Search

Abstract Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 14 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC50) of 1.9 μM. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field.

Abstract Geometric deep learning has been revolutionizing the molecular modeling field. Despite the state-of-the-art neural network models are approaching ab initio accuracy for molecular property prediction, their applications, such as drug discovery and molecular dynamics (MD) simulation, have been hindered by insufficient utilization of geometric information and high computational costs. Here we propose an equivariant geometry-enhanced graph neural network called ViSNet, which elegantly extracts geometric features and efficiently models molecular structures with low computational costs. Our proposed ViSNet outperforms state-of-the-art approaches on multiple MD benchmarks, including MD17, revised MD17 and MD22, and achieves excellent chemical property prediction on QM9 and Molecule3D datasets. Furthermore, through a series of simulations and case studies, ViSNet can efficiently explore the conformational space and provide reasonable interpretability to map geometric representations to molecular structures.

During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (https://covid19forecasthub.org). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1-4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making.

With the increasing demand for deep oil and gas exploration and CCUS (Carbon Capture, Utilization, and Storage) engineering, improving rock-crushing efficiency stands as a pivotal technology. Ultrasonic vibration-assisted drilling has emerged as a novel rock-breaking technology. The high-frequency vibrations of ultrasonic waves impact rocks, inducing resonance and accelerating their fragmentation. At the same time, ultrasonic waves generate cavitation bubbles in the liquid near rock surfaces; the expansion and collapse of these bubbles further contribute to rock damage, thereby improving crushing efficiency. Therefore, investigating the dynamics and failure characteristics of cavitation bubbles near rock surfaces under ultrasonic influence is crucial for advancing ultrasonic-assisted rock-breaking technology. This study treats the liquid as compressible flow and investigates the movement and rupture of bubbles near rock surfaces under varying ultrasonic parameters, rock properties, characteristics of the circulating medium, and other relevant factors. The findings show that ultrasonic waves induce the oscillation, translation, collapse, and rebound of bubbles near rock surfaces. Higher ultrasonic frequencies correspond to larger collapse pressures and amplitudes near surrounding rocks, as well as longer expansion times and shorter collapse durations. In addition, bubble movement and collapse lead to rock material deformation, influenced by the rheological properties of the liquid medium. The study outcomes serve as a foundation for optimizing engineering parameters in ultrasonic-assisted rock breaking and provide theoretical support for the advancement of this technology.

Objectives: Network pharmacology is essential for understanding the multi-target and multi-pathway therapeutic mechanisms of traditional Chinese medicine. This study aims to evaluate the influence of database quality on target identification and to explore the therapeutic potential of rhynchophylline (Rhy) in treating overactive bladder (OAB). Methods: An OAB dataset was constructed through extensive literature screening. Using this dataset, we applied network pharmacology to predict potential targets for Rhy, which is known for its therapeutic effects but lacks a well-defined target profile. Predicted targets were validated through in vitro experiments, including DARTS and CETSA. Results: Our analysis identified Rhy as a potential modulator of the M3 receptor and TRPM8 channel in the treatment of OAB. Validation experiments confirmed the interaction between Rhy and these targets. Additionally, the GeneCards database predicted other targets that are not directly linked to OAB, corroborated by the literature. Conclusions: We established a more accurate and comprehensive dataset of OAB targets, enhancing the reliability of target identification for drug treatments. This study underscores the importance of database quality in network pharmacology and contributes to the potential therapeutic strategies for OAB. [ABSTRACT FROM AUTHOR]

Recombinase-aided polymerase chain reaction (RAP) is a sensitive, single-tube, two-stage nucleic acid amplification method. This study aimed to develop an assay that can be used for the early diagnosis of three types of bacteremia caused by Staphylococcus aureus (SA), Pseudomonas aeruginosa (PA), and Acinetobacter baumannii (AB) in the bloodstream based on recombinant human mannan-binding lectin protein (M1 protein)-conjugated magnetic bead (M1 bead) enrichment of pathogens combined with RAP. Recombinant plasmids were used to evaluate the assay sensitivity. Common blood influenza bacteria were used for the specific detection. Simulated and clinical plasma samples were enriched with M1 beads and then subjected to multiple recombinase-aided PCR (M-RAP) and quantitative PCR (qPCR) assays. Kappa analysis was used to evaluate the consistency between the two assays. The M-RAP method had sensitivity rates of 1, 10, and 1 copies/μL for the detection of SA, PA, and AB plasmids, respectively, without cross-reaction to other bacterial species. The M-RAP assay obtained results for < 10 CFU/mL pathogens in the blood within 4 h, with higher sensitivity than qPCR. M-RAP and qPCR for SA, PA, and AB yielded Kappa values of 0.839, 0.815, and 0.856, respectively (P < 0.05). An M-RAP assay for SA, PA, and AB in blood samples utilizing M1 bead enrichment has been developed and can be potentially used for the early detection of bacteremia. [ABSTRACT FROM AUTHOR]

Real-time intelligent lithology identification while drilling is vital to realizing downhole closed-loop drilling. The complex and changeable geological environment in the drilling makes lithology identification face many challenges. This paper studies the problems of difficult feature information extraction, low precision of thin-layer identification and limited applicability of the model in intelligent lithologic identification. The author tries to improve the comprehensive performance of the lithology identification model from three aspects: data feature extraction, class balance, and model design. A new real-time intelligent lithology identification model of dynamic felling strategy weighted random forest algorithm (DFW-RF) is proposed. According to the feature selection results, gamma ray and 2 MHz phase resistivity are the logging while drilling (LWD) parameters that significantly influence lithology identification. The comprehensive performance of the DFW-RF lithology identification model has been verified in the application of 3 wells in different areas. By comparing the prediction results of five typical lithology identification algorithms, the DFW-RF model has a higher lithology identification accuracy rate and F1 score. This model improves the identification accuracy of thin-layer lithology and is effective and feasible in different geological environments. The DFW-RF model plays a truly efficient role in the realtime intelligent identification of lithologic information in closed-loop drilling and has greater applicability, which is worthy of being widely used in logging interpretation. [ABSTRACT FROM AUTHOR]

The whole-process people's democracy empowered by digital technology is a new form of democracy that conforms to the trend of the times. It has rich internal logic, mainly including the value logic of exploring the democratic field, democratic mechanism and democratic political literacy, the institutional logic of exploring the digital democratic system and its operation mechanism, the technological logic of exploring big data, artificial intelligence and blockchain technology. However, in the process of empowering people's democracy through digital technology, there are also difficulties such as the increasing imbalance of democratic development, the increasing instability of democratic operation, and the loss of the authenticity of democratic development. To alleviate the above difficulties, effective measures should be taken from various aspects, including realizing the equality of multiple participants, strengthening the mechanism construction in the process of democratic operation, cultivating Internet users' political literacy and awareness, and creating a healthy and orderly democratic operation environment, so as to realize the sound operation of whole-process people's democracy under the digital technology. [ABSTRACT FROM AUTHOR]

This document is a correction notice for an article titled "Ammonia promotes the proliferation of bone marrow-derived mesenchymal stem cells by regulating the Akt/mTOR/S6k pathway" published in Bone Research. The authors identified mistakes in Figures 1c and 5d after the article was published online. The revisions made to these figures do not affect the conclusions of the study. The corrected figures and their accompanying legends are provided in the document. The original article has been updated. [Extracted from the article]

Generative drug design facilitates the creation of compounds effective against pathogenic target proteins. This opens up the potential to discover novel compounds within the vast chemical space and fosters the development of innovative therapeutic strategies. However, the practicality of generated molecules is often limited, as many designs focus on a narrow set of drug-related properties, failing to improve the success rate of subsequent drug discovery process. To overcome these challenges, we develop TamGen, a method that employs a GPT-like chemical language model and enables target-aware molecule generation and compound refinement. We demonstrate that the compounds generated by TamGen have improved molecular quality and viability. Additionally, we have integrated TamGen into a drug discovery pipeline and identified 14 compounds showing compelling inhibitory activity against the Tuberculosis ClpP protease, with the most effective compound exhibiting a half maximal inhibitory concentration (IC 50 ) of 1.9 μM. Our findings underscore the practical potential and real-world applicability of generative drug design approaches, paving the way for future advancements in the field., (© 2024. The Author(s).)

During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (https://covid19forecasthub.org). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1-4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making., Competing Interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: APP report grants from Metabiota Inc outside the submitted work. J.S. and Columbia University declare partial ownership of SK Analytics. No other authors have competing interests to declare., (Copyright: This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.)