ISREA: An Efficient Peak-Preserving Baseline Correction Algorithm for Raman Spectra

A critical step in Raman spectroscopy is baseline correction. This procedure eliminates the background signals generated by residual Rayleigh scattering or fluorescence. Baseline correction procedures relying on asymmetric loss functions have been employed recently. They operate with a reduced penalty on positive spectral deviations that essentially push down the baseline estimates from invading Raman peak areas. However, their coupling with polynomial fitting may not be suitable over the whole spectral domain and can yield inconsistent baselines. Their requirement of the specification of a threshold and the non-convexity of the corresponding objective function further complicates the computation. Learning from their pros and cons, we have developed a novel baseline correction procedure called the iterative smoothing-splines with root error adjustment (ISREA) that has three distinct advantages. First, ISREA uses smoothing splines to estimate the baseline that are more flexible than polynomials and capable of capturing complicated trends over the whole spectral domain. Second, ISREA mimics the asymmetric square root loss and removes the need of a threshold. Finally, ISREA avoids the direct optimization of a non-convex loss function by iteratively updating prediction errors and refitting baselines. Through our extensive numerical experiments on a wide variety of spectra including simulated spectra, mineral spectra, and dialysate spectra, we show that ISREA is simple, fast, and can yield consistent and accurate baselines that preserve all the meaningful Raman peaks.

Raman Spectroscopy detects changes in Bone Mineral Quality and Collagen Cross-linkage in Staphylococcus Infected Human Bone

Diagnosis of osteomyelitis presents a formidable challenge. Lack of pathognomonic clinical sign(s) and diagnostic tests that can diagnose osteomyelitis at an early stage contribute to this difficulty. If the diagnosis is not made early, the disease becomes very difficult to eradicate and can lead to limb threatening and potentially life-threatening complications. Staphylococcus aureus is the most common organism causing osteomyelitis. Raman Spectroscopy provides information about molecular vibration that could potentially be harnessed as a spectral signature for cellular changes in specific pathologic conditions. In this study we describe a technique using Raman spectroscopy that could potentially be used to diagnose staphylococcal osteomyelitis. Human bone samples were co-cultured with Staphylococcus aureus (S. aureus) and the effects of bacterial growth on bone quality were then monitored using Raman spectroscopy. A major drop in the bone mineral quality and crystallinity was observed in the infected bones compared to the controls. S. aureus infection was also found to alter the collagen cross-linking. Our study shows that specific spectral signatures are present for the cause as well as the effect of staphylococcal osteomyelitis, opening the possibility of developing a useful diagnostic modality for early and rapid diagnosis of this condition.

Preliminary identification of unicellular algal genus by using combined confocal resonance Raman spectroscopy with PCA and DPLS analysis

The analysis of algae and dominant alga plays important roles in ecological and environmental fields since it can be used to forecast water bloom and control its potential deleterious effects. Herein, we combine in vivo confocal resonance Raman spectroscopy with multivariate analysis methods to preliminary identify the three algal genera in water blooms at unicellular scale. Statistical analysis of characteristic Raman peaks demonstrates that certain shifts and different normalized intensities, resulting from composition of different carotenoids, exist in Raman spectra of three algal cells. Principal component analysis (PCA) scores and corresponding loading weights show some differences from Raman spectral characteristics which are caused by vibrations of carotenoids in unicellular algae. Then, discriminant partial least squares (DPLS) classification method is used to verify the effectiveness of algal identification with confocal resonance Raman spectroscopy. Our results show that confocal resonance Raman spectroscopy combined with PCA and DPLS could handle the preliminary identification of dominant alga for forecasting and controlling of water blooms.

Raman Imaging of Plant Cell Walls in Sections of Cucumis sativus

Raman microspectra combine information on chemical composition of plant tissues with spatial information. The contributions from the building blocks of the cell walls in the Raman spectra of plant tissues can vary in the microscopic sub-structures of the tissue. Here, we discuss the analysis of 55 Raman maps of root, stem, and leaf tissues of Cucumis sativus, using different spectral contributions from cellulose and lignin in both univariate and multivariate imaging methods. Imaging based on hierarchical cluster analysis (HCA) and principal component analysis (PCA) indicates different substructures in the xylem cell walls of the different tissues. Using specific signals from the cell wall spectra, analysis of the whole set of different tissue sections based on the Raman images reveals differences in xylem tissue morphology. Due to the specifics of excitation of the Raman spectra in the visible wavelength range (532 nm), which is, e.g., in resonance with carotenoid species, effects of photobleaching and the possibility of exploiting depletion difference spectra for molecular characterization in Raman imaging of plants are discussed. The reported results provide both, specific information on the molecular composition of cucumber tissue Raman spectra, and general directions for future imaging studies in plant tissues.

Classification and identification of pigmented cocci bacteria relevant to the soil environment via Raman spectroscopy

A soil habitat consists of a significant number of bacteria that cannot be cultivated by conventional means, thereby posing obvious difficulties in their classification and identification. This difficulty necessitates the need for advanced techniques wherein a well-compiled biomolecular database consisting of the already cultivable bacteria can be used as a reference in an attempt to link the noncultivable bacteria to their closest phylogenetic groups. Raman spectroscopy has been successfully applied to taxonomic studies of many systems like bacteria, fungi, and plants relying on spectral differences contributed by the variation in their overall biomolecular composition. However, these spectral differences can be obscured due to Raman signatures from photosensitive microbial pigments like carotenoids that show enormous variation in signal intensity hindering taxonomic investigations. In this study, we have applied laser-induced photobleaching to expel the carotenoid signatures from pigmented cocci bacteria. Using this method, we have investigated 12 species of pigmented bacteria abundant in soil habitats belonging to three genera mainly Micrococcus, Deinococcus, and Kocuria based on their Raman spectra with the assistance of a chemometric tool known as the radial kernel support vector machine (SVM). Our results demonstrate the potential of Raman spectroscopy as a minimally invasive taxonomic tool to identify pigmented cocci soil bacteria at a single-cell level.

Chemical fixation methods for Raman spectroscopy-based analysis of bacteria

Preservation of biological samples for downstream analysis is important for analytical methods that measure the biochemical composition of a sample. One such method, Raman microspectroscopy, is commonly used as a rapid phenotypic technique to measure biomolecular composition for the purposes of identification and discrimination of species and strains of bacteria, as well as investigating physiological responses to external stressors and the uptake of stable isotope-labelled substrates in single cells. This study examines the influence of a number of common chemical fixation and inactivation methods on the Raman spectrum of six species of bacteria. Modifications to the Raman-phenotype caused by fixation were compared to unfixed control samples using difference spectra and Principal Components Analysis (PCA). Additionally, the effect of fixation on the ability to accurately classify bacterial species using their Raman phenotype was determined. The results showed that common fixatives such as glutaraldehyde and ethanol cause significant changes to the Raman spectra of bacteria, whereas formaldehyde and sodium azide were better at preserving spectral features.

Laser raman spectroscopy as a potential chair-side microbiological diagnostic device

INTRODUCTION

Culture-dependent and -independent techniques are time-consuming processes requiring highly trained personnel to identify microorganisms contained within a sample. Rapid chair-side identification of microorganisms could reduce the lag time between patient presentation and ideal treatment. As a first step toward this goal, this study aims to determine if laser Raman spectroscopy (LRS) can discern uniqueness among 10 different species of bacteria contained within a medium in unprocessed and processed samples.

METHODS

Ten bacterial species were individually grown on blood agar plates for 3 days. Checkerboard DNA-DNA hybridization was used for species verification. For the unprocessed samples, a 1.0-cm diameter agar sample, with undisturbed bacterial growth, was transferred for each species to a barium fluoride crystal (BaF(2)) slide and laser scanned for a total of 15 seconds per sample. For the processed samples, bacterial cells were harvested, washed, and resuspended in phosphate-buffered saline buffer at 10(9) cells/mL concentration. Each suspension was laser scanned for 15 seconds on a BaF(2) slide. Select regions of Raman spectra for each species/agar and species/suspension combination were processed using a two-sided t test.

RESULTS

For the 10 bacterial species, 45 bacteria pair combinations were tested for each group. In both groups, LRS was capable of statistically distinguishing among a majority of bacterial pairings based on RS signature differences of means.

CONCLUSIONS

Results show each bacterial species generated restricted ranges of unique spectral signatures that were not masked by their containing medium. Chair-side LRS is a promising technique that differentiates among oral bacterial species with a high degree of specificity.

Towards Raman-based epidemiological typing of Pseudomonas aeruginosa

Raman spectra of bacteria can be used as highly specific fingerprints, enabling discrimination at strain level. Pseudomonas aeruginosa strains can be strongly pigmented, making it difficult to obtain high quality spectra of such isolates due to high fluorescent spectral backgrounds. Furthermore, the spectra that could be measured with acceptable quality often showed large spectral variations limiting the reproducibility required for strain level discrimination. P. aeruginosa produces a characteristic yellowish green fluorescent pigment, called pyoverdin. Applying a washing procedure to reduce the amount of fluorescent pigment, enabled the highly pigmented isolates to be measured with sufficient spectral quality. Isolation of the pigment/pyoverdin spectral features, together with spectral scaling methods improved reproducibility. It will be important to analyze the range of the spectral variations that can occur and ensure the correction of all of these factors to obtain the highest reproducibility required for strain level typing.

Discriminating isogenic cancer cells and identifying altered unsaturated fatty acid content as associated with metastasis status, using k-means clustering and partial least squares-discriminant analysis of Raman maps

Raman spectroscopy is a label-free, real-time diagnostic tool that shows great promise in identifying cell differences. We have evaluated the discriminatory power of Raman spectroscopy using a unique model system consisting of two isogenic cancer cell lines derived from the MDA-MB-435 cell line. The two cell lines are equally tumorigenic in mice, but while M-4A4 gives rise to metastasis, NM-2C5 only disseminates single cells that remain dormant in distant organs. Previous comparative proteomic and transcriptomic analyses of the two cell lines have shown that they differ only in the expression level of a few proteins and genes. Raman maps were recorded of single cells after fixation and drying using 785 nm laser excitation. K-means clustering reduced the amount of data from each cell and improved the signal-to-noise ratio of cluster-averaged spectra. Spectra representing the nucleus were discarded as they showed much smaller differences between the two cell lines compared to cytoplasm spectra. Partial least squares-discriminant analysis (PLS-DA) was applied to distinguish the two cell lines. A cross-validated PLS-DA resulted in 92% correctly classified samples. Spectral differences were assigned to a higher unsaturated fatty acid content in the metastatic vs nonmetastatic cell line. Our study demonstrates the unique ability of Raman spectroscopy to distinguish minute differences at the subcellular level and yield new biological information. Our study is the first to demonstrate the association between polyunsaturated fatty acid content and metastatic ability in this unique cell model system and is in agreement with previous studies on this topic.

Characterization of pollen carotenoids with in situ and high-performance thin-layer chromatography supported resonant Raman spectroscopy

Raman signatures of the carotenoid component are studied in individual pollen grains from different species of trees. The information is obtained as differences in the strong pre-resonant Raman spectra measured before and after photodepletion of the carotenoid molecules. The results provide the first in situ evidence of interspecies differences in pollen carotenoid content, structure, and/or assembly between plant species without prior purification. The analysis of carotenoids in situ is confirmed by high-performance thin-layer chromatography (HPTLC)-supported resonance Raman data measured directly on the HPTLC plates after separation of carotenoids in pollen extracts. Utilization of the in situ, extraction-free procedure in carotenoid analysis will improve sensitivity and structural selectivity and provides insight into carotenoid structure and composition in single pollen grains.

Raman spectroscopic typing reveals the presence of carotenoids in Mycoplasma pneumoniae

Raman spectroscopy has previously been demonstrated to be a highly useful methodology for the identification and/or typing of micro-organisms. In this study, we set out to evaluate whether this technology could also be applied as a tool to discriminate between isolates of Mycoplasma pneumoniae, which is generally considered to be a genetically highly uniform species. In this evaluation, a total of 104 strains of M. pneumoniae were analysed, including two reference strains (strains M129 and FH), and 102 clinical isolates, which were isolated between 1973 and 2005 and originated from various countries. By Raman spectral analysis (Raman typing) of this strain collection, we were able to reproducibly distinguish six different clusters of strains. An unequivocal correlation between Raman typing and P1 genotyping, which is based on sequence differences in the P1 (or MPN141) gene of M. pneumoniae, was not observed. In the two major Raman clusters that we identified (clusters 3 and 6, which together harboured 81 % of the strains), the different P1 subtypes were similarly distributed, and ∼76 % isolates were of subtype 1, ∼20 % of subtype 2 and ∼5 % of variant 2a. Nevertheless, a relatively high prevalence of P1 subtype 2 strains was found in clusters 2 and 5 (100 %), as well as in cluster 1 (75 %) and cluster 4 (71 %); these clusters, however, harboured a small number of strains. Only two of the strains (2 %) could not be typed correctly. Interestingly, analysis of the Raman spectra revealed the presence of carotenoids in M. pneumoniae. This finding is in line with the identification of M. pneumoniae genes that have similarity with genes involved in a biochemical pathway leading to carotenoid synthesis, i.e. the 2-C-methyl-d-erythritol 4-phosphate (MEP) pathway. Therefore, we hypothesize that M. pneumoniae hosts an MEP-like pathway for carotenoid synthesis. We conclude that Raman spectroscopy is a convenient tool for discriminating between M. pneumoniae strains, and that it presents a promising supplement to the current methods for typing of this bacterium.

Optical fingerprinting in bacterial epidemiology: Raman spectroscopy as a real-time typing method

Hospital-acquired infections (HAI) increase morbidity and mortality and constitute a high financial burden on health care systems. An effective weapon against HAI is early detection of potential outbreaks and sources of contamination. Such monitoring requires microbial typing with sufficient reproducibility and discriminatory power. Here, a microbial-typing method is presented, based on Raman spectroscopy. This technique provides strain-specific optical fingerprints in a few minutes instead of several hours to days, as is the case with genotyping methods. Although the method is generally applicable, we used 118 Staphylococcus aureus isolates to illustrate that the discriminatory power matches that of established genotyping techniques (numerical index of diversity [D]=0.989) and that concordance with the gold standard (pulsed-field gel electrophoresis) is high (95%). The Raman clustering of isolates was reproducible to the strain level for five independent cultures, despite the various culture times from 18 h to 24 h. Furthermore, this technique was able to classify stored (-80 degrees C) and recent isolates of a methicillin-resistant Staphylococcus aureus-colonized individual during surveillance studies and did so days earlier than established genotyping techniques did. Its high throughput and ease of use make it suitable for use in routine diagnostic laboratory settings. This will set the stage for continuous, automated, real-time epidemiological monitoring of bacterial infections in a hospital, which can then be followed by timely corrective action by infection prevention teams.