Structural principles of transcriptional collisions: SEND-Seq dataset (split, paired end reads)

Raw data of a simultaneous-end sequencing dataset for "Structural principles of transcriptional collisions". Sequencing was performed on an Illumina MiSeq instrument in a paired-end mode (150 nt x2). These data have undergone some processing to split the 5' and 3' ends of the individual sequenced RNAs (R1 and R2, respectively) according to the method described in PMID: 31308523. Sample Identification Part 1/2 are from different sample replicates No effect was observed when GreB was added after the fact, however this is likely because it was done in a no-Mg2+ containing buffer which would negate any ability of GreB to cleave backtracked RNA - because of this samples S3+S4 and S5+S6 were pooled in the data analysis pipeline. S1-S1 Both Promoters, Scrambled Hairpin, No Pulldown S2-S2 Both Promoters, YnaJ-UspE Hairpin, No Pulldown S3-S3 Both Promoters, YnaJ-UspE Hairpin, Plus Pulldown S4-S4 Both Promoters, YnaJ-UspE Hairpin, Plus Pulldown, GreB wash after (no Mg present) S5-S5 Both Promoters, Scrambled Hairpin, Plus Pulldown S6-S6 Both Promoters, Scrambled Hairpin, Plus Pulldown, GreB wash after (no Mg present) S7-S7 Reverse Promoter Only S8-S8 Forward Promoter Only Processing pipeline used in the manuscript Reads were barcode-sorted to the correct template and experimental condition. The separate 5’ and 3’ end reads of each RNA sequence were aligned to both the forward and reverse candidate RNA sequences using a local pairwise aligner in the Biopython module (match_score=2, mismatch_score=-1, open_gap_score=-5, extend_gap_score=-0.5). Alignment scores were calculated by the summation of the two best candidate positions. The higher-scoring alignment was taken, as long as the distance between the 5’ and 3’ ends was greater than 15 nucleotides, as all the RNAs should be at least 17 nucleotides from the A/G/UTP only transcription initiation walkout before subsequent purification. Unmapped reads either had the wrong direction of 5’ and 3’ alignment or were too short. Furthermore, RNAs that did not initiate from within a 10-nucleotide range near the transcription start site were filtered out. These lost reads upon filtering could come from RNAP initiation from the end of the DNA template, RNA degradation of the 5’ end, or non-templated addition at the 3’ end. The Python scripts used for data analysis are included in this deposition.