Selection of representative structures from large biomolecular ensembles

Abstract

Despite the incredible progress of experimental techniques, protein structure determination still remains a challenging task. Due to the rapid improvements of computer technology, simulations are often used to complement or interpret experimental data, in particular for sparse or low-resolution data. Many such in silico methods allow to obtain highly accurate models of a protein structure either de novo or via refinement of a physical model with experimental restraints. One crucial question is how to select a representative member or ensemble out of vast number of computationally generated structures. Here, we introduce such a method. As a representative task, we add co-evolutionary contact pairs as distance restraints to a physical force field and want to select a good characterization of the resulting native-like ensemble. To generate large ensembles, we run replica-exchange molecular dynamics (REMD) on five mid-sized test proteins and over a wide temperature range. High temperatures allow overcoming energetic barriers while low temperatures perform local searches of native-like conformations. The integrated bias is based on co-evolutionary contact pairs derived from a deep residual neural network to guide the simulation towards native-like conformations. We shortly compare and discuss the achieved model precision of contact-guided REMD for mid-sized proteins. Lastly, we discuss four robust ensemble-selection algorithms in great detail which are capable to extract the representative structure models with a high certainty. To assess the performance of the selection algorithms we exemplarily mimic a "blind scenario', i.e. where the target structure is unknown, and select a representative structural ensemble of native-like folds.

Clustering Heterogeneous Conformational Ensembles of Intrinsically Disordered Proteins with t-Distributed Stochastic Neighbor Embedding

“…Here, we show that t-SNE is particularly well-suited for clustering seemingly disparate IDP conformations into homogeneous subgroups since it is designed to conserve the local neighborhood when reducing the dimension, which ensures similar data points remain equivalently similar and dissimilar data points remain equivalently dissimilar in the low-dimensional and high-dimensional space . Due to its ability to provide a very informative visualization of heterogeneity in the data, t-SNE is being increasingly employed in several applications such as clustering data from single cell transcriptomics, − mass spectrometry imaging, and mass cytometry. , Lately, t-SNE has also been used for depicting the MD trajectories of folded proteins − and for interpretation of mass-spectrometry-based experimental data on IDPs by juxtaposing with classical GROMOS-based conformation clusters from the corresponding molecular simulation trajectories of the IDP under consideration …”