From Proteins to Ligands: Decoding Deep Learning Methods for Binding Affinity Prediction.

Accurate in silico prediction of protein-ligand binding affinity is important in the early stages of drug discovery. Deep learning-based methods exist but have yet to overtake more conventional methods such as giga-docking largely due to their lack of generalizability. To improve generalizability, we need to understand what these models learn from input protein and ligand data. We systematically investigated a sequence-based deep learning framework to assess the impact of protein and ligand encodings on predicting binding affinities for commonly used kinase data sets. The role of proteins is studied using convolutional neural network-based encodings obtained from sequences and graph neural network-based encodings enriched with structural information from contact maps. Ligand-based encodings are generated from graph-neural networks. We test different ligand perturbations by randomizing node and edge properties. For proteins, we make use of 3 different protein contact generation methods (AlphaFold2, Pconsc4, and ESM-1b) and compare these with a random control. Our investigation shows that protein encodings do not substantially impact the binding predictions, with no statistically significant difference in binding affinity for KIBA in the investigated metrics (concordance index, Pearson's R Spearman's Rank, and RMSE). Significant differences are seen for ligand encodings with random ligands and random ligand node properties, suggesting a much bigger reliance on ligand data for the learning tasks. Using different ways to combine protein and ligand encodings did not show a significant change in performance.

Benchmarking Active Learning Protocols for Ligand-Binding Affinity Prediction.

Active learning (AL) has become a powerful tool in computational drug discovery, enabling the identification of top binders from vast molecular libraries. To design a robust AL protocol, it is important to understand the influence of AL parameters, as well as the features of the data sets on the outcomes. We use four affinity data sets for different targets (TYK2, USP7, D2R, Mpro) to systematically evaluate the performance of machine learning models [Gaussian process (GP) model and Chemprop model], sample selection protocols, and the batch size based on metrics describing the overall predictive power of the model (R2, Spearman rank, root-mean-square error) as well as the accurate identification of top 2%/5% binders (Recall, F1 score). Both models have a comparable Recall of top binders on large data sets, but the GP model surpasses the Chemprop model when training data are sparse. A larger initial batch size, especially on diverse data sets, increased the Recall of both models as well as overall correlation metrics. However, for subsequent cycles, smaller batch sizes of 20 or 30 compounds proved to be desirable. Furthermore, adding artificial Gaussian noise to the data up to a certain threshold still allowed the model to identify clusters with top-scoring compounds. However, excessive noise (<1σ) did impact the model's predictive and exploitative capabilities.

Sire: An interoperability engine for prototyping algorithms and exchanging information between molecular simulation programs.

Sire is a Python/C++ library that is used both to prototype new algorithms and as an interoperability engine for exchanging information between molecular simulation programs. It provides a collection of file parsers and information converters that together make it easier to combine and leverage the functionality of many other programs and libraries. This empowers researchers to use sire to write a single script that can, for example, load a molecule from a PDBx/mmCIF file via Gemmi, perform SMARTS searches via RDKit, parameterize molecules using BioSimSpace, run GPU-accelerated molecular dynamics via OpenMM, and then display the resulting dynamics trajectory in a NGLView Jupyter notebook 3D molecular viewer. This functionality is built on by BioSimSpace, which uses sire's molecular information engine to interconvert with programs such as GROMACS, NAMD, Amber, and AmberTools for automated molecular parameterization and the running of molecular dynamics, metadynamics, and alchemical free energy workflows. Sire comes complete with a powerful molecular information search engine, plus trajectory loading and editing, analysis, and energy evaluation engines. This, when combined with an in-built computer algebra system, gives substantial flexibility to researchers to load, search for, edit, and combine molecular information from multiple sources and use that to drive novel algorithms by combining functionality from other programs. Sire is open source (GPL3) and is available via conda and at a free Jupyter notebook server at https://try.openbiosim.org. Sire is supported by the not-for-profit OpenBioSim community interest company.

What geometrically constrained models can tell us about real-world protein contact maps.

The mechanisms by which a protein's 3D structure can be determined based on its amino acid sequence have long been one of the key mysteries of biophysics. Often simplistic models, such as those derived from geometric constraints, capture bulk real-world 3D protein-protein properties well. One approach is using protein contact maps (PCMs) to better understand proteins' properties. In this study, we explore the emergent behaviour of contact maps for different geometrically constrained models and compare them to real-world protein systems. Specifically, we derive an analytical approximation for the distribution of amino acid distances, denoted asP(s), using a mean-field approach based on a geometric constraint model. This approximation is then validated for amino acid distance distributions generated from a 2D and 3D version of the geometrically constrained random interaction model. For real protein data, we show how the analytical approximation can be used to fit amino acid distance distributions of protein chain lengths ofL ≈ 100,L ≈ 200, andL ≈ 300 generated from two different methods of evaluating a PCM, a simple cutoff based method and a shadow map based method. We present evidence that geometric constraints are sufficient to model the amino acid distance distributions of protein chains in bulk and amino acid sequences only play a secondary role, regardless of the definition of the PCM.

18S rDNA sequence-structure phylogeny of the Euglenophyceae (Euglenozoa, Euglenida).

The phylogeny of Euglenophyceae (Euglenozoa, Euglenida) has been discussed for decades with new genera being described in the last few years. In this study, we reconstruct a phylogeny using 18S rDNA sequence and structural data simultaneously. Using homology modeling, individual secondary structures were predicted. Sequence-structure data are encoded and automatically aligned. Here, we present a sequence-structure neighbor-joining tree of more than 300 taxa classified as Euglenophyceae. Profile neighbor-joining was used to resolve the basal branching pattern. Neighbor-joining, maximum parsimony, and maximum likelihood analyses were performed using sequence-structure information for manually chosen subsets. All analyses supported the monophyly of Eutreptiella, Discoplastis, Lepocinclis, Strombomonas, Cryptoglena, Monomorphina, Euglenaria, and Colacium. Well-supported topologies were generally consistent with previous studies using a combined dataset of genetic markers. Our study supports the simultaneous use of sequence and structural data to reconstruct more accurate and robust trees. The average bootstrap value is significantly higher than the average bootstrap value obtained from sequence-only analyses, which is promising for resolving relationships between more closely related taxa.

SILVR: Guided Diffusion for Molecule Generation.

Computationally generating new synthetically accessible compounds with high affinity and low toxicity is a great challenge in drug design. Machine learning models beyond conventional pharmacophoric methods have shown promise in the generation of novel small-molecule compounds but require significant tuning for a specific protein target. Here, we introduce a method called selective iterative latent variable refinement (SILVR) for conditioning an existing diffusion-based equivariant generative model without retraining. The model allows the generation of new molecules that fit into a binding site of a protein based on fragment hits. We use the SARS-CoV-2 main protease fragments from Diamond XChem that form part of the COVID Moonshot project as a reference dataset for conditioning the molecule generation. The SILVR rate controls the extent of conditioning, and we show that moderate SILVR rates make it possible to generate new molecules of similar shape to the original fragments, meaning that the new molecules fit the binding site without knowledge of the protein. We can also merge up to 3 fragments into a new molecule without affecting the quality of molecules generated by the underlying generative model. Our method is generalizable to any protein target with known fragments and any diffusion-based model for molecule generation.

Dynamic musculoskeletal ultrasound: slipping rib, muscle hernia, snapping hip, and peroneal tendon pathology.

Dynamic musculoskeletal ultrasound is an important diagnostic tool that allows the practitioner to observe soft tissue structures over a range of motion and identify pathology not diagnosed on other modalities. Familiarity with this modality allows health care practitioners to appropriately refer patients for this type of examination. This article will review several indications for dynamic ultrasound imaging, including slipping rib, muscle hernia, snapping hip, and peroneal tendon pathology. The examination technique and expected findings for common pathology in each location are discussed.

Truncal sensory polyneuropathy in light-chain amyloidosis.

We describe a case of truncal sensory polyneuropathy in a patient with light-chain amyloidosis. We highlight the clinical signs and differential diagnoses related to the presentation.

Spectroscopic evaluation of UVI-cement mineral interactions: ettringite and hydrotalcite.

Portland cement based grouts used for radioactive waste immobilization contain high replacement levels of supplementary cementitious materials, including blast-furnace slag and fly ash. The minerals formed upon hydration of these cements may have capacity for binding actinide elements present in radioactive waste. In this work, the minerals ettringite (Ca6Al2(SO4)3(OH)12·26H2O) and hydrotalcite (Mg6Al2(OH)16CO3·4H2O) were selected to investigate the importance of minor cement hydrate phases in sequestering and immobilizing UVI from radioactive waste streams. U LIII-edge X-ray absorption spectroscopy (XAS) was used to probe the UVI coordination environment in contact with these minerals. For the first time, solid-state 27Al magic angle spinning nuclear magnetic resonance (MAS-NMR) spectroscopy was applied to probe the Al coordination environment in these UVI-contacted minerals and make inferences on the UVI coordination, in conjunction with the X-ray spectroscopy analyses. The U LIII-edge XAS analysis of the UVI-contacted ettringite phases found them to be similar (>∼70%) to the uranyl oxyhydroxides present in a mixed becquerelite/metaschoepite mineral. Fitting of the EXAFS region, in combination with 27Al NMR analysis, indicated that a disordered Ca- or Al-bearing UVI secondary phase also formed. For the UVI-contacted hydrotalcite phases, the XAS and 27Al NMR data were interpreted as being similar to uranyl carbonate, that was likely Mg-containing.

Near-Fatal Self-Harm Among Canadian Adolescents.

OBJECTIVE: To evaluate the clinical features of Canadian adolescents admitted to the intensive care unit (ICU) for medically serious self-harm. METHODS: 2700 Canadian paediatricians were surveyed monthly over two years (January 2017 to December 2018) through the Canadian Paediatric Surveillance Program to ascertain data from eligible cases. RESULTS: Ninety-three cases (73 female; age 15.2 ± 1.5) met the case definition. Four provinces reported the majority of cases: Quebec (n = 27), Ontario (n = 26), Alberta (n = 21), and British Columbia (n = 8). There were 10 deaths, 9 by hanging. Overdose and hanging were the most frequently reported methods of self-harm (74.2% and 19.4%, respectively). Overdose was more common in females (80.8% females vs. 50% males; χ2 = 7.8 (1), p = .005), whereas hanging was more common in males (35% males vs. 15.1% females, χ2 = 3.9 (1), p = .04). More females than males had a past psychiatric diagnosis (79% vs. 58%; χ2 = 4.1 (1), p = .06), a previous suicide attempt (55.9% vs. 29.4%, χ2 = 3.8 (1), p = .05), and prior use of mental health service (69.7% vs. 27.8%, χ2 = 10.4 (1), p = .001). Family conflict was the most commonly identified precipitating factor (43%) of self-harm. CONCLUSIONS: Among Canadian adolescents admitted to the ICU with medically serious self-harm, females demonstrate a higher rate of suicide attempts and prior mental health care engagement, whereas males are more likely to die by suicide. These findings are consistent with data from other adolescent samples, as well as data from working-age and older adults. Therefore, a sex-specific approach to suicide prevention is warranted as part of a national suicide prevention strategy; family conflict may be a specific target for suicide prevention interventions among adolescents.

Current approaches to the diagnosis and management of amyloidosis.

Amyloidosis is a collection of diseases caused by the misfolding of proteins that aggregate into insoluble amyloid fibrils and deposit in tissues. While these fibrils may aggregate to form insignificant localised deposits, they can also accumulate in multiple organs to the extent that amyloidosis can be an immediately life-threatening disease, requiring urgent treatment. Recent advances in diagnostic techniques and therapies are dramatically changing the disease landscape and patient prognosis. Delays in diagnosis and treatment remain the greatest challenge, necessitating physician awareness of the common clinical presentations that suggest amyloidosis. The most common types are transthyretin (ATTR) amyloidosis followed by immunoglobulin light-chain (AL) amyloidosis. While systemic AL amyloidosis was previously considered a death sentence with no effective therapies, significant improvement in patient survival has occurred over the past 2 decades, driven by greater understanding of the disease process, risk-adapted adoption of myeloma therapies such as proteosome inhibitors (bortezomib) and monoclonal antibodies (daratumumab) and improved supportive care. ATTR amyloidosis is an underdiagnosed cause of heart failure. Technetium scintigraphy has made noninvasive diagnosis much easier, and ATTR is now recognised as the most common type of amyloidosis because of the increased identification of age-related ATTR. There are emerging ATTR treatments that slow disease progression, decrease patient hospitalisations and improve patient quality of life and survival. This review aims to update physicians on recent developments in amyloidosis diagnosis and management and to provide a diagnostic and treatment framework to improve the management of patients with all forms of amyloidosis.

Carcass characteristics and meat quality of goats fed increasing levels of crude glycerin.

Crude glycerin is a byproduct of the biodiesel industry and has been widely used in ruminant diets as a source of energy, usually in place of corn, primarily during periods of drought in tropical regions. The objective of this study was to evaluate the effect of including levels of the crude glycerin of low purity (0, 6, 12 and 18%) replacing corn in the diets of goats on the carcass characteristics, tissue composition, meat cuts yield and physicochemical parameters of meat. Forty males castrated without defined racial pattern goats an initial average weight of 19.70 ± 2.30 kg were slaughtered after 86 days. Diets content 0 and 6% crude glycerin promoted similar responses to the analyzed variables, except for pH and breast weight. No differences were observed to total digestible nutrients, slaughter body weight, commercial cut yield leg tissue composition and physicochemical parameters of meat. Crude glycerin can be included up to 12% without losses on carcass weight and meat cuts, leg composition, and meat quality. The inclusion of crude glycerin containing 63.06% glycerol and 45.57% lipids could be effective in partial replacement of corn in diets for confined goats in tropical areas.

Characteristics of Pediatric Frequent Users of Emergency Departments in Alberta and Ontario.

OBJECTIVES: Emergency department (ED) volumes have drawn attention to frequent users but less attention has been paid to children. This study examined sociodemographic and ED presentation characteristics of pediatric high-system ED users (HSUs) in 2 provinces in Canada. METHODS: Cohorts of HSUs were created from the National Ambulatory Care Reporting System in 2015/2016 for children with the top 10% of ED presentations. Controls were random samples of non-HSU patients. Factors were explored in multivariable logistic regression models. RESULTS: There were 151,497 HSUs (51.7% girls, average age, 6.4 years) and 591,545 controls (53.1% girls; average age, 7.4 years). High-system ED users were more likely to be younger (adjusted odds ratio [aOR], 0.89 per 5 years; 95% confidence interval [CI], 0.88-0.89), live in less populated areas (aOR, 1.85; 95% CI, 1.82-1.88), and from lowest income neighborhoods (aOR, 1.51; 95% CI, 1.48-1.54) than controls. High-system ED users had higher proportions of presentations for pediatric complex chronic (aOR, 1.25 per 0.25 increase; 95% CI, 1.21-1.29), respiratory (aOR, 1.14 per 0.25; 95% CI, 1.12-1.15), and mental health (aOR, 1.14 per 0.25; 95% CI, 1.13-1.16) conditions than controls. CONCLUSIONS: Complex factors underlie pediatric health care utilization decisions. Findings identified conditions to target in interventions to improve health care access and utilization. Future work should engage children and families to design interventions.

Neurology and the histiocytoses: a case of Rosai-Dorfman-Destombes disease.

The histiocytoses are a group of rare disorders characterised by the accumulation of neoplastic or non-neoplastic activated histiocytes in various tissues. Phenotypes vary widely from cutaneous lesions or lymphadenopathy that regress spontaneously to disseminated disease with poor prognosis. Neurological symptoms can be a presenting feature or appear during the course of disease. We present a challenging diagnostic and management case of Rosai-Dorfman-Destombes disease in a 48-year-old woman with a relapsing, partially steroid-responsive syndrome comprising patchy, non-length-dependent radiculoneuropathy with diffuse pachymeningitis and widespread systemic disease, and recent dramatic response to novel mitogen-activated kinase pathway inhibition. We discuss the clinical characteristics, diagnosis, recent breakthroughs in pathogenesis and emerging treatment options for Rosai-Dorfman disease and for the histiocytoses with neurological sequelae, including Langerhans cell histiocytosis and Erdheim-Chester disease.

Second-line nivolumab in relapsed small-cell lung cancer: CheckMate 331☆.

BACKGROUND: Patients with relapsed small-cell lung cancer (SCLC) have few treatment options and dismal survival. Phase I/II data show activity of nivolumab in previously treated SCLC. PATIENTS AND METHODS: CheckMate 331 is a randomized, open-label, phase III trial of nivolumab versus standard chemotherapy in relapsed SCLC. Patients with relapse after first-line, platinum-based chemotherapy were randomized 1 : 1 to nivolumab 240 mg every 2 weeks or chemotherapy (topotecan or amrubicin) until progression or unacceptable toxicity. Primary endpoint was overall survival (OS). RESULTS: Overall, 284 patients were randomized to nivolumab and 285 to chemotherapy. Minimum follow-up was 15.8 months. No significant improvement in OS was seen with nivolumab versus chemotherapy [median OS, 7.5 versus 8.4 months; hazard ratio (HR), 0.86; 95% confidence interval (CI), 0.72-1.04; P = 0.11]. A survival benefit with nivolumab was suggested in patients with baseline lactate dehydrogenase ≤ upper limit of normal and in those without baseline liver metastases. OS (nivolumab versus chemotherapy) was similar in patients with programmed death-ligand 1 combined positive score ≥1% versus <1%. Median progression-free survival was 1.4 versus 3.8 months (HR, 1.41; 95% CI, 1.18-1.69). Objective response rate was 13.7% versus 16.5% (odds ratio, 0.80; 95% CI, 0.50-1.27); median duration of response was 8.3 versus 4.5 months. Rates of grade 3 or 4 treatment-related adverse events were 13.8% versus 73.2%. CONCLUSION: Nivolumab did not improve survival versus chemotherapy in relapsed SCLC. No new safety signals were seen. In exploratory analyses, select baseline characteristics were associated with improved OS for nivolumab.

Clinical definition of acquired resistance to immunotherapy in patients with metastatic non-small-cell lung cancer.

Acquired resistance (AR) to programmed cell death protein 1/programmed death-ligand 1 [PD-(L)1] blockade is frequent in non-small-cell lung cancer (NSCLC), occurring in a majority of initial responders. Patients with AR may have unique properties of persistent antitumor immunity that could be re-harnessed by investigational immunotherapies. The absence of a consistent clinical definition of AR to PD-(L)1 blockade and lack of uniform criteria for ensuing enrollment in clinical trials remains a major barrier to progress; such clinical definitions have advanced biologic and therapeutic discovery. We examine the considerations and potential controversies in developing a patient-level definition of AR in NSCLC treated with PD-(L)1 blockade. Taking into account the specifics of NSCLC biology and corresponding treatment strategies, we propose a practical, clinical definition of AR to PD-(L)1 blockade for use in clinical reports and prospective clinical trials. Patients should meet the following criteria: received treatment that includes PD-(L)1 blockade; experienced objective response on PD-(L)1 blockade (inclusion of a subset of stable disease will require future investigation); have progressive disease occurring within 6 months of last anti-PD-(L)1 antibody treatment or rechallenge with anti-PD-(L)1 antibody in patients not exposed to anti-PD-(L)1 in 6 months.

Implementation of the QUBE Force Field in SOMD for High-Throughput Alchemical Free-Energy Calculations.

The quantum mechanical bespoke (QUBE) force-field approach has been developed to facilitate the automated derivation of potential energy function parameters for modeling protein-ligand binding. To date, the approach has been validated in the context of Monte Carlo simulations of protein-ligand complexes. We describe here the implementation of the QUBE force field in the alchemical free-energy calculation molecular dynamics simulation package SOMD. The implementation is validated by demonstrating the reproducibility of absolute hydration free energies computed with the QUBE force field across the SOMD and GROMACS software packages. We further demonstrate, by way of a case study involving two series of non-nucleoside inhibitors of HIV-1 reverse transcriptase, that the availability of QUBE in a modern simulation package that makes efficient use of graphics processing unit acceleration will facilitate high-throughput alchemical free-energy calculations.

Dynamic Profiling of ß-Coronavirus 3CL Mpro Protease Ligand-Binding Sites.

ß-coronavirus (CoVs) alone has been responsible for three major global outbreaks in the 21st century. The current crisis has led to an urgent requirement to develop therapeutics. Even though a number of vaccines are available, alternative strategies targeting essential viral components are required as a backup against the emergence of lethal viral variants. One such target is the main protease (Mpro) that plays an indispensable role in viral replication. The availability of over 270 Mpro X-ray structures in complex with inhibitors provides unique insights into ligand-protein interactions. Herein, we provide a comprehensive comparison of all nonredundant ligand-binding sites available for SARS-CoV2, SARS-CoV, and MERS-CoV Mpro. Extensive adaptive sampling has been used to investigate structural conservation of ligand-binding sites using Markov state models (MSMs) and compare conformational dynamics employing convolutional variational auto-encoder-based deep learning. Our results indicate that not all ligand-binding sites are dynamically conserved despite high sequence and structural conservation across ß-CoV homologs. This highlights the complexity in targeting all three Mpro enzymes with a single pan inhibitor.

Nerve biopsy in acquired neuropathies.

A diagnosis of neuropathy can typically be determined through clinical assessment and focused investigation. With technological advances, including significant progress in genomics, the role of nerve biopsy has receded over recent years. However, making a specific and, in some cases, tissue-based diagnosis is essential across a wide array of potentially treatable acquired peripheral neuropathies. When laboratory investigations do not suggest a definitive diagnosis, nerve biopsy remains the final step to ascertain the etiology of the disease. The present review highlights the utility of nerve biopsy in confirming a diagnosis, while further illustrating the importance of a tissue-based diagnosis in relation to treatment strategies, particularly when linked to long-term immunosuppressive therapies.

Gender-related differences in the modulation of anthropometric, biochemical, and immune markers by physical activity in hypertensive and diabetic individuals.

Systemic arterial hypertension (SAH) and type 2 diabetes mellitus (T2DM) compose the two major noncommunicable chronic inflammatory diseases. Physical activity has been shown as a promising complementary approach to control the systemic inflammation. However, it is still unclear whether this modulation is gender-dependent. The objective of this study was evaluate the gender-related influence of physical activity on the inflammatory response and biochemical profile of individuals with SAH and T2DM. An international physical activity questionnaire was applied to 376 individuals diagnosed with SAH and T2DM in order to access their exercises routine and was evaluated the influence of physical activity in biochemical, anthropometrical, and immunological markers involved in these disorders in men and women. Even though active individuals have exhibited lower serum levels of IL-1ß, IFN-γ, TNF-α, and IL-17A, the ratios between IL-10 and all inflammatory cytokines were higher in men than in women. Physically active individuals also demonstrated increased HDL/LDL and HDL/VLDL ratios. Moreover, multiple correlations revealed that in active women both IL-10 and TNF-α serum levels positively correlate with fasting glucose levels, and were negatively associated with HDL levels. Our findings suggest that gender-related differences dictate a distinct crosstalk between inflammatory and biochemical markers in physically active individuals.