PAIR-MaP visualizes RNA base pairing complexity in human and bacterial non-coding RNAs

Structure probing experiments were performed on in vitro transcripts and E. coli and human cell cultures under natively extracted (cell-free) and in-cell conditions to benchmark the performance of the newly introduced PAIR-MaP correlated chemical probing strategy for detecting RNA duplexes. Multiple-hit dimethyl sulfate (DMS) probing was done using new buffer conditions that facilitate DMS modification of all four nucleotides. Overall design: In vitro transcribed RNA, cell-free total RNA from E. coli and human Jurkat cells, and living E. coli and human Jurkat cells were treated with DMS in bicine (pH 8.0) probing buffer. Untreated control samples were also collected in parallel by substituting neat ethanol for DMS. MaP reverse-transcription was used to convert DMS-modified RNA into cDNA using random 9mers (16S and 23S rRNAs) or specific primers (all other RNAs). Sequencing libraries were constructed using a Nextera strategy (16S and 23S rRNAs) or two-step PCR strategy (all other RNAs) and sequenced on an Illumina MiSeq. When necessary, libraries were resequenced multiple times to obtain sufficient reads. The ShapeMapper (v2.1.4) pipeline was used to align reads and call modification-induced mutations. PairMapper (v1.0) was then used to calculate normalized DMS reactivities and detect correlated base-paired positions. Fasta files used for alignment of each sample are provided as supplementary data files. Final processed DMS reactivity files and PAIR-MaP files are provided as supplementary files for RNA and condition.