High-Speed Diagnosis of Bacterial Pathogens at the Single Cell Level by Raman Microspectroscopy with Machine Learning Filters and Denoising Autoencoders

2022 ◽  
Author(s):  
Jiabao Xu ◽  
Xiaofei Yi ◽  
Guilan Jin ◽  
Di Peng ◽  
Gaoya Fan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Single Cell ◽  
High Speed ◽  
Bacterial Pathogens ◽  
Raman Microspectroscopy ◽  
Single Cell Level ◽  
Cell Level

Identification of water-conditioned Pseudomonas aeruginosa by Raman microspectroscopy on a single cell level

2014 ◽  
Vol 37 (5) ◽  
pp. 360-367 ◽  
Author(s):  
Anja Silge ◽  
Wilm Schumacher ◽  
Petra Rösch ◽  
Paulo A. Da Costa Filho ◽  
Cédric Gérard ◽  
...  
Keyword(s):  
Pseudomonas Aeruginosa ◽  
Single Cell ◽  
Raman Microspectroscopy ◽  
Single Cell Level ◽  
Cell Level

scholarly journals Convolutional Neural Network Can Recognize Drug Resistance of Single Cancer Cells

2020 ◽  
Vol 21 (9) ◽  
pp. 3166
Author(s):  
Kiminori Yanagisawa ◽  
Masayasu Toratani ◽  
Ayumu Asai ◽  
Masamitsu Konno ◽  
Hirohiko Niioka ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Convolutional Neural Network ◽  
Single Cell ◽  
Tumor Cells ◽  
Antitumor Drugs ◽  
Single Cell Level ◽  
Cell Level ◽  
Single Cancer ◽  
Deep Learning Model

It is known that single or isolated tumor cells enter cancer patients’ circulatory systems. These circulating tumor cells (CTCs) are thought to be an effective tool for diagnosing cancer malignancy. However, handling CTC samples and evaluating CTC sequence analysis results are challenging. Recently, the convolutional neural network (CNN) model, a type of deep learning model, has been increasingly adopted for medical image analyses. However, it is controversial whether cell characteristics can be identified at the single-cell level by using machine learning methods. This study intends to verify whether an AI system could classify the sensitivity of anticancer drugs, based on cell morphology during culture. We constructed a CNN based on the VGG16 model that could predict the efficiency of antitumor drugs at the single-cell level. The machine learning revealed that our model could identify the effects of antitumor drugs with ~0.80 accuracies. Our results show that, in the future, realizing precision medicine to identify effective antitumor drugs for individual patients may be possible by extracting CTCs from blood and performing classification by using an AI system.

scholarly journals Stress-Induced Dinoflagellate Bioluminescence at the Single Cell Level

2020 ◽  
Author(s):  
Maziyar Jalaal ◽  
Nico Schramma ◽  
Antoine Dode ◽  
Hélène de Maleprade ◽  
Christophe Raufaste ◽  
...  
Keyword(s):  
Single Cell ◽  
High Speed ◽  
Fluid Flows ◽  
Breaking Waves ◽  
Single Cell Level ◽  
Cell Level ◽  
High Speed Imaging ◽  
Light Production ◽  
Marine Dinoflagellates ◽  
Characteristic Features

One of the characteristic features of many marine dinoflagellates is their bioluminescence, which lights up nighttime breaking waves or seawater sliced by a ship’s prow. While the internal biochemistry of light production by these microorganisms is well established, the manner by which fluid shear or mechanical forces trigger bioluminescence is still poorly understood. We report controlled measurements of the relation between mechanical stress and light production at the single-cell level, using high-speed imaging of micropipette-held cells of the marine dinoflagellate Pyrocystis lunula subjected to localized fluid flows or direct indentation. We find a viscoelastic response in which light intensity depends on both the amplitude and rate of deformation, consistent with the action of stretch-activated ion channels. A phenomenological model captures the experimental observations.

scholarly journals Adaptable microfluidic system for single-cell pathogen classification and antimicrobial susceptibility testing

2019 ◽  
Vol 116 (21) ◽  
pp. 10270-10279 ◽  
Author(s):  
Hui Li ◽  
Peter Torab ◽  
Kathleen E. Mach ◽  
Christine Surrette ◽  
Matthew R. England ◽  
...  
Keyword(s):  
Single Cell ◽  
Antimicrobial Susceptibility ◽  
Bacterial Infections ◽  
Susceptibility Testing ◽  
Bacterial Pathogens ◽  
Microfluidic System ◽  
Antimicrobial Susceptibility Testing ◽  
Microbiological Analysis ◽  
Single Cell Level ◽  
Cell Level

Infectious diseases caused by bacterial pathogens remain one of the most common causes of morbidity and mortality worldwide. Rapid microbiological analysis is required for prompt treatment of bacterial infections and to facilitate antibiotic stewardship. This study reports an adaptable microfluidic system for rapid pathogen classification and antimicrobial susceptibility testing (AST) at the single-cell level. By incorporating tunable microfluidic valves along with real-time optical detection, bacteria can be trapped and classified according to their physical shape and size for pathogen classification. By monitoring their growth in the presence of antibiotics at the single-cell level, antimicrobial susceptibility of the bacteria can be determined in as little as 30 minutes compared with days required for standard procedures. The microfluidic system is able to detect bacterial pathogens in urine, blood cultures, and whole blood and can analyze polymicrobial samples. We pilot a study of 25 clinical urine samples to demonstrate the clinical applicability of the microfluidic system. The platform demonstrated a sensitivity of 100% and specificity of 83.33% for pathogen classification and achieved 100% concordance for AST.

Using Stable Isotope Probing and Raman Microspectroscopy to measure growth rates of heterotrophic bacteria

2021 ◽  
Author(s):  
Felix Weber ◽  
Tatiana Zaliznyak ◽  
Virginia P. Edgcomb ◽  
Gordon T. Taylor
Keyword(s):  
Stable Isotope ◽  
Single Cell ◽  
Bacterial Growth ◽  
Growth Rates ◽  
Heterotrophic Bacteria ◽  
Raman Microspectroscopy ◽  
Stable Isotope Probing ◽  
Single Cell Level ◽  
Cell Level ◽  
E Coli

The suitability of stable isotope probing (SIP) and Raman microspectroscopy to measure growth rates of heterotrophic bacteria at the single-cell level was evaluated. Label assimilation into E. coli biomass during growth on a complex 13 C-labeled carbon source was monitored in time course experiments. 13 C-incorporation into various biomolecules was measured by spectral “red shifts” of Raman-scattered emissions. The 13 C- and 12 C-isotopologues of the amino acid phenylalanine (Phe) proved to be a quantitatively accurate reporter molecules of cellular isotopic fractional abundances ( f cell ). Values of f cell determined by Raman microspectroscopy and independently by isotope-ratio mass spectrometry (IRMS) over a range of isotopic enrichments were statistically indistinguishable. Progressive labeling of Phe in E. coli cells among a range of 13 C/ 12 C organic substrate admixtures occurred predictably through time. Relative isotopologue abundances of Phe determined by Raman spectral analysis enabled accurate calculation of bacterial growth rates as confirmed independently by optical density (OD) measurements. Results demonstrate that combining stable isotope probing (SIP) and Raman microspectroscopy can be a powerful tool for studying bacterial growth at the single-cell level when grown on defined or complex organic 13 C-carbon sources even in mixed microbial assemblages. Importance: Population growth dynamics and individual cell growth rates are the ultimate expressions of a microorganism’s fitness to its environmental conditions, whether natural or engineered. Natural habitats and many industrial settings harbor complex microbial assemblages. Their heterogeneity in growth responses to existing and changing conditions is often difficult to grasp by standard methodologies. In this proof of concept study, we tested whether Raman microspectroscopy can reliably quantify assimilation of isotopically-labeled nutrients into E. coli cells and enable determination of individual growth rates among heterotrophic bacteria. Raman-derived growth rate estimates were statistically indistinguishable from those derived by standard optical density measurements of the same cultures. Raman microspectroscopy also can be combined with methods for phylogenetic identification. We report development of Raman-based techniques that enable researchers to directly link genetic identity to functional traits and rate measurements of single cells within mixed microbial assemblages, currently a major technical challenge in microbiological research.

scholarly journals PIKE-R2P: Protein–protein interaction network-based knowledge embedding with graph neural network for single-cell RNA to protein prediction

2021 ◽  
Vol 22 (S6) ◽  
Author(s):  
Xinnan Dai ◽  
Fan Xu ◽  
Shike Wang ◽  
Piyushkumar A. Mundra ◽  
Jie Zheng
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Single Cell ◽  
Prior Knowledge ◽  
Protein Interactions ◽  
Interaction Network ◽  
Superior Performance ◽  
Single Cell Level ◽  
Cell Level ◽  
Protein Prediction

Abstract Background Recent advances in simultaneous measurement of RNA and protein abundances at single-cell level provide a unique opportunity to predict protein abundance from scRNA-seq data using machine learning models. However, existing machine learning methods have not considered relationship among the proteins sufficiently. Results We formulate this task in a multi-label prediction framework where multiple proteins are linked to each other at the single-cell level. Then, we propose a novel method for single-cell RNA to protein prediction named PIKE-R2P, which incorporates protein–protein interactions (PPI) and prior knowledge embedding into a graph neural network. Compared with existing methods, PIKE-R2P could significantly improve prediction performance in terms of smaller errors and higher correlations with the gold standard measurements. Conclusion The superior performance of PIKE-R2P indicates that adding the prior knowledge of PPI to graph neural networks can be a powerful strategy for cross-modality prediction of protein abundances at the single-cell level.

Sign in / Sign up

Export Citation Format

Share Document