MALDI MSI is a sensitive, high-throughput technique to image the molecular makeup of tissues
MALDI mass spectrometry presents a powerful platform for the analysis of a variety of different endogenous and exogenous molecules directly on cells or in tissue sections. It has femtomolar sensitivity to a wide range of analytes, including lipids, drugs and metabolites, proteins, peptides, carbohydrates, polymers and large organic compounds. It reveals the complex metabolic signatures present in tissue in response to disease and treatment.
In MALDI MSI, a matrix compound is applied to the sample surface, typically by a robotic sprayer or a sublimation technique. Analyte ionisation occurs when the matrix is irradiated by a focused laser beam. The energy from the laser is absorbed by the matrix, resulting in a rapid desorption of the molecules from the radiated spot. A mass spectrum of the ions produced may then be recorded. A molecular image may be formed by firing the laser in an ordered array on the sample or continuously rastering the beam across the sample surface, with typical pixel sizes of 1-50 µm.
We are leading the understanding fundamental aspects of MALDI MSI and the associated metrology challenges, including laser variables, matrix chemistry and deposition technique. We are also focusing on improving throughput, sensitivity and lateral resolution, as well as methods of quantification and normalisation.
These fundamental studies complement application of MALDI MSI to answer biological questions from collaborators and stakeholders from academia and industry This includes the localisation and quantification of drugs and metabolites and the identification of disease biomarkers.
The size and complexity of MSI data presents a challenge to data analysis. We are developing tools to convert, process and visualise MSI data sets from multiple instruments and techniques using memory efficiency process. With potentially thousands of m/z peaks per pixel, identifying ion images that reflect the pathology or anatomy in the sample is challenging. Multivariate techniques such as principal components analysis or non-negative matrix factorisation, and segmentation algorithms such as k-means, enable effective and efficient identification of ions of interest and their distribution.