Understanding gene regulation is a central goal of molecular biology, which remains only partially solved as DNA sequence alone cannot explain cell-type- or context-specific gene activity. Instead, the coordinated activities of thousands of regulatory proteins ensure expression of the correct set of cell-type-specific proteins and enable rapid adaptation to changing conditions. Previous attempts to map regulatory proteins have been limited by the ‘one-by-one’ nature of traditional methods. Current reference maps exist for a narrow set of proteins in tumour cell lines (for example, K562 and GM12858), yet have been criticized for high costs (hundreds of millions of dollars), lengthy timelines (≥10 years) and narrow returns (for example, most proteins and cell types are left unexplored). Moreover, reference datasets are of limited use for studying specific cell types, processes or diseases, and generating additional data in those contexts is not feasible because of technical challenges.

ChIP-DIP can be applied to comprehensively profile the regulatory landscape in individual cell types, facilitating the discovery of combinatorial binding patterns and/or regulatory networks involving tens to hundreds of regulatory proteins. As a proof of concept, we used ChIP-DIP to identify combinations of histone modifications that distinguish regulatory elements with varied functions, including maintaining current gene activity and cell state, setting future differentiation or development potential, and enhancing stress responsiveness. Because it profiles many proteins in a single sample, ChIP-DIP is ideal for studying biological systems with limited cell numbers (for example, rare cell types or patient-derived cells) or with dynamic changes (for example, post stimulation, during development or differentiation or during disease progression).