Standardization of complex biologically derived spectrochemical datasets

Abstract

Spectroscopic techniques, such as Fourier-transform infrared (FTIR) spectroscopy, are used to study the interaction of light with biological materials. This interaction forms the basis of many analytical assays used in disease screening and diagnosis, microbiological studies, forensic and environmental investigations. Advantages of spectrochemical analysis are its low cost, minimal sample preparation, non-destructive nature and substantially accurate results. However, there is now an urgent need for repetition and validation of these methods in large-scale studies and across different research groups, which would bring the method closer to clinical and/or industrial implementation. In order for this to succeed, it is important to understand and reduce the effect of random spectral alterations caused by inter-individual, inter-instrument and/or inter-laboratory variations, such as variations in air humidity and CO2 levels, and the aging of instrumental parts.Thus, it is evident that spectral standardization is crucial for the widespread adoption of these spectrochemical technologies. By using calibration transfer procedures, where the spectral response of a secondary instrument is standardized to resemble the spectral response of a primary instrument, different sources of variations can be normalized into a single model using computational-based methods, such as direct standardization (DS) and piecewise direct standardization (PDS); therefore, measurements performed under different conditions can generate the same result, eliminating the need for a full recalibration. In this paper, we have constructed a protocol for model standardization using different transfer technologies described for FTIR spectrochemical applications. This is a critical step towards the construction of a practical spectrochemical analysis model for daily routine analysis, where uncertain and random variations are present. 4 worldwide are developing spectrochemical approaches for diagnosis, discrimination and monitoring of diseases, as well as for other uses. Combination of multiple datasets would facilitate the conduction of large-scale studies which are still lacking in the field of bio-spectroscopy. Sensor-based technologiesSensor-based technologies are an integral part of daily life ranging from locating sensorbased technology, such as global positioning system (GPS) 6 , to image biosensors, such as X-rays 7-10 and γ-rays [11][12][13] , which are used extensively for medical applications. Other powerful approaches that make use of sensor-based technologies toward medical disease examination and diagnostics include circular dichroism (CD) spectroscopy 14-17 , ultraviolet (UV) or visible spectroscopy 18,19 , fluorescence 20-24 , nuclear magnetic resonance (NMR) spectroscopy 25-29 and ultrasound (US) 7,30- .Over the last two decades, optical biosensors employing vibrational spectroscopy, particularly IR spectroscopy, have seen tremendous progress in biomedical and biological research. A number of studies using the above-mentioned methods ha...

Spectrochemical differentiation of meningioma tumours based on attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy

“…The spectral data analysis was performed within a MATLAB R2014b environment (MathWorks, Natick, USA) using the Classification Toolbox for MATLAB [26]. The biofingerprint spectra (1800-900 cm −1 ) were pre-processed by Savitzky-Golay 2 nd derivative (window of 7 points, 2 nd -order polynomial fit) and vector normalisation, a common pre-process employed in biological-derived spectral data for correcting random noise and baseline distortions and to improve the signal-to-noise ratio [10,14]. An outlier test was performed using Hotelling's T 2 versus Q residual test [14], and no spectral outlier was observed in the dataset (see ESM Fig.…”

“…In order to obtain meaningful and reliable information, the IR spectra within the fingerprint region are processed through specific computational techniques, known as chemometrics. The spectral data initially undergo preprocessing techniques to correct the baseline and to remove possible physical variations not related to disease changes, and then chemometric models are built and validated, whereby possible spectral biomarkers as well as sensitivity and specificity metrics can be obtained [14]. Multivariate classification models, such as principal component analysis plus linear discriminant analysis (PCA-LDA) [15] and partial least squares plus discriminant analysis (PLS-DA) [16], are commonly employed to process IR spectral data, since these techniques allow to extract relevant spectral features associated with tumour differentiation and also to classify the samples into groups in a predictive fashion.…”

Towards screening Barrett’s oesophagus: current guidelines, imaging modalities and future developments

“…Vibrational spectroscopy techniques such as Fourier-transform infrared (FTIR) spectroscopy or Raman spectroscopy (RS) are used to study interactions of light with biological materials and are relatively novel [130]. Advantages of spectrochemical analysis are its low cost, minimal sample preparation, non-destructive nature and substantially accurate results [131].…”

A procedure for calibration transfer of DOSY NMR measurements: An example of molecular weight of heparin preparations

“…Historically, calibration transfer techniques were primarily developed for near‐infrared (NIR), infrared (IR), and Raman data . The focus of the recent research in this area was the correction of nonlinear signal response, day‐to‐day fluctuations, and disturbance factors from the gas background matrix in IR spectrometry .…”

“…Historically, calibration transfer techniques were primarily developed for near-infrared (NIR), infrared (IR), and Raman data. [7][8][9][10] The focus of the recent research in this area was the correction of nonlinear signal response, day-today fluctuations, and disturbance factors from the gas background matrix in IR spectrometry. 11 Attempts were also made to transfer multivariate models between benchtop Fourier transform and portable NIR spectrometers as it was shown for the determination of total soluble solid contents and quality control of polymorphs.…”