In situ identification of toxin-producing Clostridioides difficile in stool samples based on single-cell Raman spectroscopy

Frontiers in Cellular and Infection Microbiology, Journal Year: 2025, Volume and Issue: 15

Published: May 19, 2025

Clostridioides difficile (CD) has emerged as one of the most prevalent nosocomial infections in hospitals and is primary causative agent antibiotic-associated diarrhea pseudomembranous colitis. In recent years, C. -induced have resulted significant morbidity mortality worldwide, with a particularly rapid increase incidence observed China. strains are categorized into toxin-producing non-toxin-producing based on their ability to synthesize toxins, pathogenicity being strictly dependent protein toxins produced by strains. Therefore, early identification crucial for diagnosis prevention infection (CDI). Currently, detection methods carried out clinical laboratories China mainly include culture, cell culture toxin assay, assay immunological methods, glutamate dehydrogenase (GDH) nucleic acid amplification assay.However, current CDI exhibit limitations, such time-consuming, operationally complex, lacking specificity sensitivity. Raman microspectroscopy been shown potential reliable microbial diagnostics, method reducing time results less than 1 hour, including processing samples, measurement single-cell spectra, final through use training models. this study, we aimed predict situ strain virulent 24 raw stool samples constructing reference spectroscopy (SCRS) database common intestinal flora , well SCRS The showed that accuracy was 83%, prediction 80%. These findings suggest may be viable non-virulent holds promise application CDI.

Language: Английский

---

Machine Learning Approaches in Label-Free Small Extracellular Vesicles Analysis with Surface-Enhanced Raman Scattering (SERS) for Cancer Diagnostics

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 22, 2024

Abstract Early diagnosis remains of pivotal importance in reducing patient morbidity and mortality cancer. To this end, liquid biopsy is emerging as a tool to perform broad cancer screenings. Small extracellular vesicles (sEVs), also called exosomes, found bodily fluids can serve important biomarkers these Our group has recently developed label-free electrokinetic microchip purify sEVs from blood. Herein, we demonstrate the feasibility integrate approach with surface-enhanced Raman scattering (SERS) analysis. SERS be used characterized extracted through their vibrational fingerprint that changes depending on origin sEVs. While are not easily identified spectra, they modeled machine learning (ML) approaches. Common ML approaches field spectral analysis use dimensionality reduction method often function black box. avoid pitfall, Shapley additive explanations (SHAP) type explainable AI (XAI) bridges models human comprehension by calculating specific contribution individual features model’s predictions, directly correlating model/decisions original data. Using demonstrated proof-of-concept model predictive isolated sEVs, integrating device SERS. This work explores diagnostic complex data clinical samples, while reporting interpretable biochemical information.

Language: Английский

---

Raman-based machine learning platform reveals unique metabolic differences between IDHmut and IDHwt glioma

Neuro-Oncology, Journal Year: 2024, Volume and Issue: unknown

Published: June 3, 2024

Abstract BACKGROUND Formalin-fixed, paraffin-embedded (FFPE) tissue slides are routinely used in cancer diagnosis, clinical decision-making, and stored biobanks, but their utilization Raman spectroscopy-based studies has been limited due to the background coming from embedding media. METHODS Spontaneous spectroscopy was for molecular fingerprinting of FFPE 46 patient samples with known methylation subtypes. Spectra were construct tumor/non-tumor, IDH1WT/IDH1mut, methylation-subtype classifiers. Support vector machine random forest identify most discriminatory frequencies. Stimulated validate frequencies identified. Mass spectrometry glioma cell lines TCGA biological findings. RESULTS Here we develop APOLLO (rAman-based PathOLogy maLignant glioma) – a computational workflow that predicts different subtypes spontaneous spectra slides. Our novel platform distinguishes tumors nontumor identifies peaks corresponding DNA proteins more intense tumor. differentiates isocitrate dehydrogenase 1 mutant (IDH1mut) wildtype (IDH1WT) cholesterol ester levels be highly abundant IDHmut glioma. Moreover, achieves high discriminative power between finer, clinically relevant subtypes, distinguishing CpG island hypermethylated phenotype (G-CIMP)-high G-CIMP-low phenotypes within IDH1mut types. CONCLUSIONS results demonstrate potential label-free classify extract meaningful information thus opening door future applications on these archived tissues other cancers.

Language: Английский

---

Leveraging the microbiome to combat antibiotic resistant gynecological infections

npj Antimicrobials and Resistance, Journal Year: 2025, Volume and Issue: 3(1)

Published: April 23, 2025

Abstract The vaginal resistome can be considered a collection of the resistant determinants in microbiome. Here we review including microbes and genes harbored common gynecological infections, that participate horizontal gene transfer, host factors contribute to resistome, therapies. Finally, provide perspective on technologies leveraged study remaining challenges.

Language: Английский

---

Monitoring of microbial proteome dynamics using Raman stable isotope probing

Journal of Biophotonics, Journal Year: 2022, Volume and Issue: 16(4)

Published: Dec. 17, 2022

Abstract Abnormal protein kinetics could be a cause of several diseases associated with essential life processes. An accurate understanding dynamics and turnover is for developing diagnostic or therapeutic tools to monitor these changes. Raman spectroscopy in combination stable isotope probes (SIP) such as carbon‐13, deuterium has been breakthrough the qualitative quantitative study various metabolites. In this work, we are reporting utility Raman‐SIP monitoring dynamic changes proteome at community level. We have used 13 C‐labeled glucose only carbon source medium verified its incorporation microbial biomass time‐dependent manner. A visible redshift spectral vibrations major biomolecules nucleic acids, phenylalanine, tyrosine, amide I, III were observed. Temporal intensity bands demonstrating feasibility also verified. Kanamycin, synthesis inhibitor was assess identifying effects on cells. Successful application work can provide an alternate/adjunct tool proteome‐level objective nondestructive

Language: Английский

---

Application of Microfluidics for Bacterial Identification

Pharmaceuticals, Journal Year: 2022, Volume and Issue: 15(12), P. 1531 - 1531

Published: Dec. 9, 2022

Bacterial infections continue to pose serious public health challenges. Though anti-bacterial therapeutics are effective remedies for treating these infections, the emergence of antibiotic resistance has imposed new challenges treatment. Often, there is a delay in prescribing antibiotics at initial symptom presentation as it can be challenging clinically differentiate bacterial from other organisms (e.g., viruses) causing infection. Moreover, arise food, water, or sources. These have demonstrated need rapid identification bacteria liquids, clinical spaces, and environments. Conventional methods rely on culture-based approaches which require long processing times higher pathogen concentration thresholds. In past few years, microfluidic devices paired with various garnered attention addressing limitations conventional demonstrating feasibility lower biomass However, such culture-free often integration multiple steps sample preparation measurement. Research interest using growing; therefore, this review article summary current advancements field focus comparing efficacy polymerase chain reaction (PCR), loop-mediated isothermal amplification (LAMP), emerging spectroscopic methods.

Language: Английский

---

Raman-Based Antimicrobial Susceptibility Testing on Antibiotics of Last Resort

Infection and Drug Resistance, Journal Year: 2023, Volume and Issue: Volume 16, P. 5485 - 5500

Published: Aug. 1, 2023

Background: Antibiotic resistance represents a serious global health challenge, particularly with the emergence of strains resistant to last-resort antibiotics such as tigecycline, polymyxin B, and vancomycin. Urgent measures are required alleviate this situation. To facilitate judicious use antibiotics, rapid precise antimicrobial susceptibility testing (AST) is essential. Heavy water (deuterium oxide, D 2 O)-labeled Raman spectroscopy has emerged promising time-saving tool for microbiological testing. Methods: Deuterium incorporation experimental conditions were examined develop apply Raman-based AST method evaluate efficacy including vancomycin, against Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa , Enterococcus faecium . Essential agreement categorical used assess metabolism inactivation concentration based on (R-MIC)–a new metric developed in study—and minimum inhibitory (MIC) determined via traditional microdilution broth method. Spearman's rank correlation coefficient was employed measure association between R-MIC MIC values. Results: The achieved 100% (92/92) within five hours, while approximately 24 h. values shared 68.5% (63/92) consistency In addition, highly correlated (Spearman's r=0.96), resulting an essential 100%. Conclusion: Our optimized indicate that holds great promise solution overcome time-consuming challenges methods. Keywords: spectroscopy, tigecycline resistant, B vancomycin

Language: Английский

---

Simultaneous Imaging and Characterization of Polyunsaturated Fatty Acids, Carotenoids, and Microcrystalline Guanine in Single Aurantiochytrium limacinum Cells with Linear and Nonlinear Raman Microspectroscopy

The Journal of Physical Chemistry B, Journal Year: 2023, Volume and Issue: 127(12), P. 2708 - 2718

Published: March 15, 2023

Thraustochytrids are heterotrophic marine protists known for their high production capacity of various compounds with health benefits, such as polyunsaturated fatty acids and carotenoids. Although much effort has been focused on developing optimal cultivation methods efficient microbial production, these high-value interrelationships not well understood at the single-cell level. Here we used spontaneous (linear) Raman multiplex coherent anti-Stokes scattering (CARS) microspectroscopy to visualize characterize lipids (e.g., docosahexaenoic acid) carotenoids astaxanthin) accumulated in single living Aurantiochytrium limacinum cells. Spontaneous imaging help multivariate curve resolution–alternating least-squares enabled us make unambiguous assignments molecular components detected derive intracellular distributions separately. Near-IR excited CARS yielded images least an order magnitude faster than imaging, suppressed contributions As culture time increased from 2 5 days, lipid amount by a factor ∼7, whereas carotenoid did change significantly. Furthermore, observed highly localized component A. This was found be mixed crystals guanine other purine derivatives. The present study demonstrates potential linear–nonlinear hybrid approach that allows accurate identification fast label-free manner link information derived cells strategies mass useful thraustochytrids.

Language: Английский

---

Comparative study of machine-and deep-learning based classification algorithms for biomedical Raman spectroscopy (RS): case study of RS based pathogenic microbe identification

Analytical Sciences, Journal Year: 2024, Volume and Issue: 40(12), P. 2101 - 2109

Published: Aug. 29, 2024

One key aspect pushing the frontiers of biomedical RS is dedicated machine- or deep- learning (ML DL) algorithms. Yet, systematic comparative study between ML and DL algorithms has not been conducted for RS, largely due to limited availability open-source large Raman spectra dataset. Therefore we compared typical partial least square-discriminant analysis (PLS-DA) one dimensional convolution neural network (1D-CNN) based pathogenic microbe identification on 12,000 from six species (i.e., K. aerogenes (Klebsiella aerogenes), C. albicans (Candida albicans), glabrata glabrata), Group A Strep. (Group Streptococcus), E. coli1 (Escherichia coli1), coli2 coli2)) when 100%, 75%, 50% 25% were retained. The total dataset was analyzed with 80% split training 20% testing. 100% retained testing accuracy, area under curve (AUC) receiver operating characteristic (ROC) 95.25% 0.997 1D-CNN, which are higher than those (89.42% 0.979) PLS-DA. PLS-DA outperforms 1D-CNN resultant accuracies AUCs demonstrated performance reliance number. Besides, both loadings latent variables saliency maps captured peaks arising DNA proteins comparable interpretability. results current work indicated that should be explored application-wise select whichever AUCs.

Language: Английский

---

Understanding Raman Spectral Based Classifications with Convolutional Neural Networks Using Practical Examples of Fungal Spores and Carotenoid-Pigmented Microorganisms

AI, Journal Year: 2023, Volume and Issue: 4(1), P. 114 - 127

Published: Jan. 18, 2023

Numerous publications showing that robust prediction models for microorganisms based on Raman micro-spectroscopy in combination with chemometric methods are feasible, often very precise predictions. Advances machine learning and easier accessibility to software make it increasingly easy users generate predictive from complex data. However, the question regarding why those predictions so accurate receives much less attention. In our work, we use spectroscopic data of fungal spores carotenoid-containing show is not position peaks or subtle differences band ratios spectra, due small chemical composition organisms, allow classification. Rather, can be characteristic effects baselines spectra biochemically similar enhanced by certain pretreatment even neutral-looking spectral regions great importance a convolutional neural network. Using method called Gradient-weighted Class Activation Mapping, attempt peer into black box networks microbiological applications which responsible

Language: Английский

---