Accurate short-term forecasting of blood glucose is essential for safe diabetes self-management, yet existing forecasters seldom leverage nutrients estimated from free-text meal logs. This estimating nutrient is especially challenging for Chinese meals due to mixed dishes and diverse preparation styles. In this study, we investigate whether large language model (LLM)-based nutrient estimation from Chinese diet logs can improve dynamic glucose forecasting. Five Chinese-capable LLMs for Chinese texts (GPT5-mini, DeepSeekv3.1, Llama3.170B, GLM4.5, Qwen3-235B) are prompted to estimate macronutrients and weighted GI for meals. We also compare two pipelines for nutrition estimation: database-first with LLM fallback and LLM-only with retrieval-augmented generation (RAG). We propose a seasonal-patch Transformer architecture designed for multi-source feature fusion, integrating dynamic sequences, static patient data, diet word embeddings, and the LLM-estimated nutritional profiles. Experiments on 112 realworld Chinese patients (ShanghaiT1DM$\mathrm{n}=12$, ShanghaiT2DM$\mathbf{n}=\mathbf{1 0 0}$) showed that, versus a no-LLM baseline, LLMaugmented models reduced 3-h postprandial RMSE by up to 1.2$\mathbf{m g} / \mathbf{d L}(\mathbf{3 6. 0 6} \rightarrow \mathbf{3 4. 8 6 ~ m g} / \mathbf{d L})$and achieved the best full-period RMSE of$26.66 \text{mg} / \text{dL}$, comparable to a recent large pretraining forecasting model. Forecast gain correlates strongest with carbohydrate$(\rho=0.94)$and fat$(\rho=0.77)$estimation accuracy. These results demonstrate that LLM-based nutritional feature engineering provides a critical pathway toward more precise, personalized CGM forecasting and the realization of clinical digital twins.

Kolmogorov-Arnold Networks (KAN) is an emerging neural network architecture in machine learning. It has greatly interested the research community about whether KAN can be a promising alternative to the commonly used Multi-Layer Perceptions (MLP). Experiments in various fields demonstrated that KAN-based machine learning can achieve comparable if not better performance than MLP-based methods, but with much smaller parameter scales and are more explainable. In this paper, we explore the incorporation of KAN into the actor and critic networks for offline reinforcement learning (RL). We evaluated the performance, parameter scales, and training efficiency of various KAN and MLP-based conservative Q-learning (CQL) on the classical D4RL benchmark for offline RL. Our study demonstrates that KAN can achieve performance close to the commonly used MLP with significantly fewer parameters. This allows us to choose the base networks according to the offline RL task requirements.

Offline Reinforcement Learning (RL) optimizes policy using pre-collected data instead of direct environment interaction, offering a safe and cost-effective solution for sequential decision-making in the real world. However, it faces challenges such as distribution shift issues and vulnerability under perturbations. Researchers have developed various conservative methods to improve the robustness of offline RL. Nevertheless, model-based methods can result in transition distribution shift issues, while model-free value-based uncertainty penalty methods may not be sufficiently robust. To address these problems, we propose a new method called Robust Offline RL via Conservative Smoothing and Dynamics Controlling (RCSD). To achieve reliable value estimation of out-of-distribution (OOD) actions, RCSD uses both model-free uncertainty penalty and model-based simulation methods. It introduces a new one-step simulation method with conservative dynamics controlling to avoid value overestimation caused by transition distribution shifts. Moreover, RCSD considers both current and next states when generating OOD states to ensure cautious value estimation and efficient data utilization. RCSD uses conservative Q-smoothing and policy smoothing to strengthen the policy against sudden changes under perturbations. Experiments on D4RL benchmark demonstrate that RCSD can achieve state-of-the-art performance compared to baselines in either benchmark or adversarial attack tests.

Drug resistance in Mycobacterium tuberculosis (Mtb) is a significant challenge in the control and treatment of tuberculosis, making efforts to combat the spread of this global health burden more difficult. To accelerate anti-tuberculosis drug discovery, repurposing clinically approved or investigational drugs for the treatment of tuberculosis by computational methods has become an attractive strategy. In this study, we developed a virtual screening workflow that combines multiple machine learning and deep learning models, and 11 576 compounds extracted from the DrugBank database were screened against Mtb. Our screening method produced satisfactory predictions on three data-splitting settings, with the top predicted bioactive compounds all known antibacterial or anti-TB drugs. To further identify and evaluate drugs with repurposing potential in TB therapy, 15 screened potential compounds were selected for subsequent computational and experimental evaluations, out of which aldoxorubicin and quarfloxin showed potent inhibition of Mtb strain H37Rv, with minimal inhibitory concentrations of 4.16 and 20.67 μM/mL, respectively. More inspiringly, these two compounds also showed antibacterial activity against multidrug-resistant TB isolates and exhibited strong antimicrobial activity against Mtb. Furthermore, molecular docking, molecular dynamics simulation, and the surface plasmon resonance experiments validated the direct binding of the two compounds to Mtb DNA gyrase. In summary, our effective comprehensive virtual screening workflow successfully repurposed two novel drugs (aldoxorubicin and quarfloxin) as promising anti-Mtb candidates. The verification results provide useful information for the further development and clinical verification of anti-TB drugs.

Large language models have shown promise in biomedical natural language processing, yet their use in extracting structured knowledge from medical scales remains limited. This study introduces MedScaleRE-PF, a novel prompting framework designed for relation extraction in Chinese medical scale texts. The framework combines few-shot in-context learning with retrieval-augmented generation, chain-of-thought prompting, and self-verification strategies to improve contextual understanding and factual consistency. We constructed the CMedS-RE dataset, consisting of 606 full-text articles with 19,051 sentences, 29,359 annotated entities, and 7217 relation instances. Experiments were conducted on two tasks: relational triple extraction (RTE) and relation classification (RC). We evaluated both single-step and multi-step prompting, along with four self-verification strategies: direct (D-SV), stepwise (S-CoT-SV), relation-specific (R-CoT-SV), and stepwise relation-specific (SR-CoT-SV). The best results were achieved with single-step prompting and the R-CoT-SV strategy, yielding F1 scores of 42.58 % for RTE under the 32-shot setting and 65.42 % for RC under the 8-shot setting. Compared to a RAG-only baseline, this configuration improved F1 by 7.59 % on RTE and 1.07 % on RC. Additional experiments demonstrated strong performance under annotation-scarce conditions, achieving 46.99 % F1 on RTE with 20 training articles and 59.87 % on RC with 50 articles. Ablation and error analyses further confirmed that task-specific prompt structure and verification design significantly impact performance under few-shot conditions. MedScaleRE-PF also showed consistent results across multiple LLMs, confirming its stability and generalizability. These findings highlight the effectiveness of combining simple prompting and CoT-inspired verification in domain-specific information extraction. MedScaleRE-PF offers a flexible and structured approach for mining medical scale knowledge and supports prompt-based development in biomedical applications.

BACKGROUND: Stroke is one of the leading causes of death and disability worldwide. The National Institutes of Health Stroke Scale (NIHSS) scores in electronic health records (EHRs), which quantitatively describe patients' neurological deficits in evidence-based treatment, are crucial in stroke-related clinical investigations. However, the free-text format and lack of standardization inhibit their effective use. Automatically extracting the scale scores from the clinical free text so that its potential value in real-world studies is realized has become an important goal. OBJECTIVE: This study aims to develop an automated method to extract scale scores from the free text of EHRs. METHODS: We propose a two-step pipeline method to identify NIHSS items and numerical scores and validate its feasibility using a freely accessible critical care database: MIMIC-III (Medical Information Mart for Intensive Care III). First, we utilize MIMIC-III to create an annotated corpus. Then, we investigate possible machine learning methods for two subtasks, NIHSS item and score recognition and item-score relation extraction. In the evaluation, we conduct both task-specific and end-to-end evaluations and compare our method with the rule-based method using precision, recall and F1 scores as evaluation metrics. RESULTS: We use all available discharge summaries of stroke cases in MIMIC-III. The annotated NIHSS corpus contains 312 cases, 2929 scale items, 2774 scores and 2733 relations. The results show that the best F1-score of our method was 0.9006, which was attained by combining BERT-BiLSTM-CRF and Random Forest, and it outperformed the rule-based method (F1-score = 0.8098). In the end-to-end task, our method could successfully recognize the item "1b level of consciousness questions", the score "1" and their relation "('1b level of consciousness questions', '1', 'has value')" from the sentence "1b level of consciousness questions: said name = 1", while the rule-based method could not. CONCLUSIONS: The two-step pipeline method we propose is an effective approach to identify NIHSS items, scores and their relations. With its help, clinical investigators can easily retrieve and access structured scale data, thereby supporting stroke-related real-world studies.

The traditional wet-experiment-guided drug discovery is a labor-consuming and time-consuming process. Quite a few computational drug repositioning approaches have been proposed to predict potential drug-disease associations for the discovery of new indications for drugs and new therapies for diseases. Among them, heterogeneous graph neural network-based approaches can learn drug/disease topological representations on heterogeneous graphs and then give precise inferences for unconfirmed drug-disease associations. However, the existing approaches ignored the meta-paths in the drug-disease networks which could enhance the model performance and interpretability. In this study, we first proposed a multi-instance learning-based heterogeneous graph network approach for drug-disease association prediction, which is called MilGNet. Fusing with heterogeneous graph convolutional layer, the MilGNet learns meta-path-level representations for given drug-disease pairs by a novel pseudo meta-path instance generator and a bidirectional translating embedding projector. Then, an attention-based multi-scale interpretable joint predictor is assembled for precise and rational drug-disease association prediction. Comprehensive experiments have demonstrated the effectiveness of MilGNet compared to 6 advanced approaches. Meanwhile, the case study also shows the model interpretability of MilGNet by identifying high confident meta-paths. Our adopted benchmark dataset and source code are available at https://github.com/gu-yaowen/MilGNet..

Computational methods have been widely applied to resolve various core issues in drug discovery, such as molecular property prediction. In recent years, a data-driven computational method-deep learning had achieved a number of impressive successes in various domains. In drug discovery, graph neural networks (GNNs) take molecular graph data as input and learn graph-level representations in non-Euclidean space. An enormous amount of well-performed GNNs have been proposed for molecular graph learning. Meanwhile, efficient use of molecular data during training process, however, has not been paid enough attention. Curriculum learning (CL) is proposed as a training strategy by rearranging training queue based on calculated samples' difficulties, yet the effectiveness of CL method has not been determined in molecular graph learning. In this study, inspired by chemical domain knowledge and task prior information, we proposed a novel CL-based training strategy to improve the training efficiency of molecular graph learning, called CurrMG. Consisting of a difficulty measurer and a training scheduler, CurrMG is designed as a plug-and-play module, which is model-independent and easy-to-use on molecular data. Extensive experiments demonstrated that molecular graph learning models could benefit from CurrMG and gain noticeable improvement on five GNN models and eight molecular property prediction tasks (overall improvement is 4.08%). We further observed CurrMG's encouraging potential in resource-constrained molecular property prediction. These results indicate that CurrMG can be used as a reliable and efficient training strategy for molecular graph learning. Availability: The source code is available in https://github.com/gu-yaowen/CurrMG.

Nowadays computational methods in bioinformatics and cheminformatics have been widely used in molecular property prediction, advancing activities such as drug discovery. Combining to expert manual annotation of molecular features, machine learning approaches have gained satisfying prediction accuracies in most molecular property prediction tasks. Recently, Graph neural networks (GNNs) have gained increasing popularity in cheminformatics, where a chemical molecule structure is represented as a graph, and have made monumental progress in molecular property prediction. However, GNNs models requires large amounts of training samples, and the diversified molecular structure information might under-utilized when the model is trained with traditional random sampling strategies, thus leading to redundancy and inefficiency. Similar to human learning procedures, training of molecule graph learning models can benefit from an easy-to-difficult curriculum. In this study, we proposed a curriculum learning approach for graph based molecular property prediction, called CurrMG. A data-aware integrated difficulty measurer was proposed to distinguish easy molecules from complex ones. Without any model redesign or external data, our training strategy improves model efficiency and accuracy in numerous molecular property prediction tasks and shows potential for low data drug discovery.

Computational drug repositioning or repurposing is a promising and efficient tool for discovering new uses from existing drugs and holds the great potential for precision medicine in the age of big data. The explosive growth of large-scale genomic and phenotypic data, as well as data of small molecular compounds with granted regulatory approval, is enabling new developments for computational repositioning. To achieve the shortest path toward new drug indications, advanced data processing and analysis strategies are critical for making sense of these heterogeneous molecular measurements. In this review, we show recent advancements in the critical areas of computational drug repositioning from multiple aspects. First, we summarize available data sources and the corresponding computational repositioning strategies. Second, we characterize the commonly used computational techniques. Third, we discuss validation strategies for repositioning studies, including both computational and experimental methods. Finally, we highlight potential opportunities and use-cases, including a few target areas such as cancers. We conclude with a brief discussion of the remaining challenges in computational drug repositioning.

Drug-disease treatment relationships, i.e., which drug(s) are indicated to treat which disease(s), are among the most frequently sought information in PubMed®. Such information is useful for feeding the Google Knowledge Graph, designing computational methods to predict novel drug indications, and validating clinical information in EMRs. Given the importance and utility of this information, there have been several efforts to create repositories of drugs and their indications. However, existing resources are incomplete. Furthermore, they neither label indications in a structured way nor differentiate them by drug-specific properties such as dosage form, and thus do not support computer processing or semantic interoperability. More recently, several studies have proposed automatic methods to extract structured indications from drug descriptions; however, their performance is limited by natural language challenges in disease named entity recognition and indication selection. In response, we report LabeledIn: a human-reviewed, machine-readable and source-linked catalog of labeled indications for human drugs. More specifically, we describe our semi-automatic approach to derive LabeledIn from drug descriptions through human annotations with aids from automatic methods. As the data source, we use the drug labels (or package inserts) submitted to the FDA by drug manufacturers and made available in DailyMed. Our machine-assisted human annotation workflow comprises: (i) a grouping method to remove redundancy and identify representative drug labels to be used for human annotation, (ii) an automatic method to recognize and normalize mentions of diseases in drug labels as candidate indications, and (iii) a two-round annotation workflow for human experts to judge the pre-computed candidates and deliver the final gold standard. In this study, we focused on 250 highly accessed drugs in PubMed Health, a newly developed public web resource for consumers and clinicians on prevention and treatment of diseases. These 250 drugs corresponded to more than 8000 drug labels (500 unique) in DailyMed in which 2950 candidate indications were pre-tagged by an automatic tool. After being reviewed independently by two experts, 1618 indications were selected, and additional 97 (missed by computer) were manually added, with an inter-annotator agreement of 88.35% as measured by the Kappa coefficient. Our final annotation results in LabeledIn consist of 7805 drug-disease treatment relationships where drugs are represented as a triplet of ingredient, dose form, and strength. A systematic comparison of LabeledIn with an existing computer-derived resource revealed significant discrepancies, confirming the need to involve humans in the creation of such a resource. In addition, LabeledIn is unique in that it contains detailed textual context of the selected indications in drug labels, making it suitable for the development of advanced computational methods for the automatic extraction of indications from free text. Finally, motivated by the studies on drug nomenclature and medication errors in EMRs, we adopted a fine-grained drug representation scheme, which enables the automatic identification of drugs with indications specific to certain dose forms or strengths. Future work includes expanding our coverage to more drugs and integration with other resources. The LabeledIn dataset and the annotation guidelines are available at http://ftp.ncbi.nlm.nih.gov/pub/lu/LabeledIn/.

BACKGROUND: Recent in vivo studies showed new hopes of drug repositioning through causality inference from drugs to disease. Inspired by their success, here we present an in silico method for building a causal network (CauseNet) between drugs and diseases, in an attempt to systematically identify new therapeutic uses of existing drugs. METHODS: Unlike the traditional 'one drug-one target-one disease' causal model, we simultaneously consider all possible causal chains connecting drugs to diseases via target- and gene-involved pathways based on rich information in several expert-curated knowledge-bases. With statistical learning, our method estimates transition likelihood of each causal chain in the network based on known drug-disease treatment associations (e.g. bexarotene treats skin cancer). RESULTS: To demonstrate its validity, our method showed high performance (AUC = 0.859) in cross validation. Moreover, our top scored prediction results are highly enriched in literature and clinical trials. As a showcase of its utility, we show several drugs for potential re-use in Crohn's Disease. CONCLUSIONS: We successfully developed a computational method for discovering new uses of existing drugs based on casual inference in a layered drug-target-pathway-gene- disease network. The results showed that our proposed method enables hypothesis generation from public accessible biological data for drug repositioning.

drug discovery is known as a high cost and high risk process. In response, recently there is an increasing interest in discovering new indications for known drugs-a process known as drug repositioning-using computational methods. In this study, we present a new systematic approach for identifying potential new indications of an existing drug through its relation to similar drugs. Different from the previous similarity-based methods, we adapted a novel bipartite-graph based method when considering common drug targets and their interaction information. Furthermore, we added drug structure information into the calculation of drug pairwise similarity. In cross-validation experiments, our method achieved a sensitivity of 0.77 and specificity of 0.92 (AUC = 0.888) and compared favorably to the state of the art. When compared with a control group of drug uses, our drug repositioning results were found to be significantly enriched in both the biomedical literature and clinical trials. Our results indicate that combining chemical structure and drug target information results in better prediction performance and that the proposed approach successfully captures the implicit information between drug targets.

Recent progress in high-throughput genomic technologies has shifted pharmacogenomic research from candidate gene pharmacogenetics to clinical pharmacogenomics (PGx). Many clinical related questions may be asked such as 'what drug should be prescribed for a patient with mutant alleles?' Typically, answers to such questions can be found in publications mentioning the relationships of the gene-drug-disease of interest. In this work, we hypothesize that ClinicalTrials.gov is a comparable source rich in PGx related information. In this regard, we developed a systematic approach to automatically identify PGx relationships between genes, drugs and diseases from trial records in ClinicalTrials.gov. In our evaluation, we found that our extracted relationships overlap significantly with the curated factual knowledge through the literature in a PGx database and that most relationships appear on average 5 years earlier in clinical trials than in their corresponding publications, suggesting that clinical trials may be valuable for both validating known and capturing new PGx related information in a more timely manner. Furthermore, two human reviewers judged a portion of computer-generated relationships and found an overall accuracy of 74% for our text-mining approach. This work has practical implications in enriching our existing knowledge on PGx gene-drug-disease relationships as well as suggesting crosslinks between ClinicalTrials.gov and other PGx knowledge bases.

The recently proposed concept of molecular connectivity maps enables researchers to integrate experimental measurements of genes, proteins, metabolites, and drug compounds under similar biological conditions. The study of these maps provides opportunities for future toxicogenomics and drug discovery applications. We developed a computational framework to build disease-specific drug-protein connectivity maps. We integrated gene/protein and drug connectivity information based on protein interaction networks and literature mining, without requiring gene expression profile information derived from drug perturbation experiments on disease samples. We described the development and application of this computational framework using Alzheimer's Disease (AD) as a primary example in three steps. First, molecular interaction networks were incorporated to reduce bias and improve relevance of AD seed proteins. Second, PubMed abstracts were used to retrieve enriched drug terms that are indirectly associated with AD through molecular mechanistic studies. Third and lastly, a comprehensive AD connectivity map was created by relating enriched drugs and related proteins in literature. We showed that this molecular connectivity map development approach outperformed both curated drug target databases and conventional information retrieval systems. Our initial explorations of the AD connectivity map yielded a new hypothesis that diltiazem and quinidine may be investigated as candidate drugs for AD treatment. Molecular connectivity maps derived computationally can help study molecular signature differences between different classes of drugs in specific disease contexts. To achieve overall good data coverage and quality, a series of statistical methods have been developed to overcome high levels of data noise in biological networks and literature mining results. Further development of computational molecular connectivity maps to cover major disease areas will likely set up a new model for drug development, in which therapeutic/toxicological profiles of candidate drugs can be checked computationally before costly clinical trials begin.