An attention mechanism-based LSTM network for cancer kinase activity prediction

Despite the endeavours and achievements made in treating cancers during the past decades, resistance to available kinase drugs continues to be a major problem in cancer therapies. Thus, it is highly desirable to develop computational models that can predict the bioactivity of a compound against cancer kinases. Here, we present a Long Short-Term Memory (LSTM) framework for predicting the activities of lead molecules against seven different kinases. A total of 14,907 compounds from the ChEMBL database were selected for model building. Two different molecular representations, namely, 2D descriptors and MACCS fingerprints were subjected to the LSTM method for the training process. We also successfully integrated an attention mechanism into our model, which helped us to interpret the contribution of chemical features on kinase activity. The attention mechanism extracted the significant chemical moieties more effectively by taking them into consideration during the activity prediction. The recorded accuracies in the test sets for both 2D descriptors and MACCS fingerprints-based models were 0.81 and 0.78, respectively. The receiver operating characteristic curve (ROC)-area under the curve (AUC) score for both models was in the range of 0.8-0.99. The proposed framework can be a good starting point for the development of new cancer kinase drugs.

Utility of a 7- question online screener for DHA intake

The secondary analyses of two large, recently completed randomized clinical trials of DHA supplementation in pregnancy found that women with a low baseline DHA status benefited from randomization to a higher dose (800 vs 0 and 1000 vs 200 mg/day DHA). To obtain DHA status, it is necessary to obtain a blood sample and conduct an analysis using gas chromatography (GC) or GC-mass spectrometry (GCMS), both barriers to clinics where pregnant women receive advice on nutrition. Participants consuming less than 150 mg/day of DHA at baseline in our recent trial had a lower risk of early preterm birth and preterm birth when assigned to 1000 vs 200 m/day DHA. DHA intake was determined using a 7-question food frequency questionnaire administered by a trained nutritionist. Because the need for trained personnel to administer the questionnaire would be a barrier to implementing this finding in clinical management of pregnancy, the goal of this study was to determine if an online version of the questionnaire could be validly completed without assistance.

A Phytocannabinoid Formula Containing Palmitoylethanolamide, Acmella Oleracea Extract, and Bovine Colostrum Filtrate Inhibited Cytokine and Chemokine Production in Human Peripheral Blood Mononuclear Cells Stimulated Ex Vivo

A phytocannabinoid formula containing copaiba essential oil, palmitoylethanolamide, Sichuan pepper extract, Acmella oleracea extract, cruciferous vegetable extracts blend, and bovine colostrum filtrate was tested for its ex vivo effect on inhibiting stimulated cytokine release in human peripheral blood mononuclear cells. Effects of the phytocannabinoid formula on the peripheral blood mononuclear cells viability were measured by alamarBlue™ to confirm no obvious cytotoxicity. The phytocannabinoid formula was compared to reference compounds for its ability to inhibit both the phytohemagglutinin-stimulated release of the cytokines from T-cells and the lipopolysaccharide-stimulated release of the cytokines from macrophages/monocytes in human peripheral blood mononuclear cells. The T cell cytokine responses measured were those of interleukins (interleukin-1β, interleukin-2, interleukin-4, interleukin-5, interleukin-6, interleukin-8, interleukin-10, and interleukin-13) and interferon-gamma, interferon gamma-induced protein 10, macrophage inflammatory protein-1α, and tumor necrosis factor-α. The macrophage and monocyte cytokine responses measured were of interleukin-1β, interleukin-6, interleukin-8, macrophage inflammatory protein-1α, and tumor necrosis factor-α. For the phytohemagglutinin-stimulated T cell production of cytokines, the peripheral blood mononuclear cells treated with the Phytocannabinoid formula at 2 μg/mL and 20 μg/mL showed significant inhibition of interleukin-2, interleukin-6, interleukin-5, interleukin-8, interleukin-10, interleukin-13, and interferon gamma-induced protein 10 production. The phytocannabinoid formula also significantly reduced the macrophage/monocyte production of interleukin-1β and interleukin-6. Results from the present study showed that the phytocannabinoid formula inhibited cytokines and chemokines production in ex vivo stimulated human peripheral blood mononuclear cells, suggesting a potential immunomodulatory effect.

Ternary Alloys Enable Efficient Production of Methoxylated Chemicals via Selective Electrocatalytic Hydrogenation of Lignin Monomers

We explore the selective electrocatalytic hydrogenation of lignin monomers to methoxylated chemicals, of particular interest, when powered by renewable electricity. Prior studies, while advancing the field rapidly, have so far lacked the needed selectivity: when hydrogenating lignin-derived methoxylated monomers to methoxylated cyclohexanes, the desired methoxy group (-OCH₃) has also been reduced. The ternary PtRhAu electrocatalysts developed herein selectively hydrogenate lignin monomers to methoxylated cyclohexanes-molecules with uses in pharmaceutics. Using X-ray absorption spectroscopy and in situ Raman spectroscopy, we find that Rh and Au modulate the electronic structure of Pt and that this modulating steers intermediate energetics on the electrocatalyst surface to facilitate the hydrogenation of lignin monomers and suppress C-OCH₃ bond cleavage. As a result, PtRhAu electrocatalysts achieve a record 58% faradaic efficiency (FE) toward 2-methoxycyclohexanol from the lignin monomer guaiacol at 200 mA cm^-2, representing a 1.9× advance in FE and a 4× increase in partial current density compared to the highest productivity prior reports. We demonstrate an integrated lignin biorefinery where wood-derived lignin monomers are selectively hydrogenated and funneled to methoxylated 2-methoxy-4-propylcyclohexanol using PtRhAu electrocatalysts. This work offers an opportunity for the sustainable electrocatalytic synthesis of methoxylated pharmaceuticals from renewable biomass.

1552 'The Modified Cappuccini Audit Tool: Patient Safety in the Operating Theatre'

Aim

To assess our trust over a period of two weeks with regard to safe supervision of surgical and anaesthetic trainees within the operating theatre using the Cappuccini audit tool.

Method

Between 7th of December 2020 and 18th December 2020 the anaesthetic and surgical team were approached and if there was a trainee present, they were asked a set of questions as follows:

If the supervisor is present in theatre no further action would be taken. If the supervisor was not present the volunteer would attempt to contact the supervisor by the method given by the trainee, and they were asked several questions:

Results

146 trainees were identified during our period. Of these the supervisor was present in 93% of cases. Of those not present 90% were contactable by phone, the other 10% were contactable by the trainee themselves. All trainees were happy to answer the questions was explained and all consultants were happy to be contacted. The average time to attend when not present was 2 minutes and 36 seconds.

Conclusions

We have shown the Cappuccini audit tool to be effective in determining safe supervision and found it in general to be well received within surgery and anaesthetics. We found our trust to have safe supervision in the operating theatre.

423 Patient Perspective on the Use of the Independent Sector to Maintain Elective NHS Services During the COVID-19 Pandemic

Introduction

COVID-19 resulted in the suspension of planned treatments for patients worldwide leaving millions suffering the physical and mental effects of delay. Reports indicate that when services have been re-established, patients have been afraid to take up appointments. Hospitals put processes in place to counter this, notably separating emergency and elective patient cohorts. Most notably seen by the co-operation between the NHS and independent private healthcare providers in March 2020 at the height of the pandemic. We undertook a study to ascertain the perceptions of NHS patients who underwent elective treatment at independent ‘cold’ sites during the COVID-19 pandemic.

Method

A cross-sectional study with structured telephone interviews of patients who had planned elective treatments at ‘cold’ independent hospitals between March and September 2020. 1150 patients were identified, and a 20% sample formed a 230 patient study group, with 158 (70%) agreeing to participate.

Results

30% of patients delayed their treatment due to COVID related concerns, with 76% of these only accepting treatment because this was at a ‘cold’ site. 46% of patients perceived treatment at a ‘cold’ site as the most important factor contributing to their safety. 153 patients (97%) supported the paid arrangement between the NHS and the independent sector to provide separate ‘cold’ sites for elective treatments.

Discussion

Safely restarting elective services to allow important planned treatments to take place, as was the pandemic continues, is a priority. Our study indicates that physical separation of patient pathways impacted most on patient confidence, and that the use of ‘cold’ sites is a viable option.

3D-Printable Fluoropolymer Gas Diffusion Layers for CO2 Electroreduction

The electrosynthesis of value-added multicarbon products from CO₂ is a promising strategy to shift chemical production away from fossil fuels. Particularly important is the rational design of gas diffusion electrode (GDE) assemblies to react selectively, at scale, and at high rates. However, the understanding of the gas diffusion layer (GDL) in these assemblies is limited for the CO₂ reduction reaction (CO₂ RR): particularly important, but incompletely understood, is how the GDL modulates product distributions of catalysts operating in high current density regimes > 300 mA cm^-2 . Here, 3D-printable fluoropolymer GDLs with tunable microporosity and structure are reported and probe the effects of permeance, microstructural porosity, macrostructure, and surface morphology. Under a given choice of applied electrochemical potential and electrolyte, a 100× increase in the C₂ H₄ :CO ratio due to GDL surface morphology design over a homogeneously porous equivalent and a 1.8× increase in the C₂ H₄ partial current density due to a pyramidal macrostructure are observed. These findings offer routes to improve CO₂ RR GDEs as a platform for 3D catalyst design.