High‐precision mapping of the protein interaction network for the human transcription machinery reveals a novel class of cellular regulatory factors

Abstract

We performed a survey of soluble human protein complexes containing components of the transcription and RNA processing machineries using protein affinity purification coupled to mass spectrometry. Many identified interaction partners were targeted in reciprocal tagging experiments in order to confirm some interactions and to enrich the dataset. High‐confidence interactions were selected computationally using an algorithm that we developed and trained using machine learning to minimize the rate of both false‐positives and false‐negatives. The data produced with 100 affinity tagged proteins was used to (1) build a high‐definition map of interactions that connect components of the transcription and RNA processing machineries in human cells; (2) show that transcription and RNA processing factors from the soluble cellular fraction are associated with proteins that specifically regulate the formation (e.g. assembly, localization and/or stability) of protein complexes; and, (3) assign a putative function to a number of previously‐uncharacterized proteins on a ‘‘guilt by association’’ basis. A number of previously‐uncharacterized proteins that we further characterized functionally and biochemically define a novel class of regulatory factors that target RNA polymerase II and other transcription factors prior and/or after the transcription reaction on chromatin DNA.