Proteins are the workhorse molecules of the cell. They are encoded in an organism's DNA, and they play an important role in virtually every biological process in an organism. Measuring differences in their relative levels across experimental conditions can lead to novel biological insights. The dominant data-intensive, or “big data,” technology for comparing thousands of protein levels across experiments is a technology called liquid chromatography coupled mass spectrometry (LC-MS). LC-MS generates gigabytes of data in the form of millions of mass spectra in a single instrument run.  Converting these mass spectra into interpretable quantitative output that show differences in amounts of proteins across samples presents a substantial computational challenge. In my talk, I provide background on how LC-MS is applied to generate massive data sets from input protein samples. Then, I describe a series of algorithms that use techniques from computational geometry to significantly improve the speed and quality of LC-MS data analysis. I show how these algorithms can be applied to measure the differential expression of two protein variants in an individual. This measurement is based on a computational and biological observation that overcomes a central limitation of this technology. This new measurement and new application of LC-MS allows a researcher to pinpoint protein-levels affected by genetic differences. In a proof of concept study, I show how this measurement opens a new window into how genetic differences in an organism’s DNA impact protein levels.