Benefits of structural genomics for drug discovery research

Nontraditional Roles of Magnesium Ions in Modulating Sav2152: Insight from a Haloacid Dehalogenase-like Superfamily Phosphatase from Staphylococcus aureus

Methicillin-resistant Staphylococcus aureus (MRSA) infection has rapidly spread through various routes. A genomic analysis of clinical MRSA samples revealed an unknown protein, Sav2152, predicted to be a haloacid dehalogenase (HAD)-like hydrolase, making it a potential candidate for a novel drug target. In this study, we determined the crystal structure of Sav2152, which consists of a C2-type cap domain and a core domain. The core domain contains four motifs involved in phosphatase activity that depend on the presence of Mg2+ ions. Specifically, residues D10, D12, and D233, which closely correspond to key residues in structurally homolog proteins, are responsible for binding to the metal ion and are known to play critical roles in phosphatase activity. Our findings indicate that the Mg2+ ion known to stabilize local regions surrounding it, however, paradoxically, destabilizes the local region. Through mutant screening, we identified D10 and D12 as crucial residues for metal binding and maintaining structural stability via various uncharacterized intra-protein interactions, respectively. Substituting D10 with Ala effectively prevents the interaction with Mg2+ ions. The mutation of D12 disrupts important structural associations mediated by D12, leading to a decrease in the stability of Sav2152 and an enhancement in binding affinity to Mg2+ ions. Additionally, our study revealed that D237 can replace D12 and retain phosphatase activity. In summary, our work uncovers the novel role of metal ions in HAD-like phosphatase activity.

Structural genomics and the Protein Data Bank

The field of Structural Genomics arose over the last 3 decades to address a large and rapidly growing divergence between microbial genomic, functional, and structural data. Several international programs took advantage of the vast genomic sequence information and evaluated the feasibility of structure determination for expanded and newly discovered protein families. As a consequence, structural genomics has developed structure-determination pipelines and applied them to a wide range of novel, uncharacterized proteins, often from “microbial dark matter,” and later to proteins from human pathogens. Advances were especially needed in protein production and rapid de novo structure solution. The experimental three-dimensional models were promptly made public, facilitating structure determination of other members of the family and helping to understand their molecular and biochemical functions. Improvements in experimental methods and databases resulted in fast progress in molecular and structural biology. The Protein Data Bank structure repository played a central role in the coordination of structural genomics efforts and the structural biology community as a whole. It facilitated development of standards and validation tools essential for maintaining high quality of deposited structural data.

High precision protein functional site detection using 3D convolutional neural networks

Motivation

Accurate annotation of protein functions is fundamental for understanding molecular and cellular physiology. Data-driven methods hold promise for systematically deriving rules underlying the relationship between protein structure and function. However, the choice of protein structural representation is critical. Pre-defined biochemical features emphasize certain aspects of protein properties while ignoring others, and therefore may fail to capture critical information in complex protein sites.

Results

In this paper, we present a general framework that applies 3D convolutional neural networks (3DCNNs) to structure-based protein functional site detection. The framework can extract task-dependent features automatically from the raw atom distributions. We benchmarked our method against other methods and demonstrate better or comparable performance for site detection. Our deep 3DCNNs achieved an average recall of 0.955 at a precision threshold of 0.99 on PROSITE families, detected 98.89 and 92.88% of nitric oxide synthase and TRYPSIN-like enzyme sites in Catalytic Site Atlas, and showed good performance on challenging cases where sequence motifs are absent but a function is known to exist. Finally, we inspected the individual contributions of each atom to the classification decisions and show that our models successfully recapitulate known 3D features within protein functional sites.

Availability and implementation

The 3DCNN models described in this paper are available at https://simtk.org/projects/fscnn.

Supplementary information

Supplementary data are available at Bioinformatics online.

Molstack: A platform for interactive presentations of electron density and cryo-EM maps and their interpretations

In the Special Issue on Tools for Protein Science in 2018, we presented Molstack: a concept of a cloud-based platform for sharing electron density maps and their interpretations. Molstack is a web platform that allows the interactive visualization of density maps through the simultaneous presentation of multiple datasets and models in a way that allows for easy pairwise comparison. We anticipated that the users of this conceptually simple platform would find many different uses for their projects, and we were not mistaken. We have observed researchers use Molstack to present experimental evidence for their models in the form of electron density maps, omit maps, and anomalous difference density maps. Users also employed Molstack to present alternative interpretations of densities, including rerefinements and speculative interpretations. While we anticipated these types of projects to be the main use cases, we were pleased to see Molstack used to display superpositions of different models, as a tool for story-driven presentations, and for collaboration as well. Here, we present developments in the platform that were driven by user feedback, highlight several cases that used Molstack to enhance the publication, and discuss possible directions for the platform.

Survey of Predictors of Propensity for Protein Production and Crystallization with Application to Predict Resolution of Crystal Structures

Selection of proper targets for the X-ray crystallography will benefit biological research community immensely. Several computational models were proposed to predict propensity of successful protein production and diffraction quality crystallization from protein sequences. We reviewed a comprehensive collection of 22 such predictors that were developed in the last decade. We found that almost all of these models are easily accessible as webservers and/or standalone software and we demonstrated that some of them are widely used by the research community. We empirically evaluated and compared the predictive performance of seven representative methods. The analysis suggests that these methods produce quite accurate propensities for the diffraction-quality crystallization. We also summarized results of the first study of the relation between these predictive propensities and the resolution of the crystallizable proteins. We found that the propensities predicted by several methods are significantly higher for proteins that have high resolution structures compared to those with the low resolution structures. Moreover, we tested a new meta-predictor, MetaXXC, which averages the propensities generated by the three most accurate predictors of the diffraction-quality crystallization. MetaXXC generates putative values of resolution that have modest levels of correlation with the experimental resolutions and it offers the lowest mean absolute error when compared to the seven considered methods. We conclude that protein sequences can be used to fairly accurately predict whether their corresponding protein structures can be solved using X-ray crystallography. Moreover, we also ascertain that sequences can be used to reasonably well predict the resolution of the resulting protein crystals.

3D deep convolutional neural networks for amino acid environment similarity analysis

Background

Central to protein biology is the understanding of how structural elements give rise to observed function. The surfeit of protein structural data enables development of computational methods to systematically derive rules governing structural-functional relationships. However, performance of these methods depends critically on the choice of protein structural representation. Most current methods rely on features that are manually selected based on knowledge about protein structures. These are often general-purpose but not optimized for the specific application of interest.

In this paper, we present a general framework that applies 3D convolutional neural network (3DCNN) technology to structure-based protein analysis. The framework automatically extracts task-specific features from the raw atom distribution, driven by supervised labels. As a pilot study, we use our network to analyze local protein microenvironments surrounding the 20 amino acids, and predict the amino acids most compatible with environments within a protein structure. To further validate the power of our method, we construct two amino acid substitution matrices from the prediction statistics and use them to predict effects of mutations in T4 lysozyme structures.

Results

Our deep 3DCNN achieves a two-fold increase in prediction accuracy compared to models that employ conventional hand-engineered features and successfully recapitulates known information about similar and different microenvironments. Models built from our predictions and substitution matrices achieve an 85% accuracy predicting outcomes of the T4 lysozyme mutation variants. Our substitution matrices contain rich information relevant to mutation analysis compared to well-established substitution matrices. Finally, we present a visualization method to inspect the individual contributions of each atom to the classification decisions.

Conclusions

End-to-end trained deep learning networks consistently outperform methods using hand-engineered features, suggesting that the 3DCNN framework is well suited for analysis of protein microenvironments and may be useful for other protein structural analyses.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-017-1702-0) contains supplementary material, which is available to authorized users.

A public database of macromolecular diffraction experiments

The low reproducibility of published experimental results in many scientific disciplines has recently garnered negative attention in scientific journals and the general media. Public transparency, including the availability of `raw' experimental data, will help to address growing concerns regarding scientific integrity. Macromolecular X-ray crystallography has led the way in requiring the public dissemination of atomic coordinates and a wealth of experimental data, making the field one of the most reproducible in the biological sciences. However, there remains no mandate for public disclosure of the original diffraction data. The Integrated Resource for Reproducibility in Macromolecular Crystallography (IRRMC) has been developed to archive raw data from diffraction experiments and, equally importantly, to provide related metadata. Currently, the database of our resource contains data from 2920 macromolecular diffraction experiments (5767 data sets), accounting for around 3% of all depositions in the Protein Data Bank (PDB), with their corresponding partially curated metadata. IRRMC utilizes distributed storage implemented using a federated architecture of many independent storage servers, which provides both scalability and sustainability. The resource, which is accessible via the web portal at http://www.proteindiffraction.org, can be searched using various criteria. All data are available for unrestricted access and download. The resource serves as a proof of concept and demonstrates the feasibility of archiving raw diffraction data and associated metadata from X-ray crystallographic studies of biological macromolecules. The goal is to expand this resource and include data sets that failed to yield X-ray structures in order to facilitate collaborative efforts that will improve protein structure-determination methods and to ensure the availability of `orphan' data left behind for various reasons by individual investigators and/or extinct structural genomics projects.

Progress in protein crystallography

Macromolecular crystallography evolved enormously from the pioneering days, when structures were solved by "wizards" performing all complicated procedures almost by hand. In the current situation crystal structures of large systems can be often solved very effectively by various powerful automatic programs in days or hours, or even minutes. Such progress is to a large extent coupled to the advances in many other fields, such as genetic engineering, computer technology, availability of synchrotron beam lines and many other techniques, creating the highly interdisciplinary science of macromolecular crystallography. Due to this unprecedented success crystallography is often treated as one of the analytical methods and practiced by researchers interested in structures of macromolecules, but not highly competent in the procedures involved in the process of structure determination. One should therefore take into account that the contemporary, highly automatic systems can produce results almost without human intervention, but the resulting structures must be carefully checked and validated before their release into the public domain.

X-ray crystallography over the past decade for novel drug discovery - where are we heading next?

INTRODUCTION

Macromolecular X-ray crystallography has been the primary methodology for determining the three-dimensional structures of proteins, nucleic acids and viruses. Structural information has paved the way for structure-guided drug discovery and laid the foundations for structural bioinformatics. However, X-ray crystallography still has a few fundamental limitations, some of which may be overcome and complemented using emerging methods and technologies in other areas of structural biology.

AREAS COVERED

This review describes how structural knowledge gained from X-ray crystallography has been used to advance other biophysical methods for structure determination (and vice versa). This article also covers current practices for integrating data generated by other biochemical and biophysical methods with those obtained from X-ray crystallography. Finally, the authors articulate their vision about how a combination of structural and biochemical/biophysical methods may improve our understanding of biological processes and interactions.

EXPERT OPINION

X-ray crystallography has been, and will continue to serve as, the central source of experimental structural biology data used in the discovery of new drugs. However, other structural biology techniques are useful not only to overcome the major limitation of X-ray crystallography, but also to provide complementary structural data that is useful in drug discovery. The use of recent advancements in biochemical, spectroscopy and bioinformatics methods may revolutionize drug discovery, albeit only when these data are combined and analyzed with effective data management systems. Accurate and complete data management is crucial for developing experimental procedures that are robust and reproducible.

The role of protein structural analysis in the next generation sequencing era

Proteins are macromolecules that serve a cell’s myriad processes and functions in all living organisms via dynamic interactions with other proteins, small molecules and cellular components. Genetic variations in the protein-encoding regions of the human genome account for >85% of all known Mendelian diseases, and play an influential role in shaping complex polygenic diseases. Proteins also serve as the predominant target class for the design of small molecule drugs to modulate their activity. Knowledge of the shape and form of proteins, by means of their three-dimensional structures, is therefore instrumental to understanding their roles in disease and their potentials for drug development. In this chapter we outline, with the wide readership of non-structural biologists in mind, the various experimental and computational methods available for protein structure determination. We summarize how the wealth of structure information, contributed to a large extent by the technological advances in structure determination to date, serves as a useful tool to decipher the molecular basis of genetic variations for disease characterization and diagnosis, particularly in the emerging era of genomic medicine, and becomes an integral component in the modern day approach towards rational drug development.

The future of crystallography in drug discovery

INTRODUCTION

X-ray crystallography plays an important role in structure-based drug design (SBDD), and accurate analysis of crystal structures of target macromolecules and macromolecule-ligand complexes is critical at all stages. However, whereas there has been significant progress in improving methods of structural biology, particularly in X-ray crystallography, corresponding progress in the development of computational methods (such as in silico high-throughput screening) is still on the horizon. Crystal structures can be overinterpreted and thus bias hypotheses and follow-up experiments. As in any experimental science, the models of macromolecular structures derived from X-ray diffraction data have their limitations, which need to be critically evaluated and well understood for structure-based drug discovery.

AREAS COVERED

This review describes how the validity, accuracy and precision of a protein or nucleic acid structure determined by X-ray crystallography can be evaluated from three different perspectives: i) the nature of the diffraction experiment; ii) the interpretation of an electron density map; and iii) the interpretation of the structural model in terms of function and mechanism. The strategies to optimally exploit a macromolecular structure are also discussed in the context of 'Big Data' analysis, biochemical experimental design and structure-based drug discovery.

EXPERT OPINION

Although X-ray crystallography is one of the most detailed 'microscopes' available today for examining macromolecular structures, the authors would like to re-emphasize that such structures are only simplified models of the target macromolecules. The authors also wish to reinforce the idea that a structure should not be thought of as a set of precise coordinates but rather as a framework for generating hypotheses to be explored. Numerous biochemical and biophysical experiments, including new diffraction experiments, can and should be performed to verify or falsify these hypotheses. X-ray crystallography will find its future application in drug discovery by the development of specific tools that would allow realistic interpretation of the outcome coordinates and/or support testing of these hypotheses.

The quality and validation of structures from structural genomics

Quality control of three-dimensional structures of macromolecules is a critical step to ensure the integrity of structural biology data, especially those produced by structural genomics centers. Whereas the Protein Data Bank (PDB) has proven to be a remarkable success overall, the inconsistent quality of structures reveals a lack of universal standards for structure/deposit validation. Here, we review the state-of-the-art methods used in macromolecular structure validation, focusing on validation of structures determined by X-ray crystallography. We describe some general protocols used in the rebuilding and re-refinement of problematic structural models. We also briefly discuss some frontier areas of structure validation, including refinement of protein-ligand complexes, automation of structure redetermination, and the use of NMR structures and computational models to solve X-ray crystal structures by molecular replacement.

X-ray crystallography: Assessment and validation of protein-small molecule complexes for drug discovery

INTRODUCTION: Crystallography is the key initial component for structure-based and fragment-based drug design and can often generate leads that can be developed into high potency drugs. Therefore, huge sums of money are committed based on the outcome of crystallography experiments and their interpretation. AREAS COVERED: This review discusses how to evaluate the correctness of an X-ray structure, focusing on the validation of small molecule-protein complexes. Various types of inaccuracies found within the PDB are identified and the ramifications of these errors are discussed. The reader will gain an understanding of the key parameters that need to be inspected before a structure can be used in drug discovery efforts, as well as an appreciation of the difficulties of correctly interpreting electron density for small molecules. The reader will also be introduced to methods for validating small molecules within the context of a macromolecular structure. EXPERT OPINION: One of the reasons that ligand identification and positioning, within a macromolecular crystal structure, is so difficult is that the quality of small molecules widely varies in the PDB. For this reason, the PDB can not always be considered a reliable repository of structural information pertaining to small molecules, and this makes the derivation of general principles that govern small molecule-protein interactions more difficult.

High-throughput structural biology of metabolic enzymes and its impact on human diseases

The Structural Genomics Consortium (SGC) is a public-private partnership that aims to determine the three-dimensional structures of human proteins of medical relevance and place them into the public domain without restriction. To date, the Oxford Metabolic Enzyme Group at SGC has deposited the structures of more than 140 human metabolic enzymes from diverse protein families such as oxidoreductases, hydrolases, oxygenases and fatty acid transferases. A subset of our target proteins are involved in the inherited disorders of carbohydrate, fatty acid, amino acid and vitamin metabolism. This article will provide an overview of the structural data gathered from our high-throughput efforts and the lessons learnt in the structure-function relationship of these enzymes, small molecule development and the molecular basis of disease mutations.

The flow synthesis of heterocycles for natural product and medicinal chemistry applications

This article represents an overview of recent research from the Innovative Technology Centre in the field of flow chemistry which was presented at the FROST2 meeting in Budapest in October 2009. After a short introduction of this rapidly expanding field, we discuss some of our results with a main focus on the synthesis of heterocyclic compounds which we use in various natural product and medicinal chemistry programmes.

Unmet challenges of structural genomics

Structural genomics (SG) programs have developed during the last decade many novel methodologies for faster and more accurate structure determination. These new tools and approaches led to the determination of thousands of protein structures. The generation of enormous amounts of experimental data resulted in significant improvements in the understanding of many biological processes at molecular levels. However, the amount of data collected so far is so large that traditional analysis methods are limiting the rate of extraction of biological and biochemical information from 3D models. This situation has prompted us to review the challenges that remain unmet by SG, as well as the areas in which the potential impact of SG could exceed what has been achieved so far.

Impact of synchrotron radiation on macromolecular crystallography: a personal view

The introduction of synchrotron radiation sources almost four decades ago has led to a revolutionary change in the way that diffraction data from macromolecular crystals are being collected. Here a brief history of the development of methodologies that took advantage of the availability of synchrotron sources are presented, and some personal experiences with the utilization of synchrotrons in the early days are recalled.

To automate or not to automate: this is the question

New protocols and instrumentation significantly boost the outcome of structural biology, which has resulted in significant growth in the number of deposited Protein Data Bank structures. However, even an enormous increase of the productivity of a single step of the structure determination process may not significantly shorten the time between clone and deposition or publication. For example, in a medium size laboratory equipped with the LabDB and HKL-3000 systems, we show that automation of some (and integration of all) steps of the X-ray structure determination pathway is critical for laboratory productivity. Moreover, we show that the lag period after which the impact of a technology change is observed is longer than expected.

X-ray diffraction experiment--the last experiment in the structure elucidation process

Recent years have brought not only an avalanche of new macromolecular structures, but also significant advances in the protein structure determination methodology only now making its way into structure-based drug discovery. In this chapter, we review recent methodology developments in X-ray diffraction experiments that led to fast and very accurate elucidation of three-dimensional structures of macromolecules. We will discuss the role of data collection as the last experiment performed in the crystal structure determination process. A statistical analysis of diffraction experiments that are reported in the Protein Data Bank (PDB) is also presented.