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.

Reducing overconfident errors in molecular property classification using Posterior Network.

Deep-learning-based classification models are increasingly used for predicting molecular properties in drug development. However, traditional classification models using the Softmax function often give overconfident mispredictions for out-of-distribution samples, highlighting a critical lack of accurate uncertainty estimation. Such limitations can result in substantial costs and should be avoided during drug development. Inspired by advances in evidential deep learning and Posterior Network, we replaced the Softmax function with a normalizing flow to enhance the uncertainty estimation ability of the model in molecular property classification. The proposed strategy was evaluated across diverse scenarios, including simulated experiments based on a synthetic dataset, ADMET predictions, and ligand-based virtual screening. The results demonstrate that compared with the vanilla model, the proposed strategy effectively alleviates the problem of giving overconfident but incorrect predictions. Our findings support the promising application of evidential deep learning in drug development and offer a valuable framework for further research.

Gram matrix: an efficient representation of molecular conformation and learning objective for molecular pretraining.

Accurate prediction of molecular properties is fundamental in drug discovery and development, providing crucial guidance for effective drug design. A critical factor in achieving accurate molecular property prediction lies in the appropriate representation of molecular structures. Presently, prevalent deep learning-based molecular representations rely on 2D structure information as the primary molecular representation, often overlooking essential three-dimensional (3D) conformational information due to the inherent limitations of 2D structures in conveying atomic spatial relationships. In this study, we propose employing the Gram matrix as a condensed representation of 3D molecular structures and for efficient pretraining objectives. Subsequently, we leverage this matrix to construct a novel molecular representation model, Pre-GTM, which inherently encapsulates 3D information. The model accurately predicts the 3D structure of a molecule by estimating the Gram matrix. Our findings demonstrate that Pre-GTM model outperforms the baseline Graphormer model and other pretrained models in the QM9 and MoleculeNet quantitative property prediction task. The integration of the Gram matrix as a condensed representation of 3D molecular structure, incorporated into the Pre-GTM model, opens up promising avenues for its potential application across various domains of molecular research, including drug design, materials science, and chemical engineering.

In silico off-target profiling for enhanced drug safety assessment.

Ensuring drug safety in the early stages of drug development is crucial to avoid costly failures in subsequent phases. However, the economic burden associated with detecting drug off-targets and potential side effects through in vitro safety screening and animal testing is substantial. Drug off-target interactions, along with the adverse drug reactions they induce, are significant factors affecting drug safety. To assess the liability of candidate drugs, we developed an artificial intelligence model for the precise prediction of compound off-target interactions, leveraging multi-task graph neural networks. The outcomes of off-target predictions can serve as representations for compounds, enabling the differentiation of drugs under various ATC codes and the classification of compound toxicity. Furthermore, the predicted off-target profiles are employed in adverse drug reaction (ADR) enrichment analysis, facilitating the inference of potential ADRs for a drug. Using the withdrawn drug Pergolide as an example, we elucidate the mechanisms underlying ADRs at the target level, contributing to the exploration of the potential clinical relevance of newly predicted off-target interactions. Overall, our work facilitates the early assessment of compound safety/toxicity based on off-target identification, deduces potential ADRs of drugs, and ultimately promotes the secure development of drugs.

Deep representation learning of chemical-induced transcriptional profile for phenotype-based drug discovery.

Artificial intelligence transforms drug discovery, with phenotype-based approaches emerging as a promising alternative to target-based methods, overcoming limitations like lack of well-defined targets. While chemical-induced transcriptional profiles offer a comprehensive view of drug mechanisms, inherent noise often obscures the true signal, hindering their potential for meaningful insights. Here, we highlight the development of TranSiGen, a deep generative model employing self-supervised representation learning. TranSiGen analyzes basal cell gene expression and molecular structures to reconstruct chemical-induced transcriptional profiles with high accuracy. By capturing both cellular and compound information, TranSiGen-derived representations demonstrate efficacy in diverse downstream tasks like ligand-based virtual screening, drug response prediction, and phenotype-based drug repurposing. Notably, in vitro validation of TranSiGen's application in pancreatic cancer drug discovery highlights its potential for identifying effective compounds. We envisage that integrating TranSiGen into the drug discovery and mechanism research holds significant promise for advancing biomedicine.

Multi-Omics Insights into Disulfidptosis-Related Genes Reveal RPN1 as a Therapeutic Target for Liver Cancer.

Disulfidptosis, a newly identified mode of programmed cell death, is yet to be comprehensively elucidated with respect to its multi-omics characteristics in tumors, specific pathogenic mechanisms, and antitumor functions in liver cancer. This study included 10,327 tumor and normal tissue samples from 33 cancer types. In-depth analyses using various bioinformatics tools revealed widespread dysregulation of disulfidptosis-related genes (DRGs) in pan-cancer and significant associations with prognosis, genetic variations, tumor stemness, methylation levels, and drug sensitivity. Univariate and multivariate Cox regression and LASSO regression were used to screen and construct prognosis-related hub DRGs and predictive models in the context of liver cancer. Subsequently, single cell analysis was conducted to investigate the subcellular localization of RPN1, a hub DRG, in various solid tumors. Western blotting was performed to validate the expression of RPN1 at both cellular and tissue levels. Additionally, functional experiments, including CCK8, EdU, clone, and transwell assays, indicated that RPN1 knockdown promoted the proliferative and invasive capacities of liver cancer cells. Therefore, this study elucidated the multi-omics characteristics of DRGs in pan-cancer and established a prognostic model for liver cancer. Additionally, this study revealed the molecular functions of RPN1 in liver cancer, suggesting its potential as a therapeutic target for this disease.

LncRNA PVT1 facilitates the growth and metastasis of colorectal cancer by sponging with miR-3619-5p to regulate TRIM29 expression.

BACKGROUND: Colorectal cancer (CRC) is the second most common cause of cancer-related death worldwide. Long noncoding RNA (lncRNA) is involved in many malignant tumors. This study aimed to clarify the role of the lncRNA plasmacytoma variant translocation 1 (PVT1) in CRC growth and metastasis. METHODS: Differentially expressed lncRNAs in CRC were analyzed using the Cancer Genome Atlas. Gene expression profiling interactive analysis and a comprehensive resource for lncRNAs from cancer arrays databases were used to analyze lncRNA PVT1 expression and CRC prognosis, respectively. Cell counting kit-8, wound healing, colony formation, Transwell, and immunofluorescence assays were used to evaluate CRC cell proliferation, migration, invasion, and epithelial-mesenchymal transition (EMT), respectively. Tumor growth and metastasis models were used to explore the PVT1 effect on the growth and metastasis of CRC in vivo. RESULTS: PVT1 was highly expressed in CRC, associated with a poor prognosis of CRC, and showed good diagnostic value. Transfection of sh-PVT1 or pcDNA3.1-PVT1 reduced or increased the proliferation, wound healing rate, colony formation, invasion, and EMT of CRC cells. PVT1 and miR-3619-5p were co-expressed in CRC cytoplasm, and PVT1 acted as a competitive endogenous RNA (ceRNA) by sponging miR-3619-5p to up-regulate tripartite motif containing 29 (TRIM29) expression. MiR-3619-5p overexpression and TRIM29 knockdown reduced proliferation, wound healing rate, invasion, and EMT of CRC cells. However, simultaneous PVT1 and miR-3619-5p overexpression or knockdown of miR-3619-5p and TRIM29 knockdown rescued the malignant phenotype of CRC cells. CONCLUSIONS: We first clarified the ceRNA mechanism of PVT1 in CRC, which induced growth and metastasis by sponging with miR-3619-5p to regulate TRIM29.

KinomeMETA: a web platform for kinome-wide polypharmacology profiling with meta-learning.

Kinase-targeted inhibitors hold promise for new therapeutic options, with multi-target inhibitors offering the potential for broader efficacy while minimizing polypharmacology risks. However, comprehensive experimental profiling of kinome-wide activity is expensive, and existing computational approaches often lack scalability or accuracy for understudied kinases. We introduce KinomeMETA, an artificial intelligence (AI)-powered web platform that significantly expands the predictive range with scalability for predicting the polypharmacological effects of small molecules across the kinome. By leveraging a novel meta-learning algorithm, KinomeMETA efficiently utilizes sparse activity data, enabling rapid generalization to new kinase tasks even with limited information. This significantly expands the repertoire of accurately predictable kinases to 661 wild-type and clinically-relevant mutant kinases, far exceeding existing methods. Additionally, KinomeMETA empowers users to customize models with their proprietary data for specific research needs. Case studies demonstrate its ability to discover new active compounds by quickly adapting to small dataset. Overall, KinomeMETA offers enhanced kinome virtual profiling capabilities and is positioned as a powerful tool for developing new kinase inhibitors and advancing kinase research. The KinomeMETA server is freely accessible without registration at https://kinomemeta.alphama.com.cn/.

Targeting JMJD1C to selectively disrupt tumor Treg cell fitness enhances antitumor immunity.

Regulatory T (Treg) cells are critical for immune tolerance but also form a barrier to antitumor immunity. As therapeutic strategies involving Treg cell depletion are limited by concurrent autoimmune disorders, identification of intratumoral Treg cell-specific regulatory mechanisms is needed for selective targeting. Epigenetic modulators can be targeted with small compounds, but intratumoral Treg cell-specific epigenetic regulators have been unexplored. Here, we show that JMJD1C, a histone demethylase upregulated by cytokines in the tumor microenvironment, is essential for tumor Treg cell fitness but dispensable for systemic immune homeostasis. JMJD1C deletion enhanced AKT signals in a manner dependent on histone H3 lysine 9 dimethylation (H3K9me2) demethylase and STAT3 signals independently of H3K9me2 demethylase, leading to robust interferon-γ production and tumor Treg cell fragility. We have also developed an oral JMJD1C inhibitor that suppresses tumor growth by targeting intratumoral Treg cells. Overall, this study identifies JMJD1C as an epigenetic hub that can integrate signals to establish tumor Treg cell fitness, and we present a specific JMJD1C inhibitor that can target tumor Treg cells without affecting systemic immune homeostasis.

TG468: a text graph convolutional network for predicting clinical response to immune checkpoint inhibitor therapy.

Enhancing cancer treatment efficacy remains a significant challenge in human health. Immunotherapy has witnessed considerable success in recent years as a treatment for tumors. However, due to the heterogeneity of diseases, only a fraction of patients exhibit a positive response to immune checkpoint inhibitor (ICI) therapy. Various single-gene-based biomarkers and tumor mutational burden (TMB) have been proposed for predicting clinical responses to ICI; however, their predictive ability is limited. We propose the utilization of the Text Graph Convolutional Network (GCN) method to comprehensively assess the impact of multiple genes, aiming to improve the predictive capability for ICI response. We developed TG468, a Text GCN model framing drug response prediction as a text classification task. By combining natural language processing (NLP) and graph neural network techniques, TG468 effectively handles sparse and high-dimensional exome sequencing data. As a result, TG468 can distinguish survival time for patients who received ICI therapy and outperforms single gene biomarkers, TMB and some classical machine learning models. Additionally, TG468's prediction results facilitate the identification of immune status differences among specific patient types in the Cancer Genome Atlas dataset, providing a rationale for the model's predictions. Our approach represents a pioneering use of a GCN model to analyze exome data in patients undergoing ICI therapy and offers inspiration for future research using NLP technology to analyze exome sequencing data.

The emergence of machine learning force fields in drug design.

In the field of molecular simulation for drug design, traditional molecular mechanic force fields and quantum chemical theories have been instrumental but limited in terms of scalability and computational efficiency. To overcome these limitations, machine learning force fields (MLFFs) have emerged as a powerful tool capable of balancing accuracy with efficiency. MLFFs rely on the relationship between molecular structures and potential energy, bypassing the need for a preconceived notion of interaction representations. Their accuracy depends on the machine learning models used, and the quality and volume of training data sets. With recent advances in equivariant neural networks and high-quality datasets, MLFFs have significantly improved their performance. This review explores MLFFs, emphasizing their potential in drug design. It elucidates MLFF principles, provides development and validation guidelines, and highlights successful MLFF implementations. It also addresses potential challenges in developing and applying MLFFs. The review concludes by illuminating the path ahead for MLFFs, outlining the challenges to be overcome and the opportunities to be harnessed. This inspires researchers to embrace MLFFs in their investigations as a new tool to perform molecular simulations in drug design.

KinomeMETA: meta-learning enhanced kinome-wide polypharmacology profiling.

Kinase inhibitors are crucial in cancer treatment, but drug resistance and side effects hinder the development of effective drugs. To address these challenges, it is essential to analyze the polypharmacology of kinase inhibitor and identify compound with high selectivity profile. This study presents KinomeMETA, a framework for profiling the activity of small molecule kinase inhibitors across a panel of 661 kinases. By training a meta-learner based on a graph neural network and fine-tuning it to create kinase-specific learners, KinomeMETA outperforms benchmark multi-task models and other kinase profiling models. It provides higher accuracy for understudied kinases with limited known data and broader coverage of kinase types, including important mutant kinases. Case studies on the discovery of new scaffold inhibitors for membrane-associated tyrosine- and threonine-specific cdc2-inhibitory kinase and selective inhibitors for fibroblast growth factor receptors demonstrate the role of KinomeMETA in virtual screening and kinome-wide activity profiling. Overall, KinomeMETA has the potential to accelerate kinase drug discovery by more effectively exploring the kinase polypharmacology landscape.

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.

Single-cell and genetic multi-omics analysis combined with experiments confirmed the signature and potential targets of cuproptosis in hepatocellular carcinoma.

Background: Cuproptosis, as a recently discovered type of programmed cell death, occupies a very important role in hepatocellular carcinoma (HCC) and provides new methods for immunotherapy; however, the functions of cuproptosis in HCC are still unclear. Methods: We first analyzed the transcriptome data and clinical information of 526 HCC patients using multiple algorithms in R language and extensively described the copy number variation, prognostic and immune infiltration characteristics of cuproptosis related genes (CRGs). Then, the hub CRG related genes associated with prognosis through LASSO and Cox regression analyses and constructed a prognostic prediction model including multiple molecular markers and clinicopathological parameters through training cohorts, then this model was verified by test cohorts. On the basis of the model, the clinicopathological indicators, immune infiltration and tumor microenvironment characteristics of HCC patients were further explored via bioinformation analysis. Then, We further explored the key gene biological function by single-cell analysis, cell viability and transwell experiments. Meantime, we also explored the molecular docking of the hub genes. Results: We have screened 5 hub genes associated with HCC prognosis and constructed a prognosis prediction scoring model. And the model results showed that patients in the high-risk group had poor prognosis and the expression levels of multiple immune markers, including PD-L1, CD276 and CTLA4, were higher than those patients in the low-risk group. We found a significant correlation between risk score and M0 macrophages and memory CD4+ T cells. And the single-cell analysis and molecular experiments showed that BEX1 were higher expressed in HCC tissues and deletion inhibited the proliferation, invasion and migration and EMT pathway of HCC cells. Finally, it was observed that BEX1 could bind to sorafenib to form a stable conformation. Conclusion: The study not only revealed the multiomics characteristics of CRGs in HCC but also constructed a new high-accuracy prognostic prediction model. Meanwhile, BEX1 were also identified as hub genes that can mediate the cuproptosis of hepatocytes as potential therapeutic targets for HCC.

Evaluation of bladder filling effects on the dose distribution during radiotherapy for cervical cancer based on daily CT images.

PURPOSE: This study aimed to assess the effects of bladder filling during cervical cancer radiotherapy on target volume and organs at risk (OARs) dose based on daily computed tomography (daily-CT) images and provide bladder-volume-based dose prediction models. METHODS: Nineteen patients (475 daily-CTs) comprised the study group, and five patients comprised the validation set (25 daily-CTs). Target volumes and OARs were delineated on daily-CT images and the treatment plan was recalculated accordingly. The deviation from the planning bladder volume (DVB), the correlation between DVB and clinical (CTV)/planning (PTV) target volume in terms of prescribed dose coverage, and the relationship of small bowel volume and bladder dose with the ratio of bladder volume (RVB) were analyzed. RESULTS: In all cases, the prescribed dose coverage in the CTV was >95% when DVB was <200 cm3 , whereas that in the PTV was >95% when RVB was <160%. The ratio of bladder V45 Gy to the planning bladder V45 Gy (RBV45 ) exhibited a negative linear relationship with RVB (RBV45  = -0.18*RVB + 120.8; R2  = 0.80). Moreover, the ratio of small bowel volume to planning small bowel volume (RVS) exhibited a negative linear relationship with RVB (RVS = -1.06*RVB +217.59; R2  = 0.41). The validation set results showed that the linear model predicted well the effects of bladder volume changes on target volume coverage and bladder dose. CONCLUSIONS: This study assessed dosimetry and volume effects of bladder filling on target and OARs based on daily-CT images. We established a quantitative relationship between these parameters, providing dose prediction models for cervical cancer radiotherapy.

Tora3D: an autoregressive torsion angle prediction model for molecular 3D conformation generation.

Three-dimensional (3D) conformations of a small molecule profoundly affect its binding to the target of interest, the resulting biological effects, and its disposition in living organisms, but it is challenging to accurately characterize the conformational ensemble experimentally. Here, we proposed an autoregressive torsion angle prediction model Tora3D for molecular 3D conformer generation. Rather than directly predicting the conformations in an end-to-end way, Tora3D predicts a set of torsion angles of rotatable bonds by an interpretable autoregressive method and reconstructs the 3D conformations from them, which keeps structural validity during reconstruction. Another advancement of our method over other conformational generation methods is the ability to use energy to guide the conformation generation. In addition, we propose a new message-passing mechanism that applies the Transformer to the graph to solve the difficulty of remote message passing. Tora3D shows superior performance to prior computational models in the trade-off between accuracy and efficiency, and ensures conformational validity, accuracy, and diversity in an interpretable way. Overall, Tora3D can be used for the quick generation of diverse molecular conformations and 3D-based molecular representation, contributing to a wide range of downstream drug design tasks.

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.

Structural Optimization of Fibroblast Growth Factor Receptor Inhibitors for Treating Solid Tumors.

Small-molecule fibroblast growth factor receptor (FGFR) inhibitors have emerged as a promising antitumor therapy. Herein, by further optimizing the lead compound 1 under the guidance of molecular docking, we obtained a series of novel covalent FGFR inhibitors. After careful structure-activity relationship analysis, several compounds were identified to exhibit strong FGFR inhibitory activity and relatively better physicochemical and pharmacokinetic properties compared with those of 1. Among them, 2e potently and selectively inhibited the kinase activity of FGFR1-3 wildtype and high-incidence FGFR2-N549H/K-resistant mutant kinase. Furthermore, it suppressed cellular FGFR signaling, exhibiting considerable antiproliferative activity in FGFR-aberrant cancer cell lines. In addition, the oral administration of 2e in the FGFR1-amplified H1581, FGFR2-amplified NCI-H716, and SNU-16 tumor xenograft models demonstrated potent antitumor efficacy, inducing tumor stasis or even tumor regression.

Double-target neural circuit-magnetic stimulation improves motor function in spinal cord injury by attenuating astrocyte activation.

Multi-target neural circuit-magnetic stimulation has been clinically shown to improve rehabilitation of lower limb motor function after spinal cord injury. However, the precise underlying mechanism remains unclear. In this study, we performed double-target neural circuit-magnetic stimulation on the left motor cortex and bilateral L5 nerve root for 3 successive weeks in a rat model of incomplete spinal cord injury caused by compression at T10. Results showed that in the injured spinal cord, the expression of the astrocyte marker glial fibrillary acidic protein and inflammatory factors interleukin 1ß, interleukin-6, and tumor necrosis factor-α had decreased, whereas that of neuronal survival marker microtubule-associated protein 2 and synaptic plasticity markers postsynaptic densification protein 95 and synaptophysin protein had increased. Additionally, neural signaling of the descending corticospinal tract was markedly improved and rat locomotor function recovered significantly. These findings suggest that double-target neural circuit-magnetic stimulation improves rat motor function by attenuating astrocyte activation, thus providing a theoretical basis for application of double-target neural circuit-magnetic stimulation in the clinical treatment of spinal cord injury.

Computing the relative binding affinity of ligands based on a pairwise binding comparison network.

Structure-based lead optimization is an open challenge in drug discovery, which is still largely driven by hypotheses and depends on the experience of medicinal chemists. Here we propose a pairwise binding comparison network (PBCNet) based on a physics-informed graph attention mechanism, specifically tailored for ranking the relative binding affinity among congeneric ligands. Benchmarking on two held-out sets (provided by Schrödinger and Merck) containing over 460 ligands and 16 targets, PBCNet demonstrated substantial advantages in terms of both prediction accuracy and computational efficiency. Equipped with a fine-tuning operation, the performance of PBCNet reaches that of Schrödinger's FEP+, which is much more computationally intensive and requires substantial expert intervention. A further simulation-based experiment showed that active learning-optimized PBCNet may accelerate lead optimization campaigns by 473%. Finally, for the convenience of users, a web service for PBCNet is established to facilitate complex relative binding affinity prediction through an easy-to-operate graphical interface.