Special Issue for Celebrating Prof. Ruqin Yu's 90th Birthday

Prof. Ruqin Yu, born in November 1935, is a distinguished chemometrician and an academician of the Chinese Academy of Sciences. After graduating from the Department of Chemistry, St. Petersburg University, he pursued further research under Academician Shuquan Liang at the Institute of Chemistry, Chinese Academy of Sciences. Prof. Yu has made pioneering contributions to chemometrics, introducing morphological approaches and chaos concepts into algorithm design, and laying the theoretical foundation for robust chemometric multi-way resolution methods. Over his prolific career, he has published more than 1200 papers and received three National Natural Science Awards. He was awarded the Chemometrics Lifetime Achievement Prize in 2015. Besides, the 2016 Prize was awarded to another Chinese chemometrician Yizeng Liang, a former Prof. Yu's student who defended his PhD thesis under Yu's supervision in 1988. As a long-standing editor of the Journal of Chemometrics, Prof. Yu has played a vital role in shaping the development of the field worldwide, particularly fostering the growth of chemometrics in China. This special issue is dedicated to celebrating Prof. Yu's 90th birthday, paying tribute to his lifelong achievements and enduring influence on chemometrics and its community. The collected works not only highlight recent advances in chemometric theory and applications but also reflect the vibrancy and diversity of research in China, much of which has been inspired by Prof. Yu's vision and guidance. The contributions in this issue span a wide range of chemometric methodologies, including spectral analysis, chromatographic data processing, data preprocessing and variable selection, and machine learning and deep learning approaches. The first element to emerge from papers within this special issue was the inclusion of two comprehensive reviews. One review focuses on the application of near-infrared spectroscopy combined with chemometric methods to explore water structures in chemical and biological systems, illustrating how chemometrics enables the resolution of subtle spectral features and reveals molecular interactions [1]. The other review provides a panoramic overview of process analysis chemistry based on modern spectroscopies such as infrared, Raman, and LIBS, and summarizes methodologies including preprocessing, feature selection, modeling, and optimization for process monitoring and control [2]. Together, these reviews demonstrate the indispensable role of chemometrics in both fundamental structural studies and practical process analysis. The second element relates to the analysis of complex chromatographic and metabolomics data. Chemometric strategies for hyphenated data remain at the forefront of research, with multivariate curve resolution and multi-way calibration continuing to be recognized as core approaches. One study proposes a practical framework that integrates two-way and three-way methods to resolve severe peak overlaps and retention time shifts in chromatographic systems [3]. Another applies GC–MS-based untargeted metabolomics, coupled with advanced AntDAS-GCMS software, to evaluate how storage time influences the chemical profile of Artemisia argyi smoke [4]. Both papers highlight how multi-way chemometric methods remain powerful tools for extracting reliable information from complex systems. The third element concerns methodological advances in data augmentation and variable selection. One contribution introduces a discretized whale optimization algorithm for spectral variable selection, combined with extreme learning machines to improve the quantitative analysis of complex samples [5]. Another tackles the persistent problem of class imbalance in chemometric data by proposing HiBBKA, a hybrid framework that integrates advanced resampling strategies with heuristic feature selection to enhance classification performance, particularly for minority classes [6]. These studies exemplify how data augmentation and variable selection strategies can improve model robustness and predictive capability in chemometric applications. The final element highlights the growing role of deep learning and advanced AI methodologies as powerful tools within chemometrics. One study develops a one-class convolutional autoencoder for pattern recognition in excitation–emission matrix fluorescence data, achieving reliable detection of abnormal food samples [7]. Another introduces a voxel-projected area-based deep learning model for predicting collision cross sections in ion mobility spectrometry–mass spectrometry (IMS–MS), integrating geometric features with graph neural networks to deliver state-of-the-art accuracy and robustness [8]. Most notably, deep reinforcement learning is employed to drive de novo molecular generation for HIV-1 integrase inhibitors. This study demonstrates how emerging generative models can push chemometrics beyond data analysis toward active molecular discovery [9]. Collectively, these contributions demonstrate how deep learning has become an important component of the chemometric toolkit, enabling advances in pattern recognition, predictive modeling, and molecular design. In summary, this special issue brings together nine contributions that collectively represent both the continuity and the innovation of chemometrics. They demonstrate how classical foundations such as multivariate calibration and curve resolution continue to evolve, while new approaches including deep learning and generative modeling expand the boundaries of the field. We sincerely thank all authors and reviewers for their valuable efforts, and the editorial office for their professional support. Above all, we extend our warmest congratulations to Prof. Ruqin Yu on his 90th birthday, and we hope this commemorative issue will inspire further progress in chemometrics in China and worldwide. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.