Yanyan Diao

RingKin encompasses 72.27 million macrocycles generated from 495 approved or clinical-stage kinase drugs, offering the first intuitive visualization of the underexploited macrocyclic space for drug design.

Drafting chemical compound patents, particularly those involving Markush structures, is a complex task often hindered by manual workflows that are time‐consuming and error‐prone and may result in incomplete protection. To address these challenges, we introduce SpaceExpander, an open‐source tool designed to automate the generation of Markush patent claims with improved accuracy and efficiency. The tool systematically extracts molecular scaffolds, identifies R‐groups, and constructs comprehensive Markush structures to expand patent coverage. To enhance accessibility and usability, we provide a publicly available web server for online usage, along with standalone executables compatible with Windows, macOS, and Linux to support offline access.

Macrocyclic compounds hold great promise as therapeutic agents. However, their structural optimization remains constrained by the limited availability of bioactive candidates, which in turn hampers the systematic exploration of structure-activity relationships. Here we introduce CycleGPT, a generative chemical language model designed specifically to address these challenges. CycleGPT is characterized by a progressive transfer learning paradigm that incrementally transfers knowledge from pre-trained chemical language models to specialized macrocycle generation, thereby overcoming the data shortage issue. Meanwhile, it adopts an innovative probabilistic sampling strategy that effectively improves the structural novelty of generated macrocycles while ensuring domain-specific adaptability. In a prospective drug design based on CycleGPT and a JAK2 activity prediction model, we successfully developed a new JAK2 drug candidate with a good selectivity profile (inhibiting 17 wild-type kinases) and promising potential for treating polycythemia in vivo, demonstrating the practicality of deep learning methods in macrocyclic drug design.

ANO1, a calcium-activated chloride channel, is a newly reported therapeutic target for osteoporosis. Tamsulosin (Tam), an approved α₁-AR antagonist, was previously discovered to be a novel ANO1 allosteric inhibitor. Here, a series of Tam derivatives were designed and synthesized with the aim of developing selective ANO1 inhibitors for osteoporosis treatment. Among them, compound 10 exhibited the best overall activities. The potency of compound 10 against ANO1 was comparable to that of Tam (IC₅₀: 4.57 vs 7.22 μM), but its selectivity over α₁-AR was significantly improved. Compared with Tam, the potency of compound 10 against α_1A-, α_1B-, and α_1D-AR decreased by 118-fold, 642-fold, and 10,000-fold, respectively. In addition, compound 10 exhibited significant inhibitory effects on osteoclast differentiation and function. In the OVX (ovariectomy) mice, compound 10 effectively prevented OVX-induced bone loss. Collectively, these findings highlighted the therapeutic potential of 10 as a novel lead compound for antiosteoporosis by targeting ANO1.

Janus kinase 2 with the activating mutation V617F is a potential target for the treatment of myeloproliferative neoplasms. Current JAK2 inhibitors often suffer from multiple pharmacological limitations such as limited selectivity and rapid metabolic clearance. In this study, we describe the design, synthesis, and biological evaluation of a series of macrocyclic 2-amino-4-phenylaminopyrimidine derivatives as JAK2 inhibitors. Among them, compound 11 showed potent inhibitory activity against JAK2 kinase (IC₅₀ = 54.70 nM), as well as JAK2-dependent HEL and SET-2 cells, with IC₅₀ values of 0.57 μM and 1.07 μM, respectively. Besides, compound 11 downregulated the expression of p-STAT3 and p-STAT5 in HEL cells. Importantly, compound 11 also exhibited good bioavailability (F = 54.8%) and potent inhibition of rhEPO-induced extramedullary erythropoiesis and polycythemia vera in vivo. In summary, the macrocyclic design introduces structural innovation and provides a more efficient and selective strategy for JAK2 inhibition.

Macrocyclic compounds have shown great potential as therapeutic agents due to their distinctive structural and pharmacological properties. However, structural optimization of macrocycles—a critical step in the rational design of macrocyclic drugs—remains constrained by the limited availability of bioactive candidates, which in turn hampers the systematic exploration of structure-activity relationships. In this paper, we introduce CycleGPT, a generative chemical language model designed specifically to address these challenges. CycleGPT is characterized by a progressive transfer learning paradigm to incrementally transfer knowledge from pre-trained chemical language models to specialized macrocycle generation to overcome the data shortage issue; in the meantime, it adopts an innovative probabilistic sampling strategy that effectively improves the structural novelty of generated macrocycles while ensuring domain-specific adaptability. In a prospective drug design based on CycleGPT and our custom JAK2 activity predictive model, three synthesized macrocycles exhibited high inhibitory activity against JAK2, with the most potent compound 2 showing an IC₅₀ of 1.17 nM. Moreover, compound 2 exhibited a favorable selectivity profile and demonstrated in vivo efficacy in polycythemia mice model. These novel therapeutic candidates demonstrate the significant potential of CycleGPT for advancing macrocyclic drug discovery.

Retrosynthetic route planning is essential for designing efficient pathways to synthesize complex molecules, serving as a cornerstone in drug discovery and organic synthesis. Sequence-based models have become a predominant approach in retrosynthetic route planning, yet its validity and robustness remain limited by the challenges in molecular representation methods. Current methods typically treat reactants and products as independent molecules, overlooking structural relationships crucial for accurate synthesis predictions. Herein, we introduce RPSubAlign, a molecular sequence representation method specifically tailored for retrosynthetic tasks, which aligns common substructures between reactants and products to enhance the validity and robustness of sequence-based models. Compared with conventional random and root-alignment representations, RPSubAlign achieves better performance on the USPTO-50K and USPTO-MIT datasets, improving up to a 34.8% increase in Top-N accuracy (with Self-Referencing Embedded Strings representation) and demonstrating enhanced stability across various data augmentation scenarios. RPSubAlign significantly improves syntactic validity, reaching 86.64% on USPTO-50K and 96.45% on USPTO-MIT (with Simplified Molecular Input Line Entry System representation), outperforming baseline methods. These results highlight RPSubAlign as a robust, effective approach for molecular characterization method for retrosynthesis predictions. Code for RPSubAlign is available at https://github.com/Aminoacid1226/RPSubAlign.

Bruton's tyrosine kinase (BTK) is a therapeutic target for B-cell-driven malignancies. Most of the approved covalent BTK inhibitors are associated with treatment limitations due to off-target toxicity and drug resistance. Developing noncovalent BTK inhibitors is a promising strategy to address unmet clinical needs. Here, a novel series of pyrrolo[1,2-a]quinoxalin-4(5H)-one derivatives were designed and synthesized as noncovalent BTK inhibitors. Among them, representative compound 9 exhibited potent BTK inhibitory activity (IC₅₀ = 21.6 nM) and excellent selectivity against a panel of 468 kinases. Moreover, the oral exposure property of compound 9 was improved, and the antitumor efficacy of compound 9 (TGI = 64.4%) was superior to the lead S2 (TGI = 28.7%) and Ibrutinib (TGI = 41.1%) in the U-937 xenograft models at an oral dosage of 50 mg/kg. All these results suggest that compound 9 is a potent, selective, and orally available noncovalent BTK inhibitor worthy of further development.

In the synthetic laboratory, researchers typically rely on nuclear magnetic resonance (NMR) spectra to elucidate structures of synthesized products and confirm whether they match the desired target compounds. As chemical synthesis technology evolves toward intelligence and continuity, efficient computer-assisted structure elucidation (CASE) techniques are required to replace time-consuming manual analysis and provide the necessary speed. However, current CASE methods typically aim to derive precise chemical structures from spectroscopic data, yet they suffer from drawbacks such as low accuracy, high computational cost, and reliance on chemical libraries. In meticulously designed chemical synthesis reactions, researchers prioritize confirming the attainment of the target product based on NMR spectra, rather than focusing on identifying the specific product obtained. For this purpose, we innovatively developed a binary classification model, termed as MatCS, to directly predict the relationship between NMR spectra image (including ¹H NMR and ¹³C NMR) and the molecular structure of the target compound. After evaluating various feature extraction methods, MatCS employs a combination of the Graph Attention Networks and Graph Convolutional Networks to learn the structural features of molecular graphs and the pretrained ResNet101 network with a Convolutional Block Attention Module to extract features from NMR spectra images. The results show that on a challenging Test_sim data set, which poses difficulty in distinguishing spectra of similar molecular structures, MatCS achieves comprehensive evaluation metrics with an F1-score of 0.81 and an AUC value of 0.87. Simultaneously, it exhibited commendable performance on an external SDBS data set containing experimental NMR spectra, showcasing substantial potential for structural verification tasks in real automated chemical synthesis.

Construction of high-quality fragment libraries by segmenting organic compounds is an important part of the drug discovery paradigm. This article presents a new method, MacFrag, for efficient molecule fragmentation. MacFrag utilized a modified version of BRICS rules to break chemical bonds, and introduced an efficient subgraphs extraction algorithm for rapid enumeration of the fragment space. The evaluation results with ChEMBL dataset exhibited that MacFrag was overall faster than BRICS implementd in RDKit and modified molBLOCKS. Meanwhile, the fragments acquired through MacFrag were more compliant with the “Rule of Three”. https://github.com/yydiao1025/MacFrag Supplementary data are available at Bioinformatics online.