Ligand-Based Compound Activity Prediction via Few-Shot Learning.

Predicting the activities of new compounds against biophysical or phenotypic assays based on the known activities of one or a few existing compounds is a common goal in early stage drug discovery. This problem can be cast as a "few-shot learning" challenge, and prior studies have developed few-shot learning methods to classify compounds as active versus inactive. However, the ability to go beyond classification and rank compounds by expected affinity is more valuable. We describe Few-Shot Compound Activity Prediction (FS-CAP), a novel neural architecture trained on a large bioactivity data set to predict compound activities against an assay outside the training set, based on only the activities of a few known compounds against the same assay. Our model aggregates encodings generated from the known compounds and their activities to capture assay information and uses a separate encoder for the new compound whose activity is to be predicted. The new method provides encouraging results relative to traditional chemical-similarity-based techniques as well as other state-of-the-art few-shot learning methods in tests on a variety of ligand-based drug discovery settings and data sets. The code for FS-CAP is available at https://github.com/Rose-STL-Lab/FS-CAP.

Conformation selection by ATP-competitive inhibitors and allosteric communication in ERK2.

Activation of the extracellular signal-regulated kinase-2 (ERK2) by phosphorylation has been shown to involve changes in protein dynamics, as determined by hydrogen-deuterium exchange mass spectrometry (HDX-MS) and NMR relaxation dispersion measurements. These can be described by a global exchange between two conformational states of the active kinase, named 'L' and 'R,' where R is associated with a catalytically productive ATP-binding mode. An ATP-competitive ERK1/2 inhibitor, Vertex-11e, has properties of conformation selection for the R-state, revealing movements of the activation loop that are allosterically coupled to the kinase active site. However, the features of inhibitors important for R-state selection are unknown. Here, we survey a panel of ATP-competitive ERK inhibitors using HDX-MS and NMR and identify 14 new molecules with properties of R-state selection. They reveal effects propagated to distal regions in the P+1 and helix αF segments surrounding the activation loop, as well as helix αL16. Crystal structures of inhibitor complexes with ERK2 reveal systematic shifts in the Gly loop and helix αC, mediated by a Tyr-Tyr ring stacking interaction and the conserved Lys-Glu salt bridge. The findings suggest a model for the R-state involving small movements in the N-lobe that promote compactness within the kinase active site and alter mobility surrounding the activation loop. Such properties of conformation selection might be exploited to modulate the protein docking interface used by ERK substrates and effectors.

Target-Free Compound Activity Prediction via Few-Shot Learning.

Predicting the activities of compounds against protein-based or phenotypic assays using only a few known compounds and their activities is a common task in target-free drug discovery. Existing few-shot learning approaches are limited to predicting binary labels (active/inactive). However, in real-world drug discovery, degrees of compound activity are highly relevant. We study Few-Shot Compound Activity Prediction (FS-CAP) and design a novel neural architecture to meta-learn continuous compound activities across large bioactivity datasets. Our model aggregates encodings generated from the known compounds and their activities to capture assay information. We also introduce a separate encoder for the unknown compound. We show that FS-CAP surpasses traditional similarity-based techniques as well as other state of the art few-shot learning methods on a variety of target-free drug discovery settings and datasets.

Conformation Selection by ATP-competitive Inhibitors and Allosteric Communication in ERK2.

Activation of the extracellular signal regulated kinase-2 (ERK2) by phosphorylation has been shown to involve changes in protein dynamics, as determined by hydrogen-deuterium exchange mass spectrometry (HDX-MS) and NMR relaxation dispersion measurements. These can be described by a global exchange between two conformational states of the active kinase, named "L" and "R", where R is associated with a catalytically productive ATP-binding mode. An ATP-competitive ERK1/2 inhibitor, Vertex-11e, has properties of conformation selection for the R-state, revealing movements of the activation loop that are allosterically coupled to the kinase active site. However, the features of inhibitors important for R-state selection are unknown. Here we survey a panel of ATP-competitive ERK inhibitors using HDX-MS and NMR and identify 14 new molecules with properties of R-state selection. They reveal effects propagated to distal regions in the P+1 and helix αF segments surrounding the activation loop, as well as helix αL16. Crystal structures of inhibitor complexes with ERK2 reveal systematic shifts in the Gly loop and helix αC, mediated by a Tyr-Tyr ring stacking interaction and the conserved Lys-Glu salt bridge. The findings suggest a model for the R-state involving small movements in the N-lobe that promote compactness within the kinase active site and alter mobility surrounding the activation loop. Such properties of conformation selection might be exploited to modulate the protein docking interface used by ERK substrates and effectors.

Patients' views of delayed fertility care during the COVID-19 pandemic as a conception catastrophe: the experience of U.S. FertilityIQ users.

BACKGROUND: Assessment of psychological reactions to delays in fertility treatment have often utilized single clinic samples during the time that fertility treatments were paused. We, therefore, assessed emotional reactions to treatment cancelations due to COVID-19 in infertility patients across the United States after treatments had begun to resume. STUDY DESIGN: Cross-sectional survey emailed on 27 May 2020 and closed on 30 June 2020, to 53,600 FertilityIQ.com website users inquiring about their experience since the COVID-19 pandemic. A subset of FertilityIQ users (n = 13,490) opened the survey invitation and 1806 respondents participated in the survey (13.4% response rate). RESULTS: The majority of respondents (female, 67.4%; male, 61.7%) were 31-40 years old; most were planning to start treatment immediately (women, 42.6%; men, 44.7%) or were undergoing treatment (women, 34.9%; men, 29.8%) at the time of treatment cancelation. Patients (women, 21.1%; men 19.1%) or clinics (women, 57.7%; men, 40.4%) canceled treatment. Most clinics had resumed treatment at the time of the study (women, 90.0%; men, 73.7%). Cancelation resulted in sadness (women, 83.9%; men 86.7%) and anger (women, 45.4%; men, 36.7%); greater than half of the participants whose treatment was canceled (women: 66.8%, n = 630; men: 73.7%, n = 14) agreed with cancelations. Greater than 70% of respondents were at least somewhat concerned about reproductive chances (women, 84.7%; men, 72.4%) and exclusion of partners (women, 73.3%; men, 72.4%). Distress/concern was associated with clinic cancelation, disagreement with delays, age, diagnosis, and concern about delays and pregnancy chances (p <.05). CONCLUSIONS: Respondents were distressed/concerned about the effect of the pandemic on their fertility. Distress was highest in women with a poorer fertility prognosis, no control over treatment cancelation, and high concern about the effect of treatment delay on pregnancy chances. Emotional support, education regarding treatment delay and fertility, and efforts where possible, to include patients in decisions to delay treatment are warranted in future treatment delays.

Dynamic equilibria in protein kinases.

Structural changes involved in protein kinase activation and ligand binding have been determined from a wealth of X-ray crystallographic evidence. Recent solution studies using NMR, EPR, HX-MS, and fluorescence techniques have deepened this understanding by highlighting the underlying energetics and dynamics of multistate conformational ensembles. This new research is showing how activation mechanisms and ligand binding alter the internal motions of kinases and enable allosteric coupling between distal regulatory regions and the active site.

Anxiety in the management of localised prostate cancer by active surveillance.

OBJECTIVES: To describe a range of anxieties in men on active surveillance (AS) for prostate cancer and determine which of these anxieties predicted health-related quality of life (HRQL). PATIENTS AND METHODS: In all, 260 men with prostate cancer on AS were invited to complete psychological measures including the Hospital and Anxiety Depression Scale; the State-Trait Anxiety Inventory-Trait Scale; the Memorial Anxiety Scale for Prostate Cancer; and the Functional Assessment of Cancer Therapy Scale-Prostate. Overall, 86 men with a mean (sd, range) age of 65.7 (5.4, 51-75) years returned data, yielding a response rate of 33%. Outcome measures were standardised psychological measures. Pearson's correlations were used to examine bivariate relationships, while regression analyses were used to describe predictors of dependent variables. RESULTS: When compared with the findings of prior research, the men in our cohort had normal levels of general anxiety and illness-specific anxiety and high prostate cancer-related HRQL. Age, trait anxiety and fear of recurrence (FoR) were significant predictors of prostate cancer-related HRQL; trait anxiety and FoR were significant predictors of total HRQL. Results should be interpreted in context of sample characteristics and the correlational design of the study. CONCLUSIONS: Participants reported low levels of anxiety and high HRQL. Trait anxiety and FoR were significant predictors of both prostate cancer-related and total HRQL. The administration of a short trait-anxiety screening tool may help identify men with clinically significant levels of anxiety and those at risk of reduced HRQL.

Cognitive function, health-related quality of life, and symptoms of depression and anxiety sensitivity are impaired in patients with the postural orthostatic tachycardia syndrome (POTS).

The Postural Orthostatic Tachycardia Syndrome (POTS) is a condition in which heart rate increases abnormally when the individual assumes an upright position. In addition to the marked tachycardia, presyncope, and syncope, patients with POTS often complain of light-headedness, fatigue, and difficulty in concentrating. The present study assessed individuals with POTS for psychiatric comorbidity, anxiety sensitivity and health related quality of life and examined general cognitive ability. Data was obtained from patients with POTS (n = 15, 12 female, aged 30 ± 3 years) and age matched healthy subjects (n = 30, 21 female, aged 32 ± 2 years). Patients with POTS commonly presented with symptoms of depression, elevated anxiety and increased anxiety sensitivity, particularly with regards to cardiac symptoms, and had a poorer health related quality of life in both the physical and mental health domains. While patients with POTS performed worse in tests of current intellectual functioning (verbal and non-verbal IQ) and in measures of focused attention (digits forward) and short term memory (digits back), test results were influenced largely by years of education and the underlying level of depression and anxiety. Acute changes in cognitive performance in response to head up tilt were evident in the POTS patients. From results obtained, it was concluded that participants with POTS have an increased prevalence of depression and higher levels of anxiety. These underlying symptoms impact on cognition in patients with POTS, particularly in the cognitive domains of attention and short-term memory. Our results indicate that psychological interventions may aid in recovery and facilitate uptake and adherence of other treatment modalities in patients with POTS.