Multivariate Analysis of TOF-SIMS Data for Applications in Tissue Engineering and Quantifying Biomolecules

The ability to map the chemical composition at the surface of the sample with as good as sub-μm lateral resolution and without the need for labels renders TOF-SIMS a powerful approach for studying the compositions of cells and other biological materials.  Molecular ions and nearly intact molecular fragment ions that are unique to a single molecule are often employed to unambiguously identify components of interest in the sample.  Though more difficult to interpret, compositional information is also encoded in the low-mass (m/z<300) peaks that are abundant in the spectra of biological materials. Multivariate analysis of TOF-SIMS data enables utilizing the abundant, low-mass for sample classification or the identification of sample components.   Here we present our use supervised multivariate analysis of TOF-SIMS data for two applications: classifying cells according to differentiation stage, and measuring cholesterol concentration in model lipid membranes.  The factors that affect the outcome of the analysis, and the improvements that would expand its utility will be discussed.