A Method to Estimate the Distribution of Proteins across Multiple Compartments Using Data from Quantitative Proteomics Subcellular Fractionation Experiments

Knowledge of cellular location is key to understanding the biological function of proteins. One commonly used large-scale method to assign cellular locations is subcellular fractionation, followed by quantitative mass spectrometry to identify proteins and estimate their relative distribution among centrifugation fractions. In most of such subcellular proteomics studies, each protein is assigned to a single cellular location by comparing its distribution to those of a set of single-compartment reference proteins. However, in many cases, proteins reside in multiple compartments. To accurately determine the localization of such proteins, we previously introduced constrained proportional assignment (CPA), a method that assigns each protein a fractional residence over all reference compartments (Jadot et al. Mol. Cell Proteomics 2017, 16(2), 194-212. ). In this Article, we describe the principles underlying CPA, as well as data transformations to improve accuracy of assignment of proteins and protein isoforms, and a suite of R-based programs to implement CPA and related procedures for analysis of subcellular proteomics data. We include a demonstration data set that used isobariclabeling mass spectrometry to analyze rat liver fractions. In addition, we describe how these programs can be readily modified by users to accommodate a wide variety of experimental designs and methods for protein quantitation.

Automated summary

Knowledge of cellular location is crucial for understanding protein function. Subcellular fractionation followed by mass spectrometry is a commonly used method for assigning cellular locations to proteins. However, proteins can reside in multiple compartments, which requires a more accurate method for localization. In this article, the constrained proportional assignment (CPA) method is introduced, along with data transformations and R-based programs for analyzing subcellular proteomics data.