Two-Dimensional Clustering of Spectral Changes for the Interpretation of Raman Hyperspectra

Demixing and Analysis of Complex Biological Raman Hyperspectra Based on Peak Fitting, Amplitude Trend Clustering, and Spectrum Reconstruction

To better interpret the Raman spectra from mammalian cells, it is often desirable to reduce their complexity by decomposing them into the spectral contributions from individual macromolecules or types of macromolecules. Diverse methods exist for demixing complex spectra, each with different benefits and drawbacks. However, some methods require a library of component spectra that might not be available, while others are hampered by noise and peak congestion that includes many proximal overlapping peaks. Through rapid fitting of individual peaks in every spectrum of a Raman hyperspectral data set, we have obtained individual peak parameters from which we determined the trends for all the peak amplitudes. We then grouped similar trends with k-means clustering. Then we used the peak parameters of all the peaks in a given cluster to reconstruct a spectrum representative of that cluster. This method produced spectra that were less distorted by unrelated overlapping peaks or noise, were less congested than those in the hyperspectral set, and thereby improved peak identification and macromolecule recognition. We have demonstrated the application of the method with Raman spectra from a perchlorate-polystyrene model system and extended it to complex spectra from methanol-fixed mammalian cells. We were able to recover independent spectra of perchlorate and polystyrene in the model system and spectra pertaining to individual macromolecular types (proteins, nucleic acids, lipids) from the mammalian cell data. We discuss how imperfections in spectral preprocessing and peak fitting can adversely affect the results. In summary, we have provided a proof-of-concept for a novel mixture resolution method with different attributes than extant ones.

Study of interaction in dual-species biofilm of Candida glabrata and Klebsiella pneumoniae co-isolated from peripheral venous catheter using Raman characterization mapping and machine learning algorithms

Polymicrobial biofilm infections, especially associated with medical devices such as peripheral venous catheters, are challenging in clinical settings for treatment and management. In this study, we examined the mixed biofilm formed by Candida glabrata and Klebsiella pneumoniae, which were co-isolated from the same peripheral venous catheter. Our results revealed that C. glabrata can form mixed biofilms with K. pneumoniae in vitro on peripheral venous catheters and the bottom of microplate wells, as confirmed by scanning electron microscopy. Additionally, using Raman mapping, we showed the distribution of both species in mono- and dual-species biofilms and suggested the type of microbial interaction in this polymicrobial biofilm. Finally, with the assistance of appropriate machine learning (ML) algorithms, based on identified peaks of bacteria, yeast, catheter, and Microplate mapping spectra, we develop a dedicated Raman database to detect the presence of these elements in an unknown spectrum in the future.

Assessing the quality of stored red blood cells using handheld Spatially Offset Raman spectroscopy with multisource correlation analysis

In this work we employ Spatially Offset Raman Spectroscopy (SORS) to non-invasively identify storage-related changes in red blood cell concentrate (RCC) in-situ within standard plastic transfusion bags. To validate the measurements, we set up a parallel study comparing both bioanalytical data (obtained by blood-gas analysis, hematology analysis and spectrophotometric assays), and Raman spectrometry data from the same blood samples. We then employ Multisource Correlation Analysis (MuSCA) to correlate the different types of data in RCC. Our analysis confirmed a strong correlation of glucose, methemoglobin and oxyhemoglobin with their respective bioassay values in RCC units. Finally, by combining MuSCA with k-means clustering, we assessed changes in all Raman wavenumbers during cold storage in both RCC Raman data from the current study and parallel RCC supernatant Raman data previously acquired from the same units. Direct RCC quality monitoring during storage, would help to establish a basis for improved inventory management of blood products in blood banks and hospitals based on analytical data.