Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy最新文献

{"title":"Flower-like FeMoO4 nanoparticles as a peroxidase mimic for sensitive colorimetric immunoassay of CEA","authors":"Yuwei Pan ,&nbsp;Xiaoyan Deng ,&nbsp;Rong Zhang ,&nbsp;Liuyan You ,&nbsp;Jiaqing Guo ,&nbsp;Youxiu Lin ,&nbsp;Dianping Tang ,&nbsp;Wenqiang Lai","doi":"10.1016/j.saa.2026.127468","DOIUrl":"10.1016/j.saa.2026.127468","url":null,"abstract":"<div><div>Early and accessible detection of tumor biomarkers requires robust colorimetric readouts that can operate in complex matrices. We report flower-like hierarchical FeMoO₄ nanoparticles functioning as a peroxidase-like nanozyme and enable a cascade immunoassay for carcinoembryonic antigen. The nanoparticles were obtained by a solvothermal route and exhibit a single crystalline FeMoO₄ phase with uniform Fe, Mo, and O distributions, mixed Fe³⁺ and Fe²⁺ states, and a measurable fraction of surface non-lattice oxygen species (O_ads, including hydroxyl-related oxygen and adsorbed/weakly bound oxygen). These features support efficient H₂O₂ activation and TMB oxidation. FeMoO₄ was integrated with a glucose oxidase (GOx)-based enzyme signal reporter, in which GOx and the detection antibody (Ab2) were co-immobilized on gold nanoparticles (AuNPs) to form GOx-AuNP-Ab2 conjugate. In a sandwich immunoassay format, the captured enzyme-labeled conjugate generates H₂O₂ in proportion to the antigen concentration, which is subsequently converted by FeMoO₄ into a stable colorimetric signal at 652 nm. Reaction conditions were optimized for both the chromogenic branch and the enzymatic branch, and kinetic analysis gave Michaelis-Menten behavior with <em>K</em>_m values of 0.78 mM for H₂O₂ and 0.40 mM for TMB. Radical scavenging identified superoxide and hydroxyl radicals as the main active species, with singlet oxygen as a minor contributor. The CEA assay achieved a linear range of 0.1 to 60 ng mL⁻¹ and a limit of detection of 63 pg mL⁻¹, with a one month retention of 93% of the initial absorbance. Results for six clinical serum samples showed good agreement with a commercial ELISA at the 95% confidence level, and one CEA-negative serum was included as a negative control. These findings establish FeMoO₄ as a stable and manufacturable nanozyme platform for colorimetric immunoassays.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"351 ","pages":"Article 127468"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A research on applying the diffusion model algorithm for Infrared and Raman spectroscopy data augmentation to improve the accuracy of diseases","authors":"Xiangnan Chen ,&nbsp;Cheng Chen ,&nbsp;Xuguang Zhou ,&nbsp;Xiaoyi Lv ,&nbsp;Chen Chen","doi":"10.1016/j.saa.2026.127466","DOIUrl":"10.1016/j.saa.2026.127466","url":null,"abstract":"<div><div>In recent years, Raman and Infrared spectroscopy have become important tools in disease diagnosis due to their high sensitivity and non-invasive detection advantages. Combined with deep learning methods, spectral data can achieve high-precision identification of a wide range of diseases. However, deep learning models rely on large amounts of high-quality data. In practice, spectral data often faces challenges such as limited sample size, noise interference, and device variability, leading to model overfitting and poor generalization. To overcome this bottleneck, this paper proposes a spectral data generation method based on a diffusion model. By encoding temporal and category label information separately and combining it with a multi-head attention mechanism, the method extracts the overall morphology and local subtle features of the spectrum at multiple scales. In the reverse reconstruction stage, the noise distribution is estimated based on implicit modeling and cross-attention is used to generate spectra of different categories under different labels. This method achieves accurate denoising under conditional constraints and progressively reconstructs information such as the positions of characteristic spectral peaks. This method generates high-quality, continuous spectral data, effectively enhancing the scale and quality of the dataset. Experimental results demonstrate that the proposed method can generate high-fidelity and diverse synthetic spectral samples. The Pearson correlation coefficients for the thyroid and SLE infrared datasets reached high values of 0.97719 and 0.9914, respectively, and the similarities for the Thyroid and SLE Raman datasets reached high values of 0.9531 and 0.9747, respectively. This method effectively expands the training dataset and alleviates data scarcity and uneven distribution. The diffusion model was applied to two disease diagnosis tasks for validation. Under the condition of optimal gain ratio,the model trained with augmented data significantly improved generalization performance and diagnostic accuracy. In the infrared spectroscopy classification task, the accuracy of SLE disease on EfficientNet improved from 69.70% to 90.91%, and the accuracy of benign and malignant thyroid tumors on EfficientNet improved from 66.67% to 89.47%. In the Raman spectroscopy classification task, the accuracy of SLE disease on MLP improved from 87.88% to 90.91%, and the accuracy of benign and malignant thyroid tumors on the Transformer model improved from 87.22% to 92.98%.This study provides an effective generative augmentation framework for small-sample spectral data analysis, with strong theoretical value and application prospects.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"351 ","pages":"Article 127466"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146013913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Ratiometric genetically encoded fluorescent probe for ultrasensitive and self-calibrated detection of Hg2+","authors":"Dan Wang ,&nbsp;Zixin Wu ,&nbsp;Tianyu Song ,&nbsp;Qunfang Li ,&nbsp;Cuiling Zeng ,&nbsp;You Wang ,&nbsp;Xiaosu Fan ,&nbsp;Pengyun Liang","doi":"10.1016/j.saa.2026.127502","DOIUrl":"10.1016/j.saa.2026.127502","url":null,"abstract":"<div><div>Mercury ion (Hg²⁺) pollution poses severe environmental and health risks, necessitating the development of highly sensitive and reliable detection methods. Although ratiometric and genetically encoded fluorescent probes each offer distinct advantages, their integration for Hg²⁺ sensing remains largely unexplored. Herein, we report the rational design of a novel ratiometric genetically encoded fluorescent probe through the fusion of a circularly permuted green fluorescent protein (cpEGFP), a mercury-binding domain (MerBD), and a large Stokes shift reference fluorescent protein (LSSmOrange). This probe enables self-calibrated Hg²⁺ quantification by measuring the fluorescence intensity ratio of the response channel (F₅₁₂) to the stable reference channel (F₅₇₂), which effectively minimizes interferences from probe concentration, environmental fluctuations, and instrumental variations. The probe exhibits an ultra-high affinity for Hg²⁺ (apparent K_d' = 2.71 × 10⁻¹³ M) and a detection limit of 4 nM, alongside long-term stability and high selectivity against most common metal ions. Competitive titration, circular dichroism, and fluorescence lifetime analyses reveal that the probe operates through a cooperative conformational-change mechanism, which translates picomolar-level binding into a nanomolar-level fluorescence response. This work not only addresses a critical gap in ratiometric genetically encoded probes for Hg²⁺ but also provides a robust and versatile platform for accurate Hg²⁺ monitoring in environmental and biological systems.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"351 ","pages":"Article 127502"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146055791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Covalent organic polymer cascade MnO2 nanosheets-based fluorescence-colorimetric dual-mode sensing system for highly sensitive detection of organophosphorus pesticides","authors":"Suyu Li,&nbsp;Yuzhen Wang,&nbsp;Huilin Liu,&nbsp;Baoguo Sun","doi":"10.1016/j.saa.2026.127431","DOIUrl":"10.1016/j.saa.2026.127431","url":null,"abstract":"<div><div>To enable accurate visual identification of organophosphorus pesticides (OPs) in raw materials of medicinal and food homology (MFH), this study used the red-fluorescent covalent organic polymer TEPBY-DTDA to cascade MnO₂ nanosheets exhibiting oxidase-mimetic activity and suppressing fluorescence effects<strong>,</strong> enabling colorimetric and ratiometric fluorescence detection. OPs suppressed acetylcholinesterase (AChE) efficiency, consequently diminishing thiocholine (TCh) production and modulating residual MnO₂ nanosheet levels. Rising OP concentrations caused the probe's fluorescence to shift red-to-yellow, accompanied by solution yellowing. Using triazophos, methyl parathion (MP), chlorpyrifos, and phoxim as representative OPs, the probes generated a robust, linear signal over the concentration interval of 0.1–1500 μg/L. The fluorescence minimal limit of detection (LOD) was 0.058 μg/L, while the minimum LOD in colorimetric mode was 0.04 μg/L. The probe demonstrated excellent selectivity and strong anti-interference capability against OPs. Consequently, this sensing strategy not only expands the toolbox for quality monitoring of MFH materials but also holds significant potential for enhancing food safety protocols.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127431"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145936380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simultaneous determination of amoxicillin and clavulanic acid using MPA-capped AgInS₂ quantum dots: A fluorescence-based chemometric approach","authors":"Rafael C. Castro,&nbsp;Ricardo N.M.J. Páscoa,&nbsp;João L.M. Santos,&nbsp;David S.M. Ribeiro","doi":"10.1016/j.saa.2026.127457","DOIUrl":"10.1016/j.saa.2026.127457","url":null,"abstract":"<div><div>Due to the widespread clinical use of their co-formulated form, the simultaneous and accurate determination of amoxicillin (AMX) and clavulanic acid (CLA) in pharmaceutical formulations is essential, calling for the development of simple, rapid, cost-effective, and reliable analytical methodologies. Therefore, an innovative and novel spectrofluorimetric method was developed for the simultaneous determination of both AMX and CLA in pharmaceutical formulations based on a multiplexed detection, using AgInS₂ quantum dots (QDs) capped with mercaptopropionic acid (MPA) as a PL probe, combined with chemometrics. Several QDs were synthesised and evaluated to achieve the highest sensitivity towards both analytes. Different interaction mechanisms were observed for each analyte. A full factorial experimental design was employed to ensure robust results and assess the main effects and their interactions of both analytes. The acquired kinetic fluorescence data were modelled using an unfolded partial least squares (U-PLS) model. The optimisation of the U-PLS models involved the study of different spectral regions. This methodology revealed that the entire spectral region yielded the best results, with accuracies of 2 % and 11 % for AMX and CLA, respectively, in terms of the relative error of prediction (REP). Moreover, this method was compared with the official USP chromatographic method, and no statistically significant differences were observed, thereby validating the precision and reliability of the developed method. This simple, rapid, and cost-effective analytical methodology can serve as a valuable alternative to the official chromatographic method, eliminating the need for toxic solvents.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127457"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145940402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Combination of markov transition field and multi-scale feature extraction for cetane number prediction in diesel using near-infrared spectroscopy","authors":"Wei Luo ,&nbsp;Wenyoujia Li ,&nbsp;Haihua Huang ,&nbsp;Zhijian Qu ,&nbsp;Yanfang Gao ,&nbsp;Weifa Peng ,&nbsp;Hailiang Zhang ,&nbsp;Guozhu Fan","doi":"10.1016/j.saa.2025.127381","DOIUrl":"10.1016/j.saa.2025.127381","url":null,"abstract":"<div><div>Diesel, a crucial energy source for transportation and industrial applications. The cetane number (CN) plays a key role in assessing diesel quality and combustion performance. In this study, near-infrared spectroscopy (NIRS) data of diesel were encoded into images using the Markov transition field (MTF). A hybrid CNN-BiLSTM model with multi-head attention was developed to achieve multimodal feature extraction and fusion by simultaneously inputting spectral data and MTF images, enabling accurate CN prediction. For comparison, PLSR, 1D-CNN, and 2D-CNN models were established using feature wavelengths selected by variable combination population analysis (VCPA), SG1-preprocessed full spectra, and MTF images, respectively. Results demonstrated that the CNN-BiLSTM achieved the best performance (R_p² = 0.9824, RMSEP = 0.3106), whereas the VCPA-PLSR performed the worst (R_p² = 0.7541, RMSEP = 1.6425). The 1D-CNN outperformed the VCPA-PLSR, and the 2D-CNN further improved upon the 1D-CNN. This study demonstrates that converting NIRS data into images via MTF effectively leverages the potential of CNNs and improves CN prediction performance without feature wavelength extraction, while the introduced BiLSTM channel mitigates the distortion of original data caused by MTF encoding. The approach provides a non-destructive and reliable framework for liquid samples' quality evaluation in industrial applications.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127381"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145919593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A dual-responsive ratiometric fluorescence sensor for highly sensitive detection of heparin with Eu CPs and fluorescent polydopamine nanoparticles","authors":"Yan-Zhao Fang ,&nbsp;Yan Zhang ,&nbsp;Lin-Xin Du ,&nbsp;He-Han Huang ,&nbsp;Zhi-Jie Mao ,&nbsp;Pei Zhang ,&nbsp;Shuang Wu ,&nbsp;Xiang-Juan Kong ,&nbsp;Qiang Xiao","doi":"10.1016/j.saa.2025.127402","DOIUrl":"10.1016/j.saa.2025.127402","url":null,"abstract":"<div><div>In this study, a new dual-responsive ratiometric fluorescence nanoprobe was rationally designed for heparin (Hep) detection. The negatively charged europium-based coordination polymers (Eu CPs) and positively charged fluorescent polydopamine nanoparticles (FPDA NPs) could assemble via electrostatic interaction, meanwhile, the fluorescence of these both reduced slightly owing to self-quenching (SQ) and inner-filter effect (IFE), respectively. Upon addition of Hep, the Eu CPs emission at 615 nm was markedly quenched due to electron transfer and SQ, while the FPDA NPs emission at 465 nm was enhanced marginally owing to mitigation of the IFE. These opposing signal changes yielded a robust ratiometric readout (I₄₆₅/I₆₁₅) for Hep detection. The probe showed a widely linear response over 0–16 μg/mL with a detection limit as low as 3.58 ng/mL. Furthermore, high sensitivity and accuracy of the probe enabled accurate analysis of Hep in human serum with favorable recoveries, highlighting its potential for biomedical analysis and clinical diagnostics.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127402"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Smart three-in-one detection platform: Chemical sensing, image recognition, and machine learning for rapid identification of tetracycline antibiotics","authors":"Haoyuan Ma ,&nbsp;Lei Jia ,&nbsp;Jiaying Li ,&nbsp;Lina Zhang ,&nbsp;Yongxin Li ,&nbsp;Taofeng Zhu ,&nbsp;Jun Xu","doi":"10.1016/j.saa.2025.127416","DOIUrl":"10.1016/j.saa.2025.127416","url":null,"abstract":"<div><div>The hidden allergic reactions and cumulative toxic effects caused by antibiotic residues in aquatic products have become significant public health concerns. Among these, tetracycline antibiotics (TCs) pose considerable ecological risks and health threats due to their high molecular stability and prolonged environmental persistence. To address this challenge, this study developed a guanosine-5-monophosphate (GMP)-Tb/rhodamine B (RhB) multi-proportion ratiometric fluorescence probe array, integrated with computer vision and an intelligent color mapping recognition framework, to construct a comprehensive and intelligent solution for antibiotic identification. Specifically, the “You Only Look Once” version 5 s (YOLOv5s) model was utilized to automatically detect the reaction areas on the 96-well plate and extract RGB feature data. These features were then analyzed using k-Nearest Neighbor(KNN), support vector machines (SVM), and random forest(RF) classifiers for multi-dimensional modeling. Finally, the optimized model was integrated into a web application built using the Flask framework. Experimental results demonstrated that the system achieved an overall recognition accuracy of 94.4 % for TC, OTC, and CTC antibiotics within a concentration range of 0–50 μM; notably, the optimal KNN classifier achieved an F1-score of 1.00 on the validation set. This study successfully established a three-in-one analytical paradigm that integrates chemical sensing, image recognition, and intelligent decision-making, thereby providing an expandable technical platform for the rapid screening of environmental pollutants.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127416"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Human neuroglobin and H64A distal variant: How mutation and pH affect the heme pocket","authors":"Mirco Meglioli ,&nbsp;Laura Cesati ,&nbsp;Gianantonio Battistuzzi ,&nbsp;Marco Borsari ,&nbsp;Federico Sebastiani ,&nbsp;Giulietta Smulevich","doi":"10.1016/j.saa.2025.127401","DOIUrl":"10.1016/j.saa.2025.127401","url":null,"abstract":"<div><div>Neuroglobins (Ngbs) are low spin heme proteins, with His64 and His96 residues bound to the distal and proximal positions of the heme iron, respectively. His64 must be displaced to allow the ligation of exogenous molecules to the heme, thus regulating ligand binding. In fact, in the absence of the distal His, at physiological pH, the protein becomes high spin in both the ferric and ferrous states. However, the variant undergoes a pH-dependent coordination/spin state transition to a hexacoordinated low-spin species at alkaline pH, featuring Lys67 as the distal endogenous ligand. Here, we have investigated the effect of pH on the wild-type (WT) protein and the distal H64A variant of human Ngb in solution by a combination of UV–Vis electronic absorption, Magnetic Circular Dichroism and resonant Raman spectroscopies. In the WT protein, only a vinyl substituent conformational change is observed with increasing pH, due to the strong stability of the bis-histidyl heme coordination. Conversely, in the H64A variant alkaline pH induces a major structural rearrangement of the active site with the distal coordination of Lys67 to the heme, a change in the strength of the hydrogen bonds of the propionate groups and a conformational rearrangement of one vinyl substituent. We propose that these findings are due to the modifications in the <em>E</em>-helix as a consequence of Lys67 deprotonation in both the WT and the variant. These results provide further details at the molecular level on the fundamental role of distal residues for heme binding within the active site and protein stability.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127401"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145936308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the Accuracy of Partial Least Square Regression for Distillation Fractions from ATR-IR Spectra of Crude Oils, with a Focus on Selected Crudes and Fractions","authors":"Slobodan Šašić,&nbsp;Connor Grant,&nbsp;Roy Mize,&nbsp;Deepak Patel,&nbsp;Maryro Mendez","doi":"10.1016/j.saa.2025.127408","DOIUrl":"10.1016/j.saa.2025.127408","url":null,"abstract":"<div><div>Optimization software used for automated generation of partial-least-square (PLS) regression models for the distillation fractions of crude oils based on ATR-IR (attenuated total reflection infrared) spectra often yields multiple models with comparable accuracy and complexity. Selecting models solely based on highest accuracy (lowest cross-validation errors) may not always be the optimal analytical choice for certain crudes. If these crudes hold significant commercial importance, it is crucial to address suboptimal prediction quality through model updates and careful selection. This study focuses on two important crude oils, which experience these issues, and two specific fractions, 550 °C-fbp (final boiling point) and 232–342 °C, that exhibited the highest prediction errors. Updating the models with additional spectra from available crudes, and specifically adding the spectra of the two mentioned crudes, did not show any improvement as far as cross validation errors. The models with the lowest cross-validation errors were further evaluated based on their performance using previously acquired spectra of the two crudes in question. This analysis, weighted heavily by those crudes, reveals substantial differences in prediction errors among the updated models and compared to several models currently in use. Consequently, judicious selection of models for future predictions should consider both generally low cross-validation error and retroactive performance on high-volume, commercially important crudes.</div></div>","PeriodicalId":433,"journal":{"name":"Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy","volume":"350 ","pages":"Article 127408"},"PeriodicalIF":4.6,"publicationDate":"2026-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145936211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}