GPCR molecular dynamics forecasting using recurrent neural networks.

G protein-coupled receptors (GPCRs) are a large superfamily of cell membrane proteins that play an important physiological role as transmitters of extracellular signals. Signal transmission through the cell membrane depends on conformational changes in the transmembrane region of the receptor, which makes the investigation of the dynamics in these regions particularly relevant. Molecular dynamics (MD) simulations provide a wealth of data about the structure, dynamics, and physiological function of biological macromolecules by modelling the interactions between their atomic constituents. In this study, a Recurrent and Convolutional Neural Network (RNN) model, namely Long Short-Term Memory (LSTM), is used to predict the dynamics of two GPCR states and three specific simulations of each one, through their activation path and focussing on specific receptor regions. Active and inactive states of the GPCRs are analysed in six scenarios involving APO, Full Agonist (BI 167107) and Partial Inverse Agonist (carazolol) of the receptor. Four Machine Learning models with increasing complexity in terms of neural network architecture are evaluated, and their results discussed. The best method achieves an overall RMSD lower than 0.139 Å and the transmembrane helices are the regions showing the minimum prediction errors and minimum relative movements of the protein.

Layer-Wise Relevance Analysis for Motif Recognition in the Activation Pathway of the ß2-Adrenergic GPCR Receptor.

G-protein-coupled receptors (GPCRs) are cell membrane proteins of relevance as therapeutic targets, and are associated to the development of treatments for illnesses such as diabetes, Alzheimer's, or even cancer. Therefore, comprehending the underlying mechanisms of the receptor functional properties is of particular interest in pharmacoproteomics and in disease therapy at large. Their interaction with ligands elicits multiple molecular rearrangements all along their structure, inducing activation pathways that distinctly influence the cell response. In this work, we studied GPCR signaling pathways from molecular dynamics simulations as they provide rich information about the dynamic nature of the receptors. We focused on studying the molecular properties of the receptors using deep-learning-based methods. In particular, we designed and trained a one-dimensional convolution neural network and illustrated its use in a classification of conformational states: active, intermediate, or inactive, of the ß2-adrenergic receptor when bound to the full agonist BI-167107. Through a novel explainability-oriented investigation of the prediction results, we were able to identify and assess the contribution of individual motifs (residues) influencing a particular activation pathway. Consequently, we contribute a methodology that assists in the elucidation of the underlying mechanisms of receptor activation-deactivation.

Structure of the HOPS tethering complex, a lysosomal membrane fusion machinery.

Lysosomes are essential for cellular recycling, nutrient signaling, autophagy, and pathogenic bacteria and viruses invasion. Lysosomal fusion is fundamental to cell survival and requires HOPS, a conserved heterohexameric tethering complex. On the membranes to be fused, HOPS binds small membrane-associated GTPases and assembles SNAREs for fusion, but how the complex fulfills its function remained speculative. Here, we used cryo-electron microscopy to reveal the structure of HOPS. Unlike previously reported, significant flexibility of HOPS is confined to its extremities, where GTPase binding occurs. The SNARE-binding module is firmly attached to the core, therefore, ideally positioned between the membranes to catalyze fusion. Our data suggest a model for how HOPS fulfills its dual functionality of tethering and fusion and indicate why it is an essential part of the membrane fusion machinery.

Our cells break down the nutrients that they receive from the body to create the building blocks needed to keep us alive. This is done by compartments called lysosomes that are filled with a cocktail of proteins called enzymes, which speed up the breakdown process. Lysosomes are surrounded by a membrane, a barrier of fatty molecules that protects the rest of the cell from being digested. When new nutrients reach the cell, they travel to the lysosome packaged in vesicles, which have their own fatty membrane. To allow the nutrients to enter the lysosome without creating a leak, the membranes of the vesicles and the lysosome must fuse. The mechanism through which these membranes fuse is not fully clear. It is known that both fusing membranes must contain proteins called SNAREs, which wind around each other when they interact. However, this alone is not enough. Other proteins are also required to tether the membranes together before they fuse. To understand how these tethers play a role, Shvarev, Schoppe, König et al. studied the structure of the HOPS complex from yeast. This assembly of six proteins is vital for lysosomal fusion and, has a composition similar to the equivalent complex in humans. Using cryo-electron microscopy, a technique that relies on freezing purified proteins to image them with an electron microscope and reveal their structure, allowed Shvarev, Schoppe, König et al. to provide a model for how HOPS interacts with SNAREs and membranes. In addition to HOPS acting as a tether to bring the membranes together, it can also bind directly to SNAREs. This creates a bridge that allows the proteins to wrap around each other, driving the membranes to fuse. HOPS is a crucial component in the cellular machinery, and mutations in the complex can cause devastating neurological defects. The complex is also targeted by viruses ­ such as SARS-CoV-2 ­ that manipulate HOPS to reduce its activity. Shvarev, Schoppe, König et al.'s findings could help researchers to develop drugs to maintain or recover the activity of HOPS. However, this will require additional information about its structure and how the complex acts in the biological environment of the cell.

Sensitivity Analysis of Sensors in a Hydraulic Condition Monitoring System Using CNN Models.

Condition monitoring (CM) is a useful application in industry 4.0, where the machine's health is controlled by computational intelligence methods. Data-driven models, especially from the field of deep learning, are efficient solutions for the analysis of time series sensor data due to their ability to recognize patterns in high dimensional data and to track the temporal evolution of the signal. Despite the excellent performance of deep learning models in many applications, additional requirements regarding the interpretability of machine learning models are getting relevant. In this work, we present a study on the sensitivity of sensors in a deep learning based CM system providing high-level information about the relevance of the sensors. Several convolutional neural networks (CNN) have been constructed from a multisensory dataset for the prediction of different degradation states in a hydraulic system. An attribution analysis of the input features provided insights about the contribution of each sensor in the prediction of the classifier. Relevant sensors were identified, and CNN models built on the selected sensors resulted equal in prediction quality to the original models. The information about the relevance of sensors is useful for the system's design to decide timely on the required sensors.

Using machine learning tools for protein database biocuration assistance.

Biocuration in the omics sciences has become paramount, as research in these fields rapidly evolves towards increasingly data-dependent models. As a result, the management of web-accessible publicly-available databases becomes a central task in biological knowledge dissemination. One relevant challenge for biocurators is the unambiguous identification of biological entities. In this study, we illustrate the adequacy of machine learning methods as biocuration assistance tools using a publicly available protein database as an example. This database contains information on G Protein-Coupled Receptors (GPCRs), which are part of eukaryotic cell membranes and relevant in cell communication as well as major drug targets in pharmacology. These receptors are characterized according to subtype labels. Previous analysis of this database provided evidence that some of the receptor sequences could be affected by a case of label noise, as they appeared to be too consistently misclassified by machine learning methods. Here, we extend our analysis to recent and quite substantially modified new versions of the database and reveal their now extremely accurate labeling using several machine learning models and different transformations of the unaligned sequences. These findings support the adequacy of our proposed method to identify problematic labeling cases as a tool for database biocuration.

Systematic Analysis of Primary Sequence Domain Segments for the Discrimination Between Class C GPCR Subtypes.

G-protein-coupled receptors (GPCRs) are a large and diverse super-family of eukaryotic cell membrane proteins that play an important physiological role as transmitters of extracellular signal. In this paper, we investigate Class C, a member of this super-family that has attracted much attention in pharmacology. The limited knowledge about the complete 3D crystal structure of Class C receptors makes necessary the use of their primary amino acid sequences for analytical purposes. Here, we provide a systematic analysis of distinct receptor sequence segments with regard to their ability to differentiate between seven class C GPCR subtypes according to their topological location in the extracellular, transmembrane, or intracellular domains. We build on the results from the previous research that provided preliminary evidence of the potential use of separated domains of complete class C GPCR sequences as the basis for subtype classification. The use of the extracellular N-terminus domain alone was shown to result in a minor decrease in subtype discrimination in comparison with the complete sequence, despite discarding much of the sequence information. In this paper, we describe the use of Support Vector Machine-based classification models to evaluate the subtype-discriminating capacity of the specific topological sequence segments.

Label noise in subtype discrimination of class C G protein-coupled receptors: A systematic approach to the analysis of classification errors.

BACKGROUND: The characterization of proteins in families and subfamilies, at different levels, entails the definition and use of class labels. When the adscription of a protein to a family is uncertain, or even wrong, this becomes an instance of what has come to be known as a label noise problem. Label noise has a potentially negative effect on any quantitative analysis of proteins that depends on label information. This study investigates class C of G protein-coupled receptors, which are cell membrane proteins of relevance both to biology in general and pharmacology in particular. Their supervised classification into different known subtypes, based on primary sequence data, is hampered by label noise. The latter may stem from a combination of expert knowledge limitations and the lack of a clear correspondence between labels that mostly reflect GPCR functionality and the different representations of the protein primary sequences. RESULTS: In this study, we describe a systematic approach, using Support Vector Machine classifiers, to the analysis of G protein-coupled receptor misclassifications. As a proof of concept, this approach is used to assist the discovery of labeling quality problems in a curated, publicly accessible database of this type of proteins. We also investigate the extent to which physico-chemical transformations of the protein sequences reflect G protein-coupled receptor subtype labeling. The candidate mislabeled cases detected with this approach are externally validated with phylogenetic trees and against further trusted sources such as the National Center for Biotechnology Information, Universal Protein Resource, European Bioinformatics Institute and Ensembl Genome Browser information repositories. CONCLUSIONS: In quantitative classification problems, class labels are often by default assumed to be correct. Label noise, though, is bound to be a pervasive problem in bioinformatics, where labels may be obtained indirectly through complex, many-step similarity modelling processes. In the case of G protein-coupled receptors, methods capable of singling out and characterizing those sequences with consistent misclassification behaviour are required to minimize this problem. A systematic, Support Vector Machine-based method has been proposed in this study for such purpose. The proposed method enables a filtering approach to the label noise problem and might become a support tool for database curators in proteomics.

Reducing the n-gram feature space of class C GPCRs to subtype-discriminating patterns.

G protein-coupled receptors (GPCRs) are a large and heterogeneous superfamily of receptors that are key cell players for their role as extracellular signal transmitters. Class C GPCRs, in particular, are of great interest in pharmacology. The lack of knowledge about their full 3-D structure prompts the use of their primary amino acid sequences for the construction of robust classifiers, capable of discriminating their different subtypes. In this paper, we investigate the use of feature selection techniques to build Support Vector Machine (SVM)-based classification models from selected receptor subsequences described as n-grams. We show that this approach to classification is useful for finding class C GPCR subtype-specific motifs.

Copeptin, a stable peptide derived from the vasopressin precursor, correlates with the individual stress level.

BACKGROUND: During stress, vasopressin is a potent synergistic factor of CRH as a hypothalamic stimulator of the HPA axis. The measurements of CRH and vasopressin levels are cumbersome because of their instability and short half-life. Copeptin is a more stable peptide stoichiometrically released from the same precursor molecule. The aim of our study was to compare copeptin and cortisol levels in different stress situations. METHODS: Three groups of patients with increasing stress levels were investigated: a) healthy controls without apparent stress (n=20), b) hospitalized medical patients with moderate stress (n=25) and c) surgical patients 30 minutes after extubation, with maximal stress (n=29). In all patients we assessed cortisol and copeptin levels. Copeptin levels were measured with a new sandwich immunoassay. RESULTS: Cortisol levels in controls were (median, IQ range, 486 [397-588] nmol/L), not significantly different as compared to medical patients (438 [371-612] nmol/L, p=0.69). Cortisol levels in surgical patients after extubation were higher (744 [645-1062] nmol/L p<0.01 vs controls and medical patients). Copeptin levels in controls were 4.3 [3.2-5.5] pmol/L, which was lower as compared to medical patients (17.5 [6.4-24.1], p<0.001) and surgical patients after extubation (67.5 [37.8-110.0] pmol/L, p<0.001). The correlation between copeptin levels and cortisol was r=0.46, p<0.001. CONCLUSION: Copeptin is a novel marker of the individual stress level. It more subtly mirrors moderate stress as compared to cortisol values.

Measurement of serum free cortisol shows discordant responsivity to stress and dynamic evaluation.

CONTEXT: Routinely available assays of adrenal function measure serum total cortisol (TC) and not the biologically active free cortisol (FC). However, there are few data on FC levels during surgical stress and in response to standard pharmacological tests. OBJECTIVE: Our objective was to evaluate TC and FC levels in different states of physical stress. DESIGN AND SETTING: We conducted a prospective observational study in a university hospital. PARTICIPANTS AND MAIN OUTCOME MEASURES: We measured TC and FC levels in 64 patients: group A, 17 healthy controls without stress; group B, 23 medical patients with moderate stress; and group C, 24 surgical patients undergoing coronary bypass grafting. Cortisol levels in group C were measured basally and at several time points thereafter and were compared with responsivity to a pharmacological dose of ACTH. FC was measured using equilibrium dialysis. RESULTS: In group C patients after extubation, the relative increase above basal FC was higher than the increase in TC levels (399 +/- 266 vs. 247 +/- 132% of initial values, respectively; mean +/- sd; P = 0.02) and then fell more markedly, FC levels falling to 67 +/- 49% and TC levels to 79 +/- 36% (P = 0.04). After ACTH stimulation, TC levels increased to 680 +/- 168 nmol/liter, which was similar to the increase with major stress (811 +/- 268 nmol/liter). In contrast, FC levels increased to 55 +/- 16 nmol/liter after ACTH stimulation but significantly greater with surgical stress to 108 +/- 56 nmol/liter (P < 0.001). CONCLUSION: The more pronounced increase in FC seen during stress as compared with the ACTH test suggests that this test does not adequately anticipate the FC levels needed during severe stress.

Cortisol response in relation to the severity of stress and illness.

BACKGROUND: The aim of the study was to compare the adrenal response, the course of the ACTH/cortisol ratio, as well as the variance and the diagnostic performance of different cutoffs after 1 and 250 microg ACTH stimulation in different stress situations. METHODS: We investigated three groups with increasing stress levels: ambulatory controls (group A; n = 20), hospitalized medical patients (group B; n = 25), and patients undergoing coronary artery bypass grafting (group C; n = 29). All subjects underwent four consecutive ACTH stimulation tests and were randomized to either a 1- or 250-microg dose. RESULTS: Stimulated cortisol levels in group A were similar to basal cortisol levels under maximal stress (C3; P = 0.8). Peak cortisol concentrations were higher after 250 microg compared with 1 microg ACTH in group B (P = 0.006) and under maximal stress after extubation (group C3; P = 0.027), whereas there were no differences in group A. The ACTH/cortisol ratio was lower in surgical patients after extubation compared with unstressed conditions (P < or = 0.03) The within-subject variance was similar in ambulatory controls and medical patients and after both ACTH doses (all 17-36% of total variance). Cutoff dependent, the diagnosis of relative adrenal insufficiency would have been made in 0-58.3%, respectively. CONCLUSION: In moderate and major stress situations, cortisol concentrations in patients without hypothalamic-pituitary-adrenal disease were higher after stimulation with 250 microg compared with 1 mug ACTH. Data from our study give insight into the physiological adaptations of the hypothalamic-pituitary-adrenal axis to stress.