Anal. Chem. 2003, 75, 460-468
High Mass Measurement Accuracy Determination for Proteomics Using Multivariate Regression Fitting: Application to Electrospray Ionization Time-Of-Flight Mass Spectrometry Eric F. Strittmatter,* Nestor Rodriguez,† and Richard D. Smith*
Environmental and Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, MSIN: K8-98, Richland, Washington 99352
Important factors that limit the mass measurement accuracy from a mass spectrometer are related to (1) the type of mass analyzer used and (2) the data processing/ calibration methods used to obtain mass values from the raw data. Here, two data processing methods are presented that correct for systematic deviations when the mass of ions is measured using a time-of-flight (TOF) mass spectrometer. The first fitting method is one where m/z values are obtained from fitting peak distributions using double Gaussian functions. A second calibration method takes into account the slight nonlinear response of the TOF analyzer in addition to the drift in the calibration over time. Using multivariate regression, both of these two effects can be corrected for using a single calibration formula. Achievable performance was evaluated with a trypsin digestion of serum albumin and proteins from the organism D. radiodurans that was analyzed using gradient reversed-phase liquid chromatography combined with an electrospray ionization orthogonal TOF mass spectrometer. The root-mean-square deviation between the theoretical and experimental m/z values for serum albumin tryptic peptides was found to be 8 ppm using the double Gaussian-multivariate method compared to 29 ppm determined using linear calibration and normal peak centroiding. An advantage of the methods presented here is that no calibrant compounds need to be added to the mobile phase, thereby avoiding interference effects and signal suppression of analytes. Mass spectrometry-based strategies are increasingly being implemented that exploit high mass measurement accuracy (MMA) to solve complex biological problems. Mass spectrometry, in combination with high-performance separations, is playing an important role in proteomicssthe identification of expressed proteins and the elucidation of their signaling pathways.1 High MMA for a polypeptide can be crucial for identifying its sequence among the thousands of possibilities that may be present in a database.2-7 Conrads et al.5 have examined the role of MMA for † Present address: West Pharmaceuticals, Gordon Drive, Lionville, PA 19341. (1) Aebersold, R.; Goodlett, D. R. Chem. Rev. 2001, 101, 269-295.
460 Analytical Chemistry, Vol. 75, No. 3, February 1, 2003
effective identification of tryptic peptides from various organisms based solely on high-accuracy mass measurements. For example, if a MMA of 1 ppm is obtainable, then ∼50% of the tryptic peptides of Saccharomyces cerevisae can potentially be uniquely identified.5 A MMA of 10 ppm only allows ∼5% of the tryptic peptides to be uniquely identified.5 The limit in MMA obtainable from a spectrometer is traditionally recognized to be dependent on the type of mass analyzer employed. The greatest accuracy is attributed to Fourier transform ion cyclotron resonance instruments (1 ppm-1 ppb),8,9 while quadrupole and ion trap instruments afford lower routine accuracy (>100 ppm).10 The resolution and MMA ability of time-of-flight (TOF) instruments have increased throughout the past decade. Routine accuracy of
