Identifying features that effectively represent the energetic contribution of an individual interface residue to the interactions between proteins remains problematic. Here, we present several new features and show that they are more effective than conventional features. By combining the proposed features with conventional features, we develop a predictive model for interaction hot spots. Initially, 54 multifaceted features, composed of different levels of information including structure, sequence and molecular interaction information, are quantified. Then, to identify the best subset of features for predicting hot spots, feature selection is performed using a decision tree. Based on the selected features, a predictive model for hot spots is created using support vector machine (SVM) and tested on an independent test set. Our model shows better overall predictive accuracy than previous methods such as the alanine scanning methods Robetta and FOLDEF, and the knowledge-based method KFC. Subsequent analysis yields several findings about hot spots. As expected, hot spots have a larger relative surface area burial and are more hydrophobic than other residues. Unexpectedly, however, residue conservation displays a rather complicated tendency depending on the types of protein complexes, indicating that this feature is not good for identifying hot spots. Of the selected features, the weighted atomic packing density, relative surface area burial and weighted hydrophobicity are the top 3, with the weighted atomic packing density proving to be the most effective feature for predicting hot spots. Notably, we find that hot spots are closely related to pi-related interactions, especially pi . . . pi interactions.

Repeat proteins have recently been of great interest as potential alternatives to immunoglobulin antibodies due to their unique structural and biophysical features. We here present the development of a binding scaffold based on variable lymphocyte receptors, which are nonimmunoglobulin antibodies composed of Leucine-rich repeat modules in jawless vertebrates, by module engineering. A template scaffold was first constructed by joining consensus repeat modules between the N- and C-capping motifs of variable lymphocyte receptors. The N-terminal domain of the template scaffold was redesigned based on the internalin-B cap by analyzing the modular similarity between the respective repeat units using a computational approach. The newly designed scaffold, termed “Repebody,” showed a high level of soluble expression in bacteria, displaying high thermodynamic and pH stabilities. Ease of molecular engineering was shown by designing repebodies specific for myeloid differentiation protein-2 and hen egg lysozyme, respectively, by a rational approach. The crystal structures of designed repebodies were determined to elucidate the structural features and interaction interfaces. We demonstrate general applicability of the scaffold by selecting repebodies with different binding affinities for interleukin-6 using phage display.

Prediction of protein-ligand interactions is a critical step during the initial phase of drug discovery. We propose a novel deep-learning-based prediction model based on a graph convolutional neural network, named GraphBAR, for protein-ligand binding affinity. Graph convolutional neural networks reduce the computational time and resources that are normally required by the traditional convolutional neural network models. In this technique, the structure of a protein-ligand complex is represented as a graph of multiple adjacency matrices whose entries are affected by distances, and a feature matrix that describes the molecular properties of the atoms. We evaluated the predictive power of GraphBAR for protein-ligand binding affinities by using PDBbind datasets and proved the efficiency of the graph convolution. Given the computational efficiency of graph convolutional neural networks, we also performed data augmentation to improve the model performance. We found that data augmentation with docking simulation data could improve the prediction accuracy although the improvement seems not to be significant. The high prediction performance and speed of GraphBAR suggest that such networks can serve as valuable tools in drug discovery.

Three-dimensional (3D) chromatin structure is an emerging paradigm for understanding gene regulation mechanisms. Hi-C (high-throughput chromatin conformation capture), a method to detect long-range chromatin interactions, allows extensive genome-wide investigation of 3D chromatin structure. However, broad application of Hi-C data have been hindered by the level of complexity in processing Hi-C data and the large size of raw sequencing data. In order to overcome these limitations, we constructed a database named 3DIV (a 3D-genome Interaction Viewer and database) that provides a list of long-range chromatin interaction partners for the queried locus with genomic and epigenomic annotations. 3DIV is the first of its kind to collect all publicly available human Hi-C data to provide 66 billion uniformly processed raw Hi-C read pairs obtained from 80 different human cell/tissue types. In contrast to other databases, 3DIV uniquely provides normalized chromatin interaction frequencies against genomic distance dependent background signals and a dynamic browsing visualization tool for the listed interactions, which could greatly advance the interpretation of chromatin interactions. ‘3DIV’ is available at http://kobic.kr/3div.

Introduction: In contrast to traditional molecular docking, inverse or reverse docking is used for identifying receptors for a given ligand among a large number of receptors. Reverse docking can be used to discover new targets for existing drugs and natural compounds, explain polypharmacology and the molecular mechanism of a substance, find alternative indications of drugs through drug repositioning, and detecting adverse drug reactions and drug toxicity. Areas covered: In this review, the authors examine how reverse docking methods have evolved over the past fifteen years and how they have been used for target identification and related applications for drug discovery. They discuss various aspects of target databases, reverse docking tools and servers. Expert opinion: There are several issues related to reverse docking methods such as target structure dataset construction, computational efficiency, how to include receptor flexibility, and most importantly, how to properly normalize the docking scores. In order for reverse docking to become a truly useful tool for the drug discovery, these issues need to be adequately resolved.

We developed a computational method named Molecule Optimization by Reinforcement Learning and Docking (MORLD) that automatically generates and optimizes lead compounds by combining reinforcement learning and docking to develop predicted novel inhibitors. This model requires only a target protein structure and directly modifies ligand structures to obtain higher predicted binding affinity for the target protein without any other training data. Using MORLD, we were able to generate potential novel inhibitors against discoidin domain receptor 1 kinase (DDR1) in less than 2 days on a moderate computer. We also demonstrated MORLD’s ability to generate predicted novel agonists for the D₄dopamine receptor (D4DR) from scratch without virtual screening on an ultra large compound library. The free web server is available athttp://morld.kaist.ac.kr.