Search

Abstract Generative models are undergoing rapid research and application to de novo drug design. To facilitate their application and evaluation, we present MolScore. MolScore already contains many drug-design-relevant scoring functions commonly used in benchmarks such as, molecular similarity, molecular docking, predictive models, synthesizability, and more. In addition, providing performance metrics to evaluate generative model performance based on the chemistry generated. With this unification of functionality, MolScore re-implements commonly used benchmarks in the field (such as GuacaMol, MOSES, and MolOpt). Moreover, new benchmarks can be created trivially. We demonstrate this by testing a chemical language model with reinforcement learning on three new tasks of increasing complexity related to the design of 5-HT2a ligands that utilise either molecular descriptors, 266 pre-trained QSAR models, or dual molecular docking. Lastly, MolScore can be integrated into an existing Python script with just three lines of code. This framework is a step towards unifying generative model application and evaluation as applied to drug design for both practitioners and researchers. The framework can be found on GitHub and downloaded directly from the Python Package Index. Scientific Contribution MolScore is an open-source platform to facilitate generative molecular design and evaluation thereof for application in drug design. This platform takes important steps towards unifying existing benchmarks, providing a platform to share new benchmarks, and improves customisation, flexibility and usability for practitioners over existing solutions. Graphical Abstract

Upper limb amputations are frequently the result of traumatic events, often associated with car accidents or incidents involving industrial machinery. The development of prosthetic technologies is essential to provide physical assistance to amputees and offer psychological support. In such circumstances, cosmetic gloves can be utilized to improve personal and social comfort by enhancing visual characteristics similar to those of a biological hand. In this work, we developed formulations based on natural rubber latex (NRL) with antibacterial properties by the incorporation of an additive consisting of zinc oxide coated with silver nanoparticles to manufacture cosmetic gloves designed for upper limb prostheses. The dispersion efficiency of the additive was investigated by microscopic techniques. Tensile and tear strength tests, Shore A hardness measurements, swelling indices, hydrophobicity and inhibition of bacterial growth were also conducted to ascertain the suitability and adaptability of the developed material for integration into the daily activities of patients. Moreover, processing parameters for the immersion molding production of cosmetic gloves were assessed and validated, highlighting the crucial relationship between manufacturing techniques and material properties. Overall, the results indicated that the combination of the additive with a high crosslinking density of NRL effectively improved the material mechanical properties, resulting in a tensile strength of up to 20.7 MPa and a tear strength of approximately 40 kN m−1. In addition, and according to the antibacterial tests, the gloves exhibited bacteriostatic effects against Staphylococcus aureus and bactericidal effects against Escherichia coli. As a conclusion, the dataset indicates that the manufactured cosmetic gloves can be used to covering upper limb prostheses. Additionally, it underscores the considerable potential for the cosmetic glove manufacturing industry through the use of appropriate formulations and processing techniques for NRL.

This study investigates the effect of extrusion screw speed and carbon nanotube (CNT) concentration on the thermal, mechanical, and electromagnetic interference shielding effectiveness (EMI SE) properties of Polycarbonate (PC)/acrylonitrile-butadiene-styrene (ABS) and its polymer nanocomposites (PNCs) by means of design of experiments (DoE) approach. A masterbatch method was employed to obtain the best dispersion of the CNTs throughout the polymer matrix. This study evaluates the thermo-mechanical characterisation of the polymers and PNCs at varying screw speeds to assess filler matrix bonding. The results highlight that CNT concentration has a significant effect on all mechanical properties, while screw speed only affects the Charpy impact strength and flexural properties of the samples. Compounding at 200 rpm has the best flexural and tensile strength, which is attributed to the best filler matrix bonding (highest storage modulus) of the PNCs. The best EMI SE results were obtained at 10 wt.% CNTs. This research contributes valuable insights into the effect of CNT concentration and extrusion screw speed on the mechanical, thermal and EMI SE properties of PC/ABS and its PNCs.

Additive manufacturing (AM) allows for the creation of novel complex structures previously impossible using traditional subtractive methods; however, there is a need for new design approaches to fully exploit this potential. Metal lattice structures have attracted attention for their broad customisable range of configurations and potential properties, and to take advantage of these possibilities there is a need for intelligent design tools to optimise the lattice according to the desired application and properties. This study presents and demonstrates a method to automatically design lattice structures which achieve a desired compressive stress–strain curve and satisfy user-defined manufacturing limits. A genetic algorithm iterates selected design variables (unit cell aspect ratio, strut taper, and thickness gradient) and minimises the error between the target stress–strain curve and predicted curve which is determined via automated finite element (FE) modelling of each lattice design. The optimised design is validated through manufacture and compression testing of 17-4PH stainless steel lattice structures, and micro-CT imaging to assess build quality. The tool's versatility is demonstrated by successfully designing lattices to fit a range of target curves. This work demonstrates the potential of this smart inverse design tool to exploit AM and optimise lattice design based on desired performance. The custom written design tool presented in this study is available open source at https://github.com/I-Form/LiDOT.

Abstract The Manawatu region experiences deep tremor and long‐term slow slip events (SSEs); however, tremor is adjacent to, and not co‐located with, long‐term SSEs. Observations of episodic tremor and slip (ETS) elsewhere suggest it is possible smaller short‐term SSEs below the current detection threshold occur where tremor is observed. Therefore, we sought to determine if small SSEs occurred with Manawatu tremor. We decomposed GNSS data using times of tremor to assess average surface displacements and performed a static slip inversion to model the displacement during tremor. The slip inversion suggested small slow slip partially coincided with tremor and long‐term SSEs may influence these small SSEs by increasing slip rates. We suggest that the interface below deep long‐term SSEs may slip often, in small ETS‐like SSEs that are not individually detectable geodetically. The question remains as to the nature of the strong variability in SSE behavior with depth and duration in the southern Hikurangi margin.

Abstract A plethora of AI-based techniques now exists to conduct de novo molecule generation that can devise molecules conditioned towards a particular endpoint in the context of drug design. One popular approach is using reinforcement learning to update a recurrent neural network or language-based de novo molecule generator. However, reinforcement learning can be inefficient, sometimes requiring up to 105 molecules to be sampled to optimize more complex objectives, which poses a limitation when using computationally expensive scoring functions like docking or computer-aided synthesis planning models. In this work, we propose a reinforcement learning strategy called Augmented Hill-Climb based on a simple, hypothesis-driven hybrid between REINVENT and Hill-Climb that improves sample-efficiency by addressing the limitations of both currently used strategies. We compare its ability to optimize several docking tasks with REINVENT and benchmark this strategy against other commonly used reinforcement learning strategies including REINFORCE, REINVENT (version 1 and 2), Hill-Climb and best agent reminder. We find that optimization ability is improved ~ 1.5-fold and sample-efficiency is improved ~ 45-fold compared to REINVENT while still delivering appealing chemistry as output. Diversity filters were used, and their parameters were tuned to overcome observed failure modes that take advantage of certain diversity filter configurations. We find that Augmented Hill-Climb outperforms the other reinforcement learning strategies used on six tasks, especially in the early stages of training or for more difficult objectives. Lastly, we show improved performance not only on recurrent neural networks but also on a reinforcement learning stabilized transformer architecture. Overall, we show that Augmented Hill-Climb improves sample-efficiency for language-based de novo molecule generation conditioning via reinforcement learning, compared to the current state-of-the-art. This makes more computationally expensive scoring functions, such as docking, more accessible on a relevant timescale.

Introduction HIV infection is an established risk factor for the development of cardiovascular diseases. Although increasing evidence implicates a higher prevalence of myocardial fibrosis (MF) among patients living with HIV (PLWH) compared with the HIV-negative population, there is a paucity of knowledge regarding its determinants and factors associated with higher odds of MF development. We aim to perform a systematic review to estimate the prevalence of MF among PLWH. Additionally, we will determine the factors associated with higher odds of MF among PLWH compared with the HIV-negative population.Methods A systematic review will be performed by consulting the Cochrane handbook for systematic reviews of interventional studies reporting a confirmed diagnosis of MF among PLWH. Articles will be eligible if they provide the prevalence of MF among PLWH and HIV-negative populations or the odds ratio (OR) and 95% confidence interval (CI) of MF development in relation to HIV. Depending on the quality of the data and the heterogeneity among the included studies, a random-effects or fixed-effects model will be used to pool and compare the ORs of MF among PLWH and HIV-negative population. Factors associated with higher odds of MF in relation to HIV will also be determined.Ethics and dissemination Ethical approval and obtaining informed consent are not required for this systematic review as it does not use individual patients’ data. Results of this study will be published in a peer-reviewed medical journal.

ABSTRACT Nontuberculous mycobacteria (NTM), including Mycobacterium avium, are clinically important pathogens in cystic fibrosis (CF). The innate immune response to M. avium remains incompletely understood. We evaluated the role of complement opsonization in neutrophil-mediated killing of M. avium. Killing assays were performed using neutrophils from healthy donors (HDs) and persons with CF (pwCF). Clinical isolates of M. avium were opsonized with plasma from HDs or pwCF, which was intact or heat-treated to inactivate complement. HD neutrophils had killing activity against M. avium opsonized with intact HD plasma and killing was significantly reduced when M. avium was opsonized with heat-inactivated HD plasma. When opsonized with HD plasma, CF neutrophils had killing activity against M. avium that was not different than HD neutrophils. When opsonized with intact plasma from pwCF, HD neutrophil killing of M. avium was significantly reduced. Opsonization of M. avium with C3-depleted serum or IgM-depleted plasma resulted in significantly reduced killing. Plasma C3 levels were elevated in pwCF with NTM infection compared to pwCF without NTM infection. These studies demonstrate that human neutrophils efficiently kill M. avium when opsonized in the presence of plasma factors from HD that include C3 and IgM. Killing efficiency is significantly lower when the bacteria are opsonized with plasma from pwCF. This indicates a novel role for opsonization in neutrophil killing of M. avium and a deficiency in complement opsonization as a mechanism of impaired M. avium killing in CF. IMPORTANCE Mycobacterium avium is a member of a group of bacterial species termed nontuberculous mycobacteria (NTM) that cause lung disease in certain populations, including persons with cystic fibrosis (CF). NTM infections are challenging to diagnose and can be even more difficult to treat. This study investigated how the immune system responds to M. avium infection in CF. We found that neutrophils, the most abundant immune cell in the lungs in CF, can effectively kill M. avium in individuals both with and without CF. Another component of the immune response called the complement system is also required for this process. Levels of complement proteins are altered in persons with CF who have a history of NTM compared to those without a history of NTM infection. These results add to our understanding of how the immune system responds to M. avium, which can help pave the way toward better diagnostic and treatment strategies.