Exploration of Principal Component Analysis: Deriving Principal Component Analysis Visually Using Spectra.

Spectroscopy rapidly captures a large amount of data that is not directly interpretable. Principal component analysis is widely used to simplify complex spectral datasets into comprehensible information by identifying recurring patterns in the data with minimal loss of information. The linear algebra underpinning principal component analysis is not well understood by many applied analytical scientists and spectroscopists who use principal component analysis. The meaning of features identified through principal component analysis is often unclear. This manuscript traces the journey of the spectra themselves through the operations behind principal component analysis, with each step illustrated by simulated spectra. Principal component analysis relies solely on the information within the spectra, consequently the mathematical model is dependent on the nature of the data itself. The direct links between model and spectra allow concrete spectroscopic explanation of principal component analysis , such as the scores representing "concentration" or "weights". The principal components (loadings) are by definition hidden, repeated and uncorrelated spectral shapes that linearly combine to generate the observed spectra. They can be visualized as subtraction spectra between extreme differences within the dataset. Each PC is shown to be a successive refinement of the estimated spectra, improving the fit between PC reconstructed data and the original data. Understanding the data-led development of a principal component analysis model shows how to interpret application specific chemical meaning of the principal component analysis loadings and how to analyze scores. A critical benefit of principal component analysis is its simplicity and the succinctness of its description of a dataset, making it powerful and flexible.

Raman spectroscopy as a predictive tool for monitoring osteoporosis therapy in a rat model of postmenopausal osteoporosis.

Pharmacological therapy of osteoporosis reduces bone loss and risk of fracture in patients. Modulation of bone mineral density cannot explain all effects. Other aspects of bone quality affecting fragility and ways to monitor them need to be better understood. Keratinous tissue acts as surrogate marker for bone protein deterioration caused by oestrogen deficiency in rats. Ovariectomised rats were treated with alendronate (ALN), parathyroid hormone (PTH) or estrogen (E2). MicroCT assessed macro structural changes. Raman spectroscopy assessed biochemical changes. Micro CT confirmed that all treatments prevented ovariectomy-induced macro structural bone loss in rats. PTH induced macro structural changes unrelated to ovariectomy. Raman analysis revealed ALN and PTH partially protect against molecular level changes to bone collagen (80% protection) and mineral (50% protection) phases. E2 failed to prevent biochemical change. The treatments induced alterations unassociated with the ovariectomy; increased beta sheet with E2, globular alpha helices with PTH and fibrous alpha helices with both ALN and PTH. ALN is closest to maintaining physiological status of the animals, while PTH (comparable protective effect) induces side effects. E2 is unable to prevent molecular level changes associated with ovariectomy. Raman spectroscopy can act as predictive tool for monitoring pharmacological therapy of osteoporosis in rodents. Keratinous tissue is a useful surrogate marker for the protein related impact of these therapies.The results demonstrate utility of surrogates where a clear systemic causation connects the surrogate to the target tissue. It demonstrates the need to assess broader biomolecular impact of interventions to examine side effects.

A Preliminary Evaluation of the Ability of Keratotic Tissue to Act as a Prognostic Indicator of Hip Fracture Risk.

Studies have shown that Raman spectroscopic analysis of fingernail clippings can help differentiate between post-menopausal women who have and who have not suffered a fracture. However, all studies to date have been retrospective in nature, comparing the proteins in nails sourced from women, post-fracture. The objective of this study was to investigate the potential of a prospective test for hip fracture based on spectroscopic analysis of nail tissue. Archived toenail samples from post-menopausal women aged 50 to 63 years in the Nurses' Health Study were obtained and analysed by Raman spectroscopy. Nails were matched case-controls sourced from 161 women; 82 who underwent a hip fracture up to 20 years after nail collection and 81 age-matched controls. A number of clinical risk factors (CRFs) from the Fracture Risk Assessment (FRAX) tool had been assessed at toenail collection. Using 80% of the spectra, models were developed for increasing time periods between nail collection and fracture. Scores were calculated from these models for the other 20% of the sample and the ability of the score to predict hip fracture was tested in model with and without the CRFs by comparing the odds ratios (ORs) per 1 SD increase in standardised predictive values. The Raman score successfully distinguished between hip fracture cases and controls. With only the score as a predictor, a statistically significant OR of 2.2 (95% confidence interval [CI]: 1.5-3.1) was found for hip fracture for up to 20 years after collection. The OR increased to 3.8 (2.6-5.4) when the CRFs were added to the model. For fractures limited to 13 years after collection, the OR was 6.3 (3.0-13.1) for the score alone. The test based on Raman spectroscopy has potential for identifying individuals who may suffer hip fractures several years in advance. Higher powered studies are required to evaluate the predictive capability of this test.

Raman spectroscopy predicts the link between claw keratin and bone collagen structure in a rodent model of oestrogen deficiency.

Osteoporosis is a common disease characterised by reduced bone mass and an increased risk of fragility fractures. Low bone mineral density is known to significantly increase the risk of osteoporotic fractures, however, the majority of non-traumatic fractures occur in individuals with a bone mineral density too high to be classified as osteoporotic. Therefore, there is an urgent need to investigate aspects of bone health, other than bone mass, that can predict the risk of fracture. Here, we successfully predicted association between bone collagen and nail keratin in relation to bone loss due to oestrogen deficiency using Raman spectroscopy. Raman signal signature successfully discriminated between ovariectomised rats and their sham controls with a high degree of accuracy for the bone (sensitivity 89%, specificity 91%) and claw tissue (sensitivity 89%, specificity 82%). When tested in an independent set of claw samples the classifier gave 92% sensitivity and 85% specificity. Comparison of the spectral changes occurring in the bone tissue with the changes occurring in the keratin showed a number of common features that could be attributed to common changes in the structure of bone collagen and claw keratin. This study established that systemic oestrogen deficiency mediates parallel structural changes in both the claw (primarily keratin) and bone proteins (primarily collagen). This strengthens the hypothesis that nail keratin can act as a surrogate marker of bone protein status where systemic processes induce changes.

Raman spectroscopy for the detection of AGEs/ALEs.

Raman spectroscopy is a noninvasive, nondestructive tool for capturing multiplexed biochemical information across diverse molecular species including proteins, lipids, DNA, and mineralizations. Based on light scattering from molecules, cells, and tissues, it is possible to detect molecular fingerprints and discriminate between subtly different members of each biochemical class. Raman spectroscopy is ideal for detecting perturbations from the expected molecular structure such as those occurring during senescence and the modification of long-lived proteins by metabolic intermediates as we age. Here, we describe the sample preparation, data acquisition, signal processing, data analysis and interpretation involved in using Raman spectroscopy for detecting age-related protein modifications in complex biological tissues.

Sclera as a surrogate marker for determining AGE-modifications in Bruch's membrane using a Raman spectroscopy-based index of aging.

PURPOSE: Raman spectroscopy is an effective probe of advanced glycation end products (AGEs) in Bruch's membrane. However, because it is the outermost layer of the retina, this extracellular matrix is difficult to analyze in vivo with current technology. The sclera shares many compositional characteristics with Bruch's membrane, but it is much easier to access for in vivo Raman analysis. This study investigated whether sclera could act as a surrogate tissue for Raman-based investigation of pathogenic AGEs in Bruch's membrane. METHODS: Human sclera and Bruch's membrane were dissected from postmortem eyes (n = 67) across a wide age range (33-92 years) and were probed by Raman spectroscopy. The biochemical composition, AGEs, and their age-related trends were determined from data reduction of the Raman spectra and compared for the two tissues. RESULTS: Raman microscopy demonstrated that Bruch's membrane and sclera are composed of a similar range of biomolecules but with distinct relative quantities, such as in the heme/collagen and the elastin/collagen ratios. Both tissues accumulated AGEs, and these correlated with chronological age (R(2) = 0.824 and R(2) = 0.717 for sclera and Bruch's membrane, respectively). The sclera accumulated AGE adducts at a lower rate than Bruch's membrane, and the models of overall age-related changes exhibited a lower rate (one-fourth that of Bruch's membrane) but a significant increase with age (P < 0.05). CONCLUSIONS: The results suggest that the sclera is a viable surrogate marker for estimating AGE accumulation in Bruch's membrane and for reliably predicting chronological age. These findings also suggest that sclera could be a useful target tissue for future patient-based, Raman spectroscopy studies.

Multiplex analysis of age-related protein and lipid modifications in human Bruch's membrane.

Aging of the human retina is characterized by progressive pathology, which can lead to vision loss. This progression is believed to involve reactive metabolic intermediates reacting with constituents of Bruch's membrane, significantly altering its physiochemical nature and function. We aimed to replace a myriad of techniques following these changes with one, Raman spectroscopy. We used multiplexed Raman spectroscopy to analyze the age-related changes in 7 proteins, 3 lipids, and 8 advanced glycation/lipoxidation endproducts (AGEs/ALEs) in 63 postmortem human donors. We provided an important database for Raman spectra from a broad range of AGEs and ALEs, each with a characteristic fingerprint. Many of these adducts were shown for the first time in human Bruch's membrane and are significantly associated with aging. The study also introduced the previously unreported up-regulation of heme during aging of Bruch's membrane, which is associated with AGE/ALE formation. Selection of donors ranged from ages 32 to 92 yr. We demonstrated that Raman spectroscopy can identify and quantify age-related changes in a single nondestructive measurement, with potential to measure age-related changes in vivo. We present the first directly recorded evidence of the key role of heme in AGE/ALE formation.

Identifying the spatial distribution of vitamin E, pulmonary surfactant and membrane lipids in cells and tissue by confocal Raman microscopy.

Every organ compromises of several different cell types. When studying the effects of a chosen compound within this organ or tissue uptake, localisation, metabolism, and the effect itself can be expected to differ between cells. Using the example of Vitamin E in pulmonary tissue we introduce confocal Raman Microscopy as a superior method to localise lipid-soluble compounds within tissues and cells. We describe the analyses of vitamin E, its oxidation products, and metabolites as well as pulmonary surfactant phospholipids in fixed lung tissue sections. Examples of main structural membrane lipids (PC, cholesterol) and an example of a lipid-signalling molecule (ceramide) are also included. Confocal Raman microscopy is a non-destructive optical method of analysing chemical and physical composition of solids, liquids, gases, gels, and solutions. The method is rich in information allowing discrimination of chemically similar molecules (including geometric isomers) and sensitive monitoring of subtle physical interactions. Additionally, Raman spectroscopy is relatively insensitive to water allowing the analysis of aqueous solutions and suspensions typical in biochemistry. In contrast, Raman spectroscopy is sensitive to non-polar molecules making it ideal for lipidomics research.

Preliminary investigations on the effects of ageing and cooking on the Raman spectra of porcine longissimus dorsi.

The influence of ageing and cooking on the Raman spectrum of porcine longissimus dorsi was investigated. The rich information contained in the Raman spectrum was highlighted, with numerous changes attributed to changes in the environment and conformations of the myofibrillar proteins. Predictions equations for shear force and cooking loss were developed from the Raman spectra of both raw and cooked pork. Good correlations and standard errors of prediction were obtained for both WB shear force and cooking loss, with the raw and the cooked samples showing almost identical results R(2)=0.77, root mean standard error of prediction (RMSEP)% of mean=12% for shear force; R(2)=0.71, RMSEP% of mean=10% for cooking loss. The Raman spectra were also able to predict the extent of cooking that occurred within the pork (R(2)(val)=0.94, RMSEP% of range=5.5%). Raman spectroscopy has considerable potential as a method for non-destructive and rapid determination of pork quality parameters such as tenderness. Raman spectroscopy may provide a means of determining changes during cooking and the extent to which foods have been cooked.

Raman microscopy of porcine inner retinal layers from the area centralis.

PURPOSE: To characterize the Raman spectra of porcine inner retinal layers, specifically, the inner nuclear, inner plexiform, ganglion cell, and nerve fiber layers. METHODS: Raman microscopy was employed at three excitation wavelengths, 785, 633, and 514 nm to measure Raman spectra in a high resolution grid across the inner layers of 4% paraformaldehyde cryoprotected porcine retina. Multivariate statistics were used to summarize the principal spectral signals within those layers and to map the distribution of each of those signals. RESULTS: The detected Raman scattering was dominated by a signal characteristic of the protein population present in each layer. As expected, a significant nucleotide contribution was observed in the inner nuclear layer, while the inner plexiform layer displayed a minor contribution from fatty acid based lipid, which would be characteristic of the axonal and synaptic connection resident in this layer. Blood vessels were readily characterized by their distinct heme-derived spectral signature, which increased at 633 and 514 nm excitation compared to 785 nm. Discrete isolated nucleotide signals were identified in the ganglion cell layer, while the nerve fiber layer exhibited a homogenous profile, which is indicative of its broadly uniform axonal and cytoplasmic Muller cell components. CONCLUSIONS: The present study demonstrated the potential of Raman microscopy as a tool to study the biochemical composition of pathologically normal retina. Specifically, the method allowed a unique method of analyzing the network of neurons involved in relaying information from the photoreceptor population to the ganglion cell derived nerve fiber layer. The study has demonstrated the ability of Raman microscopy to generate simultaneously information on a range of specific biochemical entities within the stratified normal retina.

Confocal Raman microscopy can quantify advanced glycation end product (AGE) modifications in Bruch's membrane leading to accurate, nondestructive prediction of ocular aging.

The modification of proteins by nonenzymatic glycation leading to accumulation of advanced glycation end products (AGEs) is a well-established phenomenon of aging. In the eyes of elderly patients, these adducts have been observed in retinal pigment epithelium (RPE), particularly within the underlying pentalaminar substrate known as Bruch's membrane. AGEs have also been localized to age-related subcellular deposits (drusen and basal laminar deposits) and are thought to play a pathogenic role in progression of the major sight-threatening condition known as age-related macular degeneration (AMD). The current study has quantified AGEs in Bruch's membrane from postmortem eyes and established age-related correlations. In particular, we investigated the potential of confocal Raman microscopy to identify and quantify AGEs in Bruch's membrane in a nondestructive, analytical fashion. Bruch's membrane and the innermost layers of the underlying choroid (BM-Ch) were dissected from fresh postmortem eye-cups (n=56). AGE adducts were quantified from homogenized tissue using reverse-phase HPLC and GC/MS in combination with immunohistochemistry. For parallel Raman analysis, BM-Ch was flat-mounted on slides and evaluated using a Raman confocal microscope and spectra analyzed by a range of statistical approaches. Quantitative analysis showed that the AGEs pentosidine, carboxymethyllysine (CML), and carboxyethyllysine (CEL) occurred at significantly higher levels in BM-Ch with age (P<0.05-0.01). Defined Raman spectral "fingerprints" were identified for various AGEs and these were observed in the clinical samples using confocal Raman microscopy. The Raman data set successfully modeled AGEs and not only provided quantitative data that compared with conventional analytical approaches, but also provided new and complementary information via a nondestructive approach with high spatial resolution. It was shown that the Raman approach could be used to predict chronological age of the clinical samples (P<0.001) and a difference in the Raman spectra between genders was highly significant (P<0.000001). With further development, this Raman-based approach has the potential for noninvasive examination of AGE adducts in living eyes and ultimately to assess their precise pathogenic role in age-related diseases.

Classification of adipose tissue species using Raman spectroscopy.

In this study multivariate analysis of Raman spectra has been used to classify adipose tissue from four different species (chicken, beef, lamb and pork). The adipose samples were dissected from the carcass and their spectra recorded without further preparation. 102 samples were used to create and compare a range of statistical models, which were then tested on 153 independent samples. Of the classical multivariate methods employed, Partial Least Squares Discriminant Analysis (PLSDA) performed best with 99.6% correct classification of species in the test set compared with 96.7% for Principal Component Linear Discrimination Analysis (PCLDA). Kohenen and Feed-forward artificial neural networks compared well with the PLSDA, giving 98.4 and 99.2% correct classification, respectively.

Investigation into the subambient behavior of aqueous mannitol solutions using temperature-controlled Raman microscopy.

The aim was twofold; to demonstrate the ability of temperature-controlled Raman microscopy (TRM) to locate mannitol within a frozen system and determine its form; to investigate the annealing behavior of mannitol solutions at -30 degrees C. The different polymorphic forms of anhydrous mannitol as well as the hemihydrate and amorphous form were prepared and characterized using crystal or powder X-ray diffractometry (XRD) as appropriate and Raman microscopy. Mannitol solutions (3% w/v) were cooled before annealing at -30 degrees C. TRM was used to map the frozen systems during annealing and was able to differentiate between the different forms of mannitol and revealed the location of both beta and delta polymorphic forms within the structure of the frozen material for the first time. TRM also confirmed that the crystalline mannitol is preferentially deposited at the edge of the frozen drop, forming a rim that thickens upon annealing. While there is no preference for one form initially, the study has revealed that the mannitol preferentially transforms to the beta form with time. TRM has enabled observation of spatially resolved behavior of mannitol during the annealing process for the first time. The technique has clear potential for studying other crystallization processes, with particular advantage for frozen systems.

The use of Raman microscopy to determine and localize vitamin E in biological samples.

Alpha-tocopherol (aT), the predominant form of vitamin E in mammals, is thought to prevent oxidation of polyunsaturated fatty acids. In the lung, aT is perceived to be accumulated in alveolar type II cells and secreted together with surfactant into the epithelial lining fluid. Conventionally, determination of aT and related compounds requires extraction with organic solvents. This study describes a new method to determine and image the distribution of aT and related compounds within cells and tissue sections using the light-scattering technique of Raman microscopy to enable high spatial as well as spectral resolution. This study compared the nondestructive analysis by Raman microscopy of vitamin E, in particular aT, in biological samples with data obtained using conventional HPLC analysis. Raman spectra were acquired at spatial resolutions of 2-0.8 microm. Multivariate analysis techniques were used for analyses and construction of corresponding maps showing the distribution of aT, alpha-tocopherol quinone (aTQ), and other constituents (hemes, proteins, DNA, and surfactant lipids). A combination of images enabled identification of colocalized constituents (heme/aTQ and aT/surfactant lipids). Our data demonstrate the ability of Raman microscopy to discriminate between different tocopherols and oxidation products in biological specimens without sample destruction. By enabling the visualization of lipid-protein interactions, Raman microscopy offers a novel method of investigating biological characterization of lipid-soluble compounds, including those that may be embedded in biological membranes such as aT.

Raman microscopy for the chemometric analysis of tumor cells.

Raman spectroscopy is recognized as a tool for chemometric analysis of biological materials due to the high information content relating to specific physical and chemical qualities of the sample. Thirty cells belonging to two different prostatic cell lines, PNT1A (immortalized normal prostate cell line) and LNCaP (malignant cell line derived from prostate metastases), were mapped using Raman microscopy. A range of spectral preprocessing methods (partial least-squares discriminant analyses (PLSDAs), principal component analyses (PCAs), and adjacent band ratios (ABRs)) were compared for input into linear discriminant analysis to model and classify the two cell lines. PLSDA and ABR were able to correctly classify 100% of cells into benign and malignant groups, while PLSDA correctly classified a greater proportion of individual spectra. PCA was used to image the distribution of various biochemicals inside each cell and confirm differences in composition/distribution between benign and malignant cell lines. This study has demonstrated that PLSDAs and ABRs of Raman data can identify subtle differences between benign and malignant prostatic cells in vitro.

Prediction of adipose tissue composition using Raman spectroscopy: average properties and individual fatty acids.

Raman spectroscopy has been used for the first time to predict the FA composition of unextracted adipose tissue of pork, beef, lamb, and chicken. It was found that the bulk unsaturation parameters could be predicted successfully [R2 = 0.97, root mean square error of prediction (RMSEP) = 4.6% of 4 sigma], with cis unsaturation, which accounted for the majority of the unsaturation, giving similar correlations. The combined abundance of all measured PUFA (> or = 2 double bonds per chain) was also well predicted with R2 = 0.97 and RMSEP = 4.0% of 4 sigma. Trans unsaturation was not as well modeled (R2 = 0.52, RMSEP = 18% of 4 sigma); this reduced prediction ability can be attributed to the low levels of trans FA found in adipose tissue (0.035 times the cis unsaturation level). For the individual FA, the average partial least squares (PLS) regression coefficient of the 18 most abundant FA (relative abundances ranging from 0.1 to 38.6% of the total FA content) was R2 = 0.73; the average RMSEP = 11.9% of 4 sigma. Regression coefficients and prediction errors for the five most abundant FA were all better than the average value (in some cases as low as RMSEP = 4.7% of 4 sigma). Cross-correlation between the abundances of the minor FA and more abundant acids could be determined by principal component analysis methods, and the resulting groups of correlated compounds were also well-predicted using PLS. The accuracy of the prediction of individual FA was at least as good as other spectroscopic methods, and the extremely straightforward sampling method meant that very rapid analysis of samples at ambient temperature was easily achieved. This work shows that Raman profiling of hundreds of samples per day is easily achievable with an automated sampling system.

Effect of excitation wavelength on the Raman spectroscopy of the porcine photoreceptor layer from the area centralis.

PURPOSE: Raman microscopy, based upon the inelastic scattering (Raman) of light by molecular species, has been applied as a specific structural probe in a wide range of biomedical samples. The purpose of the present investigation was to assess the potential of the technique for spectral characterization of the porcine outer retina derived from the area centralis, which contains the highest proportion of cone:rod cell ratio in the pig retina. METHODS: Retinal cross-sections, immersion-fixed in 4% (w/v) PFA and cryoprotected, were placed on salinized slides and air-dried prior to direct Raman microscopic analysis at three excitation wavelengths, 785 nm, 633 nm, and 514 nm. RESULTS: Raman spectra of each of the photoreceptor inner and outer segments (PIS, POS) and of the outer nuclear layer (ONL) of the retina acquired at 785 nm were dominated by vibrational features characteristic of proteins and lipids. There was a clear difference between the inner and outer domains in the spectroscopic regions, amide I and III, known to be sensitive to protein conformation. The spectra recorded with 633 nm excitation mirrored those observed at 785 nm excitation for the amide I region, but with an additional pattern of bands in the spectra of the PIS region, attributed to cytochrome c. The same features were even more enhanced in spectra recorded with 514 nm excitation. A significant nucleotide contribution was observed in the spectra recorded for the ONL at all three excitation wavelengths. A Raman map was constructed of the major spectral components found in the retinal outer segments, as predicted by principal component analysis of the data acquired using 633 nm excitation. Comparison of the Raman map with its histological counterpart revealed a strong correlation between the two images. CONCLUSIONS: It has been demonstrated that Raman spectroscopy offers a unique insight into the biochemical composition of the light-sensing cells of the retina following the application of standard histological protocols. The present study points to the considerable promise of Raman microscopy as a component-specific probe of retinal tissue.

A critical evaluation of Raman spectroscopy for the analysis of lipids: fatty acid methyl esters.

The work presented here is aimed at determining the potential and limitations of Raman spectroscopy for fat analysis by carrying out a systematic investigation of C4-C24 FAME. These provide a simple, well-characterized set of compounds in which the effect of making incremental changes can be studied over a wide range of chain lengths and degrees of unsaturation. The effect of temperature on the spectra was investigated over much larger ranges than would normally be encountered in real analytical measurements. It was found that for liquid FAME the best internal standard band was the carbonyl stretching vibration v(C=O), whose position is affected by changes in sample chain length and physical state; in the samples studied here, it was found to lie between 1729 and 1748 cm(-1). Further, molar unsaturation could be correlated with the ratio of the nu(C=O) to either nu(C=C) or delta(H-C=) with R2 > 0.995. Chain length was correlated with the delta(CH2)tw/v(C=O) ratio, (where "tw" indicates twisting) but separate plots for odd- and even-numbered carbon chains were necessary to obtain R2 > 0.99 for liquid samples. Combining the odd- and even-numbered carbon chain data in a single plot reduced the correlation to R2 = 0.94-0.96, depending on the band ratios used. For molal unsaturation the band ratio that correlated linearly with unsaturation (R2 > 0.99) was nu(C=C)/delta(CH2)sc (where "sc" indicates scissoring). Other band ratios show much more complex behavior with changes in chemical and physical structure. This complex behavior results from the fact that the bands do not arise from simple vibrations of small, discrete regions of the molecules but are due to complex motions of large sections of the FAME so that making incremental changes in structure does not necessarily lead to simple incremental changes in spectra.