The Impact of Data on Structure-Based Binding Affinity Predictions Using Deep Neural Networks.

Artificial intelligence (AI) has gained significant traction in the field of drug discovery, with deep learning (DL) algorithms playing a crucial role in predicting protein-ligand binding affinities. Despite advancements in neural network architectures, system representation, and training techniques, the performance of DL affinity prediction has reached a plateau, prompting the question of whether it is truly solved or if the current performance is overly optimistic and reliant on biased, easily predictable data. Like other DL-related problems, this issue seems to stem from the training and test sets used when building the models. In this work, we investigate the impact of several parameters related to the input data on the performance of neural network affinity prediction models. Notably, we identify the size of the binding pocket as a critical factor influencing the performance of our statistical models; furthermore, it is more important to train a model with as much data as possible than to restrict the training to only high-quality datasets. Finally, we also confirm the bias in the typically used current test sets. Therefore, several types of evaluation and benchmarking are required to understand models' decision-making processes and accurately compare the performance of models.

Untargeted Metabolomics for Disease-Specific Signatures.

Human diseases account for complex traits that usually exhibit markedly diverse clinical manifestations coming from a series of pathogenic processes that shape heterogeneous phenotypes. Considering that correlation does not imply causation as well as population differences and/or inter-individual variability, disease-specific signatures are becoming critical for biomarker discovery. Untargeted metabolomics is deemed to be a powerful approach to delineate molecular pathways of prime interest. Metabotypes capture the interplay of genomics and environmental influences per se. Untargeted metabolomics share the charm of being not only hypothesis-driven but also hypothesis-generating. Notwithstanding, the applicability of untargeted metabolomics toward clinically relevant outcomes depend on wet- and dry-lab procedures in the context of elegant study designs with clear rationale. As ideal may be far from feasible, herein we provide recommendations to combat sample mishandling that adversely affect data outcomes and if so, deal with imbalanced datasets toward data integrity.

Integrative structural and functional analysis of human malic enzyme 3: A potential therapeutic target for pancreatic cancer.

Malic enzymes (ME1, ME2, and ME3) are involved in cellular energy regulation, redox homeostasis, and biosynthetic processes, through the production of pyruvate and reducing agent NAD(P)H. Recent studies have implicated the third and least well-characterized isoform, mitochondrial NADP+-dependent malic enzyme 3 (ME3), as a therapeutic target for pancreatic cancers. Here, we utilized an integrated structure approach to determine the structures of ME3 in various ligand-binding states at near-atomic resolutions. ME3 is captured in the open form existing as a stable tetramer and its dynamic Domain C is critical for activity. Catalytic assay results reveal that ME3 is a non-allosteric enzyme and does not require modulators for activity while structural analysis suggests that the inner stability of ME3 Domain A relative to ME2 disables allostery in ME3. With structural information available for all three malic enzymes, the foundation has been laid to understand the structural and biochemical differences of these enzymes and could aid in the development of specific malic enzyme small molecule drugs.

Correction to 'TMSNP: a web server to predict pathogenesis of missense mutations in the transmembrane region of membrane proteins'.

[This corrects the article DOI: 10.1093/nar/lqab008.].

Three-Dimensional Cell Metabolomics Deciphers the Anti-Angiogenic Properties of the Radioprotectant Amifostine.

Aberrant angiogenesis is a hallmark for cancer and inflammation, a key notion in drug repurposing efforts. To delineate the anti-angiogenic properties of amifostine in a human adult angiogenesis model via 3D cell metabolomics and upon a stimulant-specific manner, a 3D cellular angiogenesis assay that recapitulates cell physiology and drug action was coupled to untargeted metabolomics by liquid chromatography-mass spectrometry and nuclear magnetic resonance spectroscopy. The early events of angiogenesis upon its most prominent stimulants (vascular endothelial growth factor-A or deferoxamine) were addressed by cell sprouting measurements. Data analyses consisted of a series of supervised and unsupervised methods as well as univariate and multivariate approaches to shed light on mechanism-specific inhibitory profiles. The 3D untargeted cell metabolomes were found to grasp the early events of angiogenesis. Evident of an initial and sharp response, the metabolites identified primarily span amino acids, sphingolipids, and nucleotides. Profiles were pathway or stimulant specific. The amifostine inhibition profile was rather similar to that of sunitinib, yet distinct, considering that the latter is a kinase inhibitor. Amifostine inhibited both. The 3D cell metabolomics shed light on the anti-angiogenic effects of amifostine against VEGF-A- and deferoxamine-induced angiogenesis. Amifostine may serve as a dual radioprotective and anti-angiogenic agent in radiotherapy patients.

New approach methods (NAMs) supporting read-across: Two neurotoxicity AOP-based IATA case studies

Read-across approaches are considered key in moving away from in vivo animal testing towards addressing data-gaps using new approach methods (NAMs). Ample successful examples are still required to substantiate this strategy. Here we present and discuss the learnings from two OECD IATA endorsed read-across case studies. They involve two classes of pesticides ­ rotenoids and strobilurins ­ each having a defined mode-of-action that is assessed for its neurological hazard by means of an AOP-based testing strategy coupled to toxicokinetic simulations of human tissue concentrations. The endpoint in question is potential mitochondrial respiratory chain mediated neurotoxicity, specifically through inhibition of complex I or III. An AOP linking inhibition of mitochondrial respiratory chain complex I to the degeneration of dopaminergic neurons formed the basis for both cases but was deployed in two different regulatory contexts. The two cases also exemplify several different read-across concepts: analogue versus category approach, consolidated versus putative AOP, positive versus negative prediction (i.e., neurotoxicity versus low potential for neurotoxicity), and structural versus biological similarity. We applied a range of NAMs to explore the toxicodynamic properties of the compounds, e.g., in silico docking as well as in vitro assays and readouts ­ including transcriptomics ­ in various cell systems, all anchored to the relevant AOPs. Interestingly, although some of the data addressing certain elements of the read-across were associated with high uncertainty, their impact on the overall read-across conclusion remained limited. Coupled to the elaborate regulatory review that the two cases underwent, we propose some generic learnings of AOP-based testing strategies supporting read-across.

Flame: an open source framework for model development, hosting, and usage in production environments.

This article describes Flame, an open source software for building predictive models and supporting their use in production environments. Flame is a web application with a web-based graphic interface, which can be used as a desktop application or installed in a server receiving requests from multiple users. Models can be built starting from any collection of biologically annotated chemical structures since the software supports structural normalization, molecular descriptor calculation, and machine learning model generation using predefined workflows. The model building workflow can be customized from the graphic interface, selecting the type of normalization, molecular descriptors, and machine learning algorithm to be used from a panel of state-of-the-art methods implemented natively. Moreover, Flame implements a mechanism allowing to extend its source code, adding unlimited model customization. Models generated with Flame can be easily exported, facilitating collaborative model development. All models are stored in a model repository supporting model versioning. Models are identified by unique model IDs and include detailed documentation formatted using widely accepted standards. The current version is the result of nearly 3 years of development in collaboration with users from the pharmaceutical industry within the IMI eTRANSAFE project, which aims, among other objectives, to develop high-quality predictive models based on shared legacy data for assessing the safety of drug candidates.

TMSNP: a web server to predict pathogenesis of missense mutations in the transmembrane region of membrane proteins.

The massive amount of data generated from genome sequencing brings tons of newly identified mutations, whose pathogenic/non-pathogenic effects need to be evaluated. This has given rise to several mutation predictor tools that, in general, do not consider the specificities of the various protein groups. We aimed to develop a predictor tool dedicated to membrane proteins, under the premise that their specific structural features and environment would give different responses to mutations compared to globular proteins. For this purpose, we created TMSNP, a database that currently contains information from 2624 pathogenic and 196 705 non-pathogenic reported mutations located in the transmembrane region of membrane proteins. By computing various conservation parameters on these mutations in combination with annotations, we trained a machine-learning model able to classify mutations as pathogenic or not. TMSNP (freely available at http://lmc.uab.es/tmsnp/) improves considerably the prediction power of commonly used mutation predictors trained with globular proteins.

GPCRmd uncovers the dynamics of the 3D-GPCRome.

G-protein-coupled receptors (GPCRs) are involved in numerous physiological processes and are the most frequent targets of approved drugs. The explosion in the number of new three-dimensional (3D) molecular structures of GPCRs (3D-GPCRome) over the last decade has greatly advanced the mechanistic understanding and drug design opportunities for this protein family. Molecular dynamics (MD) simulations have become a widely established technique for exploring the conformational landscape of proteins at an atomic level. However, the analysis and visualization of MD simulations require efficient storage resources and specialized software. Here we present GPCRmd (http://gpcrmd.org/), an online platform that incorporates web-based visualization capabilities as well as a comprehensive and user-friendly analysis toolbox that allows scientists from different disciplines to visualize, analyze and share GPCR MD data. GPCRmd originates from a community-driven effort to create an open, interactive and standardized database of GPCR MD simulations.

Publisher Correction: GPCRmd uncovers the dynamics of the 3D-GPCRome.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

GPCR-SAS: A web application for statistical analyses on G protein-coupled receptors sequences.

G protein-coupled receptors (GPCRs) are one of the largest protein families in mammals. They mediate signal transduction across cell membranes and are important targets for the pharmaceutical industry. The G Protein-Coupled Receptors-Sequence Analysis and Statistics (GPCR-SAS) web application provides a set of tools to perform comparative analysis of sequence positions between receptors, based on a curated structural-informed multiple sequence alignment. The analysis tools include: (i) percentage of occurrence of an amino acid or motif and entropy at a position or range of positions, (ii) covariance of two positions, (iii) correlation between two amino acids in two positions (or two sequence motifs in two ranges of positions), and (iv) snake-plot representation for a specific receptor or for the consensus sequence of a group of selected receptors. The analysis of conservation of residues and motifs across transmembrane (TM) segments may guide the design of more selective ligands or help to rationalize activation mechanisms, among others. As an example, here we analyze the amino acids of the "transmission switch", that initiates receptor activation following ligand binding. The tool is freely accessible at http://lmc.uab.cat/gpcrsas/.