The topology and function of long non-coding RNAs (lncRNAs) have been poorly characterized, because it usually involves the laborious generation of deletion mutants. To circumvent this drawback, Quinn et al. developed domain-specific chromatin isolation by RNA purification (dChIRP). This technique simultaneously maps RNA–chromatin interactions and chromatin-based RNA–RNA and RNA–protein interactions, from which RNA domain functionality can be inferred. In dChIRP, biotinylated antisense oligonucleotides are designed in pools that target distinct regions of the RNA, cells are crosslinked to preserve protein–nucleic acid interactions, nucleic acids are sheared (their size delineates the domain-mapping resolution) and the different oligonucleotide pools are then added separately. Oligonucleotides that hybridized to their target RNA fragments are purified, and RNA, protein and DNA are extracted. Using dChIRP, the authors found that the fly lncRNA roX1, which is essential for X chromosome dosage compensation, has many target sites clustered in a dosage compensation territory on the chromosome. roX1 is organized into modular, functional subunits — a core ('palm'), and extending 'fingers' that bind to proteins and chromatin. The finger domains rescue the phenotype of roX1-null male flies as efficiently as does full-length roX1.