Prof. Michael MacCoss
Professor of Genome Sciences, University of Washington
Having your cake and eating it too: Can proteomics be both comprehensive and quantitative?
Proteomics technology has improved dramatically over the last decade. The technology developments have largely been directed around instrument hardware, where instruments have been developed that scan faster, are more sensitive, and have greater mass measurement accuracy. However, the basic workflow has remained largely unchanged — mass spectrometers are directed toward the acquisition of tandem mass spectra by applying data dependent acquisition (DDA) on the most abundant molecular species eluting from a chromatography column. Alternatively, efforts have focused on the acquisition of mass spectrometry data on target peptides of interest. An alternative to DDA is an acquisition approach known as data independent acquisition (DIA). Data independent acquisition (DIA) acquires comprehensive MS/MS information in a single LC-MS/MS run using a repeated cycle of wide-window MS/MS scans. With improvements in instrument hardware and instrument control software, the practical experimental difference between a targeted and discovery proteomics is beginning to blur. DIA aims to combine the unbiased acquisition capabilities of data dependent acquisition with the sensitivity and reproducibility of a fully targeted data acquisition. However, performing analyses on these new unbiased acquisition strategies provides a significant change from the traditional proteomics workflow and have required the development of novel computational strategies to analyze, visualize, and interpret these data. I will present work illustrating our efforts in the development of new proteomics acquisition strategies and workflows. Additionally I provide a vision for challenges that still need to be overcome before these analyses become routine.
After completing his postdoctoral fellowship with John Yates at the Scripps Institute, Michael MacCoss joined the University of Washington in 2004 as an Assistant Professor of Genome Sciences, and was promoted to full Professor in 2014. His research focuses on the development and application of mass spectrometry based technologies for the high throughput characterization of complex protein mixtures.
He has made a number of seminal contributions to the field of proteomics, and most importantly on the development of methods and technologies for peptide quantification and analysis software. Software tools developed by the MacCoss’s laboratory facilitate many aspects of mass spectrometry data analysis, and include tools for liquid chromatography mass spectrometry (LC-MS) feature finding, spectrum library searching, peak detection, post-processors for peptide database searching. He also developed the Percolator algorithm, which improves peptide identifications from proteomic analyses through semi-supervised machine learning. Percolator became widely adopted in part because of its use of a liberal open source license that encourages companies to build on Percolator and promotes incorporation into commercial software packages (e.g. Mascot and Proteome Discoverer). His lab also developed and continue supporting an integrated set of software tools called Skyline that enables methods to be easily transferred and tested across labs and instrument platforms. More recently, his lab also substantially advanced the new area of data-independent MS analyses. Mike MacCoss’s philosophy on making software freely available, providing support, and enabling regular training for proteomics has greatly benefited the field of protein sciences.
Date(s) - May 7, 2018
6:00 pm - 9:00 pm
Emplacement / Location
Morris and Rosalind Goodman Agora