Communications Chemistry
·
2024
·Vol. 7
·期 1
·1-11
·
IF 5.9
Yang Yiming,Sun Shuang,Yang Shuyuan,Yang Qin,Lu Xinqiong,Wang Xiaohao,Yu Quan,Huo Xinming,Qian Xiang
Structural annotation of small molecules in tandem mass spectrometry has always been a central challenge in mass spectrometry analysis, especially using a miniaturized mass spectrometer for on-site testing. Here, we propose the Transformer enabled Fragment Tree (TeFT) method, which combines various types of fragmentation tree models and a deep learning Transformer module. It is aimed to generate the specific structure of molecules de novo solely from mass spectrometry spectra. The evaluation results on different open-source databases indicated that the proposed model achieved remarkable results in that the majority of molecular structures of compounds in the test can be successfully recognized. Also, the TeFT has been validated on a miniaturized mass spectrometer with low-resolution spectra for 16 flavonoid alcohols, achieving complete structure prediction for 8 substances. Finally, TeFT confirmed the structure of the compound contained in a Chinese medicine substance called the Anweiyang capsule. These results indicate that the TeFT method is suitable for annotating fragmentation peaks with clear fragmentation rules, particularly when applied to on-site mass spectrometry with lower mass resolution.
研究领域:
分析化学
·
人工智能
·
生物信息学
✓ 已验证
DOI
PubMed
高影响因子
MICROCHEMICAL JOURNAL
·
2025
·
IF 5.1
Ying Yang,Liang Qi,Guanhua Xu,Zheng Li,Yi Tao
Ethanol precipitation is a pivotal manufacturing step in the production of Guanxinning tablets (GXN). However, current quality control methods for this process remain inadequate. To address this gap, we propose a novel analytical framework that combines Raman spectroscopy with state-of-the-art deep learning architectures, specifically integrating a Mamba State Space model with a Transformer model, to enable non-destructive and real-time monitoring of seven critical quality attributes (CQAs) throughout the ethanol precipitation process of GXN. A total of 160 ethanol precipitation samples were collected, and their Raman spectra were acquired using a 785 nm laser (35 mW) across a wavenumber range of 200–3200 cm −1 , enabling rapid spectral acquisition within 1 s per sample. The content of five bioactive compounds (danshensu, ligustrazine, ferulic acid, rosmarinic acid, and salvianolic acid B) was quantified via HPLC, while solid content and alcohol content were determined using standard methods. Spectral outliers were removed using PCA-Mahalanobis distance analysis, and the Kennard-Stone algorithm was employed to partition datasets into training and test sets (4:1 ratio) to ensure model generalizability. Comparative model evaluation revealed that Mamba State Space model outperformed Transformer model and linear models, achieving prediction coefficients ( R p 2 ) of 0.9419–0.9978 and root mean square error of prediction (RMSEP) values of 0.0129–0.0501, with exceptional accuracy for solid content ( R p 2 > 0.99). This approach reduces reliance on time-consuming HPLC analytical method, offering a 99.9 % reduction in analysis time while maintaining predictive accuracy (RMSEP ≤0.0501) and robustness ( R p 2 > 0.94). By embedding the Mamba State Space model into Raman spectroscopy-based ethanol precipitation process monitoring, this study establishes a robust quality by artificial intelligence framework for real-time quality control in traditional Chinese medicine (TCM) manufacturing, paving the way for smart process analytical technology (PAT) applications.
研究领域:
分析化学
·
制药工程
·
中药学
·
人工智能
✓ 已验证
产品关联
DOI
高影响因子
Cell Reports Medicine
·
2025
·
IF 10.6
Dongjie Zhang,Zhaoyang Cheng,Yali Song,Huandi Li,Lin Shi,Nan Wang,Yingwen Peng,Renan Chen,Nianzheng Sun,Min Han,Fengjiao Hu,Chuntao Zong,Rui Zhang,Si Chen,Conghui Zhu,Xiaoli Zhang,Xiaobo Li,Xiaopeng Ma,Changbei Shi,Xiaofei Zhang,Rui Liu,Ziqi Ren,Lin Wang,Qi Zeng,Tingting Zeng,Xueli Chen
Rapid identification and accurate diagnosis are critical for individuals with acute leukemia (AL). Here, we propose a combined deep learning and surface-enhanced Raman scattering (DL-SERS) classification strategy to achieve rapid and sensitive identification of AL with various subtypes and genetic abnormalities. More than 390 of cerebrospinal fluid (CSF) samples are collected as targets, encompassing healthy control, AL patients, and individuals with other diseases. Sensitive SERS detection could be achieved within 5 min, using only 0.5 μL volume of CSF. Through integrated feature fusion (1D spectra and 2D image) with a transformer model, the classification method is developed to screen and diagnose AL patients, demonstrating exceptional classification performances of accuracy, sensitivity, specificity, or reliability. Also, this approach demonstrates remarkable versatility and could be extended to the classifications of meningitis diseases. The sensitive DL-SERS classification platform has the potential to be a powerful auxiliary in vitro diagnostic tool.
研究领域:
生物医学工程
·
肿瘤学
·
医学人工智能
·
分析化学
·
分子诊断
✓ 已验证
产品关联
DOI
PubMed
高影响因子
MICROCHEMICAL JOURNAL
·
2025
·
IF 5.1
Zengshan Yu,Xuemei Lin,Zelong Li,Shan Guo,Jiahao Cui,Jianjun Gu,Mingli Wang,Guochao Shi
Carcinoembryonic antigen (CEA), a critical colorectal cancer biomarker, playing a pivotal role in the early diagnosis, therapeutic intervention and longitudinal monitoring of malignancies. Conventional detection methods such as ELISA exhibit inherent limitations including inadequate analytical sensitivity and prolonged assay duration. This study proposes the development of a SERS-based immunosensor integrated with artificial intelligence (AI) frameworks, aiming to achieve precise and quantitative detection of CEA. The antibody-conjugated gold nanoprobes (Au NPs/4-MBA/Ab 1 ) were immobilized onto the surface of another antibody-functionalized Ag/Cu/PSS substrate (Ab 2 /Ag/Cu/PSS) via specific biorecognition. This sandwich-structured immunosensor, engineered with blockage of non-specific binding sites, demonstrated enhanced capacity to capture CEA antigens through dual antibody recognition layers. The immunosensor enabled efficient amplification of SERS signals and exhibited excellent linearity within the CEA concentration range of 0.1–1000 ng/mL. Importantly, Support Vector Machines (SVM), Convolutional Neural Networks (CNN) and Transformer frameworks, were employed for hierarchical processing of CEA Raman data. The Transformer model demonstrated superior performance with a classification accuracy of 99.06 % and a low loss value of 0.1056. These results underscore the model's exceptional capability to discriminate CEA concentrations with high precision and robustness. The integration of AI-driven signal processing boosts analytical sensitivity of SERS-based immunosensors. This synergy between nanophotonic sensing and deep learning algorithms represents a transformative leap in early cancer detection via non-invasive biofluid analysis.
研究领域:
生物传感器
·
纳米光子学
·
癌症诊断
·
人工智能
·
免疫传感器
·
分析化学
·
生物医学工程
·
机器学习
·
早期疾病检测
·
生物医学信号处理
✓ 已验证
产品关联
DOI
高影响因子
危险品化学品经营许可证（带存储）