Molecular signatures collected from intact tissue sections by MALDI imaging mass spectrometry (MALDI IMS) have shown high potential for use as a prognostic or diagnostic pathology tool in the clinical setting. A major obstacle to the widespread deployment of MALDI IMS, however, is the difficulty of identifying the proteins contributing to the signatures. Researchers have tried a number of approaches, including in situ digestion, MALDI TOF/TOF tandem mass spectrometry, and top-down proteomics on specific image regions. In preliminary work, we have obtained promising experimental results using top-down proteomics on intact proteins in the 2 - 20 kDa range. However, the lack of successful algorithms and software to identify the proteins in IMS mass signatures poses a major bottleneck. In particular, available top-down proteomics software relies heavily on high-accuracy mass spectrometry. The requirement for high accuracy precludes the use of some of the most sensitive mass analyzers such as linear ion traps, especially useful for these very small and complex samples. Protein Metrics Inc. is a new software company building on six years of algorithms and software research at Palo Alto Research Center. We plan to extend Byonic, our next- generation proteomics search engine, to intact proteins up to about 20 kDa. For proteins larger than 20 kDa, we will also build software for middle-down proteomics, specifically for assembling large peptides (2 - 20 kDa) produced by limited digestion to recover the identity of the intact proteins observed in IMS. The proposed Phase I feasibility study will allow us to perform controlled studies to determine the best experimental and bioinformatics approaches. Phase II will then build commercial-grade software. The proposed project will advance the state of the art in imaging mass spectrometry. Translation of imaging mass spectrometry to routine clinical pathology use will advance the state-of-the-art in disease diagnosis and treatment, and advance medical imaging and public health.
The project will develop commercial software that will improve our ability to identify the proteins and modifications represented in imaging mass spectrometry molecular signatures. Project success will make imaging mass spectrometry much more useful as a clinical pathology tool.