mzMD: A New Storage and Retrieval System for Mass Spectrometry Data

Viewing peaks in LC–MS data sets always happens in liquid-chromatography mass spectrometry data analysis. It is challenging to develop an LC–MS data storage and retrieval tool that can get some peaks in a data window whose image presented with their intensities in a display window resembles as closely as possible that of all peaks in the data window at a speed of real-time level. We present mzMD, a new storage system that is intended to store a set of mass spectral peaks in multiple database tables with different numbers of peaks selected from the set of peaks. We present an algorithm for visualization applications to query for summaries (representative peak sets) of given data windows. We propose criteria to evaluate the quality of summaries of data windows. Experimental statistics on real LC–MS data sets verified the high speed of mzMD in retrieving high-quality summaries of data windows. The source code is freely available at https://github.com/yrm9837/mzMD-java .