Automated reduction and interpretation of high resolution electrospray mass spectra of large molecules

Abstract

Here a fully automated computer algorithm is applied to complex mass spectra of peptides and proteins. This method uses a subtractive peak finding routine to locate possible isotopic clusters in the spectrum, subjecting these to a combination of the previous Fourier transform/Patterson method for primary charge determination and the method for least-squares fitting to a theoretically derived isotopic abundance distribution for m/z determination of the most abundant isotopic peak, and the statistical reliability of this determination. If a predicted protein sequence is available, each such m/z value is checked for assignment as a sequence fragment. A new signal-to-noise calculation procedure has been devised for accurate determination of baseline and noise width for spectra with high peak density. In 2 h, the program identified 824 isotopic clusters representing 581 mass values in the spectrum of a GluC digest of a 191 kDa protein; this is >50% more than the number of mass values found by the extremely tedious operator-applied methodology used previously. The program should be generally applicable to classes of large molecules, including DNA and polymers. Thorough high resolution analysis of spectra by Horn (THRASH) is proposed as the program's verb.