Biology subjects

Blazquez-Sanchez, M.

Publications and source records attributed to Blazquez-Sanchez, M..

4 recordsLinked to original sources

bioRxiv2025

MARISMa: a routine MALDI-TOF MS database from 2018 to 2024

Clinical microbiology laboratories play a crucial role in identifying pathogens, guiding antibiotic treatment, and managing antimicrobial resistance (AMR). Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) has become essential for rapid, accurate, and cost-effective microbial identification. Recent advances in integrating MALDI-TOF MS with Artificial Intelligence (AI) show promise in improving microbial detection and prediction of AMR. However, progress is limited by the lack of comprehensive and openly accessible datasets that restrict the validation, reproducibility, and applicability of the model. To address this gap, we introduce a publicly available MALDI-TOF MS dataset comprising 202,700 unique spectra from isolates collected between 2018 and 2024 at the Hospital General Universitario Gregorio Maranon, Spain. This dataset includes 186,213 bacteria, 16,163 fungal, and 371 mycobacterial samples, of which 29,679 contain AMR annotations. This resource is openly and freely shared, rigorously curated, and designed to support a wide range of machine learning. By ensuring unrestricted access to high-quality, standardized data, this dataset aims to promote transparency, reproducibility, comparative benchmarking, and collaborative progress in AI-driven clinical microbiology.

microbiology↗

bioRxiv2024

Overcoming Challenges of Reproducibility and Variability for the Clostridioides difficile typification

Machine learning (ML) approaches applied to Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS) spectra have shown promise for the typing of Clostridioides difficile, yet their deployment in routine clinical settings remains challenging due to strong sensitivity to acquisition variability. Differences in culture media, incubation time, protein extraction protocols, and instrumentation across hospitals often lead to substantial performance degradation when models are evaluated under heterogeneous or previously unseen conditions. In this work, we systematically analyze the impact of methodological and technical variability on ML-based C. difficile typing and investigate whether data augmentation (DA) strategies can mitigate these effects. Using a dedicated dataset of 60 isolates acquired under diverse conditions, we show that DA substantially improves robustness to variability when training on spectra from selective C. difficile agar media. Importantly, models trained with DA achieve performance levels approaching those obtained using enriched Schaedler agar media, while relying exclusively on standard 24-hour incubation. Evaluation on an independent cohort of 28 newly acquired isolates confirms that DA significantly reduces performance degradation under real-world domain shift. To facilitate adoption and reproducibility, we release MAL-DIDA, an open-source Python library for DA of MALDI-TOF MS spectra.

microbiology↗

bioRxiv2024

Characterization of a nosocomial outbreak caused by VIM-1 Klebsiella michiganensis using Fourier-Transform Infrared (FT-IR) Spectroscopy

Healthcare-associated infections (HAIs) are a significant concern worldwide due to their impact on patient safety and healthcare costs. Klebsiella spp., particularly Klebsiella pneumoniae and Klebsiella oxytoca, are frequently implicated in HAIs and often exhibit multidrug resistance mechanisms, posing challenges for infection control. In this study, we evaluated Fourier-transform Infrared (FT-IR) spectroscopy as a rapid method for characterizing a nosocomial outbreak caused by VIM-1-producing K. oxytoca. A total of 47 isolates, including outbreak strains and controls, were collected from Hospital Universitario Gregorio Maranon, Spain and the University Hospital Basel, Switzerland. FT-IR spectroscopy was employed for bacterial typing, offering rapid and accurate results compared to conventional methods like pulsed-field gel electrophoresis (PFGE) and correlating with whole-genome sequencing (WGS) results. The FT-IR spectra analysis revealed distinct clusters corresponding to outbreak strains, suggesting a common origin. Subsequent WGS analysis identified Klebsiella michiganensis as the causative agent of the outbreak, challenging the initial assumption based on FT-IR results. However, both FT-IR and WGS methods showed high concordance, with an Adjusted Rand index (AR) of 0.882 and an Adjusted Wallace coefficient (AW) of 0.937, indicating the reliability of FT-IR in outbreak characterization. Furthermore, FT-IR spectra visualization highlighted discriminatory features between outbreak and non-outbreak isolates, facilitating rapid screening in case and outbreak is suspected. In conclusion, FT-IR spectroscopy offers a rapid and cost-effective alternative to traditional typing methods, enabling timely intervention and effective management of nosocomial outbreaks. Its integration with WGS enhances the accuracy of outbreak investigations, demonstrating its utility in clinical microbiology and infection control practices.

microbiology↗

bioRxiv2024

Analysis of high-molecular-weight proteins using MALDI-TOF MS and Machine Learning for the differentiation of clinically relevant Clostridioides difficile ribotypes

Clostridioides difficile is the main cause of antibiotic related diarrhea and some ribotypes (RT), such as RT027, RT181 or RT078, are considered high risk clones. A fast and reliable approach for C. difficile ribotyping is needed for a correct clinical approach. This study analyses high-molecular-weight proteins for C. difficile ribotyping with MALDI-TOF MS. Sixty-nine isolates representative of the most common ribotypes in Europe were analyzed in the 17,000-65,000 m/z region and classified into 4 categories (RT027, RT181, RT078 and Other RTs). Five supervised Machine Learning algorithms were tested for this purpose: K-Nearest Neighbors, Support Vector Machine, Partial Least Squares-Discriminant Analysis, Random Forest and Light-Gradient Boosting Machine. All algorithms yielded cross-validation results >70%, being RF and Light-GBM the best performing, with 88% of agreement. Area under the ROC curve of these two algorithms was >0.9. RT078 was correctly classified with 100% accuracy and isolates from the RT181 category could not be differentiated from RT027.

microbiology↗