A Statistical Pipeline for an Activity-Based Protein Profiling Assay

• Activity Number: 193
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): William Forrest* and Johanna Heideker and Taylur Ma and Kebing Yu and Jennie Lill and Ingrid Wertz
• Companies: Genentech, Inc. and Genentech, Inc. and Genentech, Inc. and Genentech, Inc. and Genentech, Inc. and Genentech, Inc.
• Keywords: enzyme ; proteomics ; spectrometry ; assay ; bioinformatics ; linear

Abstract

Intracellular enzymes are a diverse source of targets for new therapies. Some enzymes, however, are relevant as drug targets only under specific "activated" conditions. Activity-based protein profiling (ABPP) is an assay strategy in functional proteomics to isolate for study an active subset of enzymes, allowing identification of candidate targets against a background of relatively inactive (albeit perhaps more numerous) similar enzymes.

We provide a brief overview of the assay and proteomic technologies, then trace development of a statistical analysis pipeline for an ABPP screen. Strategies for visualization and normalization are considered. Finally, researchers' questions are addressed using established statistical tools rooted in linear modeling to quantify salient experimental effects with standard errors, along with strategies to rank results for follow-up. The central roles of experimental design, graphical investigation, and open-source software are emphasized.