G protein-coupled receptors (GPCRs) play vital roles in various physiological processes, making them attractive drug discovery targets. Meanwhile, deep learning techniques have revolutionized drug discovery by facilitating efficient tools for expediting the identification and optimization of ligands. However, existing models for the GPCRs often focus on single-target or a small subset of GPCRs or employ binary classification, constraining their applicability for high throughput virtual screening. To address these issues, we introduce AiGPro, a novel multitask model designed to predict small molecule agonists (EC50) and antagonists (IC50) across the 231 human GPCRs, making it a first-in-class solution for large-scale GPCR profiling. Leveraging multi-scale context aggregation and bidirectional multi-head cross-attention mechanisms, our approach demonstrates that ensemble models may not be necessary for predicting complex GPCR states and small molecule interactions. Through extensive validation using stratified tenfold cross-validation, AiGPro achieves robust performance with Pearson's correlation coefficient of 0.91, indicating broad generalizability. This breakthrough sets a new standard in the GPCR studies, outperforming previous studies. Moreover, our first-in-class multi-tasking model can predict agonist and antagonist activities across a wide range of GPCRs, offering a comprehensive perspective on ligand bioactivity within this diverse superfamily. To facilitate easy accessibility, we have deployed a web-based platform for model access at https://aicadd.ssu.ac.kr/AiGPro . Scientific Contribution We introduce a deep learning-based multi-task model to generalize the agonist and antagonist bioactivity prediction for GPCRs accurately. The model is implemented on a user-friendly web server to facilitate rapid screening of small-molecule libraries, expediting GPCR-targeted drug discovery. Covering a diverse set of 231 GPCR targets, the platform delivers a robust, scalable solution for advancing GPCR-focused therapeutic development. The proposed framework incorporates an innovative dual-label prediction strategy, enabling the simultaneous classification of molecules as agonists, antagonists, or both. Each prediction is further accompanied by a confidence score, offering a quantitative measure of activity likelihood. This advancement moves beyond conventional models focusing solely on binding affinity, providing a more comprehensive understanding of ligand-receptor interactions. At the core of our model lies the Bi-Directional Multi-Head Cross-Attention (BMCA) module, a novel architecture that captures forward and backward contextual embeddings of protein and ligand features. By leveraging BMCA, the model effectively integrates structural and sequence-level information, ensuring a precise representation of molecular interactions. Results show that this approach is highly accurate in binding affinity predictions and consistent across diverse GPCR families. By unifying agonist and antagonist bioactivity prediction into a single model architecture, we bridge a critical gap in GPCR modeling. This enhances prediction accuracy and accelerates virtual screening workflows, offering a valuable and innovative solution for advancing GPCR-targeted drug discovery.

Unlike conventional fluorophores experiencing aggregation-caused quenching (ACQ) in the solid state, aggregation-induced emission (AIE) luminogens (AIEgens) exhibit enhanced fluorescence upon aggregation. This property renders AIEgens promising for various technological applications. However, developing red-emitting AIEgens with high fluorescence efficiency remains challenging due to structural constraints. Compact red-light-emitting AIEgens hold great potential, as their small, simple fluorescent frameworks enable facile modification and precise color tuning. This study presents a novel compact indolizine-based AIE system with solid-state emission tunable from cyan (496 nm) to deep red (669 nm), with an average molecular weight (MW) of 312.8 g/mol. The deep-red-emitting derivative is uncommon among short-conjugation, low-molecular-weight (∼280 g/mol) systems and demonstrates exceptional photophysical characteristics. Through carbonyl functional groups, we modulated molecular packing, resulting in distinct optical properties and emission tunability in solid and aggregated states. Structural investigations using single-crystal X-ray diffraction (SCXRD) and computational modeling revealed that molecular packing and intermolecular interactions significantly influence emission behavior. Integration of these AIEgens into blue-light-emitting diode chips facilitated white-light production with CIE coordinates (0.360, 0.362) and a color temperature of 4533 K. These findings underscore the potential of indolizine-based AIEgens for next-generation emissive materials, providing insights for their rational design and broader technological applications.

The emergence of cancer resistance to targeted therapy represents a significant challenge in cancer treatment. Therefore, identifying new anticancer candidates, particularly those addressing oncogenic mutants, is an urgent medical demand. A campaign of structural modifications has been conducted to further optimize our previously reported 2-anilinoquinoline-diarylamides conjugate VII as a B-RAFV600E/C-RAF inhibitor. Considering the incorporation of a methylene bridge between the terminal phenyl and cyclic diamine, focused quinoline-based arylamides have been tailored, synthesized, and biologically evaluated. Among them, the 5/6-hydroxyquinolines 17b and 18a stood out as the most potent members, with IC₅₀ values of 0.128 µM, 0.114 µM against B-RAF^V600E, and 0.0653 µM, 0.0676 µM against C-RAF. Most importantly, 17b elicited remarkable inhibitory potency against the clinically resistant B-RAF^V600K mutant with an IC₅₀ value of 0.0616 µM. The putative binding mode of 17b and 18a were studied by molecular docking and molecular dynamics (MD). Moreover, the antiproliferative activity of all target compounds has been examined over a panel of NCI-60 human cancer cell lines. In agreement with cell-free assays, the designed compounds exerted superior anticancer impact over the lead quinoline VII against all cell lines at a 10 µM dose. Notably, both 17b and 18b showed highly potent antiproliferative activity against melanoma cell lines with growth percent under −90% (SK-MEL-29, SK-MEL-5, and UACC-62) at a single dose, while 17b maintained potency with GI₅₀ values of 1.60–1.89 µM against melanoma cell lines. Taken together, 17b, a promising B-RAF^V600E/V600K and C-RAF kinase inhibitor, may serve as a valuable candidate in the arsenal of anticancer chemotherapeutics.

A modified InChI (International Chemical Identifier) string scheme, yaInChI (yet another InChI), is suggested as a method for including the structural information of a given molecule, making it straightforward and more easily readable. The yaInChI theme is applicable for checking the structural identity with higher sensitivity and generating three-dimensional (3-D) structures from the one-dimensional (1-D) string with less ambiguity than the general InChI method. The modifications to yaInChI provide non-rotatable single bonds, stereochemistry of organometallic compounds, allene and cumulene, and parity of atoms with a lone pair. Additionally, yaInChI better preserves the original information of the given input file (SDF) using the protonation information, hydrogen count +1, and original bond type, which are not considered or restrictively considered in InChI and SMILES. When yaInChI is used to perform a duplication check on a 3D chemical structure database, Ligand.Info, it shows more discriminating power than InChI. The structural information provided by yaInChI is in a compact format, making it a promising solution for handling large chemical structure databases.