Visualizing ToF-SIMS Hyperspectral Imaging Data Using Color-Tagged Toroidal Self-Organizing Maps

Abstract

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) is a powerful surface characterization technique capable of producing high spatial resolution hyperspectral images, in which each pixel comprises an entire mass spectrum. Such images can provide insight into the chemical composition across a surface. However, issues arise due to the size and complexity of the data produced. Data are particularly complicated for biological samples, primarily due to overlapping spectra produced by similar components. The traditional approach of selecting individual ion peaks as representative of particular components is insufficient for such complex data sets. Multivariate analysis (MVA) can help to overcome this significant hurdle. We demonstrate that Kohonen self-organizing maps (SOMs) with a toroidal topology can be used to analyze a ToF-SIMS hyperspectral imaging data set and identify spectral similarities between pixels. We present a method for color-tagging the toroidal SOM output, which reduces the entire data set to a single RGB image in which similar pixelsbased on their associated mass spectraare assigned a similar color. This method was exemplified using a ToF-SIMS image of dried large multilamellar vesicles (LMVs), loaded with the antibiotic cefditoren pivoxil (CP). We successfully identified CP-loaded and empty LMVs without the need for any prior knowledge of the sample, despite their highly similar spectra. We also identified which specific ion peaks were most important in differentiating the two LMV populations. This approach is entirely unsupervised and requires minimal experimenter input. It was developed with the aim of providing a user-friendly yet sophisticated workflow for understanding complex biological samples using ToF-SIMS images.

Controlling orientation, conformation, and biorecognition of proteins on silane monolayers, conjugate polymers, and thermo-responsive polymer brushes: investigations using TOF-SIMS and principal component analysis

“…PCA could also be applied to an analysis of TOF-SIMS 2D images [33,[58][59][60] and even 3D TOF-SIMS sputtering data [50]. Other multivariate methods already applied for TOF-SIMS data analysis are non-negative matrix factorization (NMF) [61,62], the k-means cluster method [63], discriminant analysis [64,65], and artificial neuronal networks [55,66] involving self-organizing maps [67][68][69][70][71].…”

Understanding mass spectrometry images: complexity to clarity with machine learning

“…ToF-SIMS data. [12,13,[62][63][64][65][66][67] Initially, investigations were limited to 1D spectra, focusing on the performance of the SOM-with reference to techniques such as PCA and MCR-for the unsupervised discrimination of highly similar spectra. Results indicated that the SOM, because of its inherent non-linearity, performed better than its linear counterparts for this task, exhibiting greater tolerance for noise and apparently requiring minimal data preprocessing.…”

“…More recently we extended the use of toroidal SOMs to the analysis of hyperspectral ToF-SIMS images. This approach is summarized in Gardner et al [12,13] and uses the same principles described earlier for MALDI MSI data. [1] However, unlike for the 3D SOM, it is not possible to apply a linear color scheme to the neurons in a toroidal SOM.…”

“…Perhaps most notable of the latter has been the application of non-linear machine and deep learning methods to MSI data. [1][2][3][4][5][6][7][8][9][10][11][12][13][14] These methods have become crucial tools in enabling a deeper and more objective understanding of MSI data, while also reducing experimenter bias and allowing more robust data processing workflows.…”

“…It was noted that a key fragment ion associated with the prodrug, C 5 H 9 O 2 − , was also produced by lipid fragmentation and hence was associated with the liposome bilayers. [12] Focusing on this fragment alone would have suggested the drug was well distributed among the liposomes in the sample. However, more sophisticated and unsupervised machine learning approaches revealed that this was not the case.…”