Fine-tuning large language models for chemical text mining.

Extracting knowledge from complex and diverse chemical texts is a pivotal task for both experimental and computational chemists. The task is still considered to be extremely challenging due to the complexity of the chemical language and scientific literature. This study explored the power of fine-tuned large language models (LLMs) on five intricate chemical text mining tasks: compound entity recognition, reaction role labelling, metal-organic framework (MOF) synthesis information extraction, nuclear magnetic resonance spectroscopy (NMR) data extraction, and the conversion of reaction paragraphs to action sequences. The fine-tuned LLMs demonstrated impressive performance, significantly reducing the need for repetitive and extensive prompt engineering experiments. For comparison, we guided ChatGPT (GPT-3.5-turbo) and GPT-4 with prompt engineering and fine-tuned GPT-3.5-turbo as well as other open-source LLMs such as Mistral, Llama3, Llama2, T5, and BART. The results showed that the fine-tuned ChatGPT models excelled in all tasks. They achieved exact accuracy levels ranging from 69% to 95% on these tasks with minimal annotated data. They even outperformed those task-adaptive pre-training and fine-tuning models that were based on a significantly larger amount of in-domain data. Notably, fine-tuned Mistral and Llama3 show competitive abilities. Given their versatility, robustness, and low-code capability, leveraging fine-tuned LLMs as flexible and effective toolkits for automated data acquisition could revolutionize chemical knowledge extraction.

Transfer learning enhanced graph neural network for aldehyde oxidase metabolism prediction and its experimental application.

Aldehyde oxidase (AOX) is a molybdoenzyme that is primarily expressed in the liver and is involved in the metabolism of drugs and other xenobiotics. AOX-mediated metabolism can result in unexpected outcomes, such as the production of toxic metabolites and high metabolic clearance, which can lead to the clinical failure of novel therapeutic agents. Computational models can assist medicinal chemists in rapidly evaluating the AOX metabolic risk of compounds during the early phases of drug discovery and provide valuable clues for manipulating AOX-mediated metabolism liability. In this study, we developed a novel graph neural network called AOMP for predicting AOX-mediated metabolism. AOMP integrated the tasks of metabolic substrate/non-substrate classification and metabolic site prediction, while utilizing transfer learning from 13C nuclear magnetic resonance data to enhance its performance on both tasks. AOMP significantly outperformed the benchmark methods in both cross-validation and external testing. Using AOMP, we systematically assessed the AOX-mediated metabolism of common fragments in kinase inhibitors and successfully identified four new scaffolds with AOX metabolism liability, which were validated through in vitro experiments. Furthermore, for the convenience of the community, we established the first online service for AOX metabolism prediction based on AOMP, which is freely available at https://aomp.alphama.com.cn.

LogD7.4 prediction enhanced by transferring knowledge from chromatographic retention time, microscopic pKa and logP.

Lipophilicity is a fundamental physical property that significantly affects various aspects of drug behavior, including solubility, permeability, metabolism, distribution, protein binding, and toxicity. Accurate prediction of lipophilicity, measured by the logD7.4 value (the distribution coefficient between n-octanol and buffer at physiological pH 7.4), is crucial for successful drug discovery and design. However, the limited availability of data for logD modeling poses a significant challenge to achieving satisfactory generalization capability. To address this challenge, we have developed a novel logD7.4 prediction model called RTlogD, which leverages knowledge from multiple sources. RTlogD combines pre-training on a chromatographic retention time (RT) dataset since the RT is influenced by lipophilicity. Additionally, microscopic pKa values are incorporated as atomic features, providing valuable insights into ionizable sites and ionization capacity. Furthermore, logP is integrated as an auxiliary task within a multitask learning framework. We conducted ablation studies and presented a detailed analysis, showcasing the effectiveness and interpretability of RT, pKa, and logP in the RTlogD model. Notably, our RTlogD model demonstrated superior performance compared to commonly used algorithms and prediction tools. These results underscore the potential of the RTlogD model to improve the accuracy and generalization of logD prediction in drug discovery and design. In summary, the RTlogD model addresses the challenge of limited data availability in logD modeling by leveraging knowledge from RT, microscopic pKa, and logP. Incorporating these factors enhances the predictive capabilities of our model, and it holds promise for real-world applications in drug discovery and design scenarios.

Enable Multi-Stimuli-Responsive Biomimetic Actuation with Asymmetric Design of Graphene-Conjugated Conductive Polymer Gradient Film.

In this paper, multiresponsive actuators based on asymmetric design of graphene-conjugated poly(3,4-ethylene dioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) gradient films have been developed by a simple drop casting method. The biomimetic actuation is attributed to the hygroscopic expansion property of PEDOT:PSS and the gradient distribution of graphene sheets within the film, which resembles the hierarchical swelling tissues of some plants in nature. Graphene-conjugated PEDOT:PSS (GCP) actuators exhibit reversible bending behavior under multistimuli such as moisture, organic vapor, electrothermal, and photothermal heating. Noticeably, the bending curvature reaches 2.15 cm-1 under applied voltage as low as 1.5 V owing to the high electrical conductivity of GCP actuator. To mimic the motions of nyctinastic plants, a GCP artificial flower that spreads its petals under sunlight illumination has been fabricated. GCP actuators have been also demonstrated as intelligent light-controlled switches for light-emitting diodes and smart curtains for thermal management. Not only do the GCP gradient films exhibit potential applications in flexible electronics and energy harvesting/storage devices but also the facile fabrication of multiresponsive GCP actuators may shed light on the development of soft robotics, artificial muscles, wearable electronics, and smart sensors.

MolFilterGAN: a progressively augmented generative adversarial network for triaging AI-designed molecules.

Artificial intelligence (AI)-based molecular design methods, especially deep generative models for generating novel molecule structures, have gratified our imagination to explore unknown chemical space without relying on brute-force exploration. However, whether designed by AI or human experts, the molecules need to be accessibly synthesized and biologically evaluated, and the trial-and-error process remains a resources-intensive endeavor. Therefore, AI-based drug design methods face a major challenge of how to prioritize the molecular structures with potential for subsequent drug development. This study indicates that common filtering approaches based on traditional screening metrics fail to differentiate AI-designed molecules. To address this issue, we propose a novel molecular filtering method, MolFilterGAN, based on a progressively augmented generative adversarial network. Comparative analysis shows that MolFilterGAN outperforms conventional screening approaches based on drug-likeness or synthetic ability metrics. Retrospective analysis of AI-designed discoidin domain receptor 1 (DDR1) inhibitors shows that MolFilterGAN significantly increases the efficiency of molecular triaging. Further evaluation of MolFilterGAN on eight external ligand sets suggests that MolFilterGAN is useful in triaging or enriching bioactive compounds across a wide range of target types. These results highlighted the importance of MolFilterGAN in evaluating molecules integrally and further accelerating molecular discovery especially combined with advanced AI generative models.

Multi-instance learning of graph neural networks for aqueous pKa prediction.

MOTIVATION: The acid dissociation constant (pKa) is a critical parameter to reflect the ionization ability of chemical compounds and is widely applied in a variety of industries. However, the experimental determination of pKa is intricate and time-consuming, especially for the exact determination of micro-pKa information at the atomic level. Hence, a fast and accurate prediction of pKa values of chemical compounds is of broad interest. RESULTS: Here, we compiled a large-scale pKa dataset containing 16 595 compounds with 17 489 pKa values. Based on this dataset, a novel pKa prediction model, named Graph-pKa, was established using graph neural networks. Graph-pKa performed well on the prediction of macro-pKa values, with a mean absolute error around 0.55 and a coefficient of determination around 0.92 on the test dataset. Furthermore, combining multi-instance learning, Graph-pKa was also able to automatically deconvolute the predicted macro-pKa into discrete micro-pKa values. AVAILABILITY AND IMPLEMENTATION: The Graph-pKa model is now freely accessible via a web-based interface (https://pka.simm.ac.cn/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Structure-Based Virtual Screening and Identification of Potential Inhibitors of SARS-CoV-2 S-RBD and ACE2 Interaction.

The emergence and rapid spread of SARS-CoV-2 have caused a worldwide public health crisis. Designing small molecule inhibitors targeting SARS-CoV-2 S-RBD/ACE2 interaction is considered as a potential strategy for the prevention and treatment of SARS-CoV-2. But to date, only a few compounds have been reported as SARS-CoV-2 S-RBD/ACE2 interaction inhibitors. In this study, we described the virtual screening and experimental validation of two novel inhibitors (DC-RA016 and DC-RA052) against SARS-CoV-2 S-RBD/ACE2 interaction. The NanoBiT assays and surface plasmon resonance (SPR) assays demonstrated their capabilities of blocking SARS-CoV-2 S-RBD/ACE2 interaction and directly binding to both S-RBD and ACE2. Moreover, the pseudovirus assay revealed that these two compounds possessed significant antiviral activity (about 50% inhibition rate at maximum non-cytotoxic concentration). These results indicate that the compounds DC-RA016 and DC-RA052 are promising inhibitors against SARS-CoV-2 S-RBD/ACE2 interaction and deserve to be further developed.

Graph neural networks for automated de novo drug design.

The goal of de novo drug design is to create novel chemical entities with desired biological activities and pharmacokinetics (PK) properties. Over recent years, with the development of artificial intelligence (AI) technologies, data-driven methods have rapidly gained in popularity in this field. Among them, graph neural networks (GNNs), a type of neural network directly operating on the graph structure data, have received extensive attention. In this review, we introduce the applications of GNNs in de novo drug design from three aspects: molecule scoring, molecule generation and optimization, and synthesis planning. Furthermore, we also discuss the current challenges and future directions of GNNs in de novo drug design.

Construction land suitability assessment in rapid urbanizing cities for promoting the implementation of United Nations sustainable development goals: a case study of Nanchang, China.

Cities, the main place of human settlements, are required to offer high-quality environments to citizens. To achieve this, it is essential to overcome several mega challenges of urbanization, population growth, economic development, environmental deterioration, and climate change. Urban infrastructure construction is capable of enhancing economic growth and promoting urban sustainability, while it will lead to many environmental problems if the infrastructure construction is not properly planned and designed. To address this challenge, this study aims to understand how to ensure the construction land expansion sustainably in rapidly urbanizing cities. In particular, this study analyzed the suitability of construction land expansion in Nanchang, a rapid urbanizing city in China, from 1995 to 2015. The results indicate that the urban expansion speed from 1995 to 2005 was faster than that from 2005 to 2015. The construction land in Nanchang was expanding towards "all directions" and sprawled towards surrounding districts and counties from the original core areas. Nevertheless, about 70% of the Nanchang area was allowable construction area (highly suitable expansion, relatively suitable expansion, and basically suitable expansion areas), indicating that the abundant reserved land resources for urban construction. This study also identified multiple suitability expansion paths of construction land, providing a scientific guidance for the land use planning of Nanchang city. Overall, this study provides a reference to the understanding of the construction land expansion for the achievement of United Nations sustainable development goals. It can also promote the understanding of spatial territory planning and practically enhance the capabilities of land use planning and design.

Cyanine-based NIR fluorescent probe for monitoring H2S and imaging in living cells and in vivo.

As a critical gaseous signaling molecule, H2S is involved in various biological processes. To deeper study the physiological and pathological roles of H2S, convenient and efficient detection techniques for endogenous H2S in vivo are still in urgent demand. Herein, we reported a new turn-on Near-infrared (NIR) fluorescence probe NIR-H2S based on thiolysis reactions for detection of H2S. The probe possessed many excellent properties including high sensitivity and selectivity, good cell-membrane permeability, and low cytotoxicity. In vitro, NIR-H2S showed a 58-fold fluorescence enhancement when reacted with H2S in a buffer and displayed a good linear relationship (r = 0.9925) in a rather wide concentration range of H2S (0-500 µM). Furthermore, NIR-H2S was successfully employed in monitoring endogenous H2S induced by D-Cys in living cancer cells and mice. These results indicated that NIR-H2S had great potentiality in detecting cellular H2S in living animals and being applied to cancer diagnosis.