Quantification of homologous proteins with unprecedented accuracy

21 Mar 2014

Nikolai Slavov
Departments of Biology & Physics, MIT, and
Department of Statistics & FAS Center for Systems Biology, Harvard

Abstract

Mass spectrometry (MS) is the most promising high-throughput method for quantifying protein levels and modifications. Existing MS methods quantify the abundances and ratios among different proteins based on the number and intensity of their peptides. However, the number of quantified peptides and their intensities depend on numerous extraneous factors (such as large variations in protein digestion and in peptide ionization) that undermine the accuracy of protein quantification. We eliminate the influence of these extraneous factors by deriving a first-principles model (Pquant) for quantifying ratios between different proteins. We prove the conditions under which Pquant has a unique solution, derive an algorithm for its optimal solution, and demonstrate experimentally unprecedented accuracy (error < 10%) in quantifying ratios between different proteins. Time permitting, I will describe experiments and results on protein translational-regulation enabled by Pquant.

current theory lunch schedule