clusters, which are well separated in the descriptor space. The identification of projections of these clusters on the low-dimensional space, which still provides good separation of active and nonactive clusters, can help the understanding and interpretation of the activities of the molecules in the original space.…”

“…Generative Topographic Maps (GTM): [21][22][23][24][25][26][27] GTM is a specific unsupervised density network based on generative modeling. It can be considered as a probabilistic extension of Kohonen Self-Organizing Maps.…”

“…G score: The G score [27] takes into account k nearest neighbors of each projection. The more neighbors of each point that belong to the same class, the higher the G score is.…”

“…Whereas IM relates to geodesic distances (for example, the length of the path that lies on the surface of a low-dimensional manifold), DM implies diffusion ones that reflect probabilities to jump from one point to another according to random walk Markov chains. GTM [21][22][23][24][25][26][27] and Laplacian Eigenmaps (LE) [28] are related to the topology-based techniques. This group of methods is based on the preservation of topology that is concerned with relative proximities: compounds that are close in the data space remain close in visualization.…”

Multivariate Analysis of Side Effects of Drug Molecules Based on Knowledge of Protein Bindings and ProteinProtein Interactions

“…Firstly, L‐shaped PLS (LPLS) was used to construct a multivariate model of the side effects and the protein bindings of the drug molecules. LPLS is an extension of standard PLS regression,9 where, in addition to the response matrix Y and the regressor matrix X , an extra data matrix Z is constructed that summarizes the background information of X 10,11 . X and Y are matrices comprising drugs‐target proteins, and drugs‐side effcts, respectively.…”