Stress-Induced Transcriptional Memory Accelerates Promoter-Proximal Pause-Release and Decelerates Termination over Mitotic Divisions

Anniina Vihervaara; Dig Bijay Mahat; Samu V. Himanen; Malin A.H. Blom; John T. Lis; Lea Sistonen

Pol II densities as bigWig and bedgraph files were visualized with Integrative Genomics 605! Viewer (IGV; Thorvaldsdóttir et al., 2013) and an in-house browser (Hojoong Kwak, Cornell 606! University, Ithaca, NY, USA). The read counts in defined genomic regions were obtained, and 607! composite profiles generated using bigWig package 608! (https://github.com/andrelmartins/bigWig/). The average intensities in composite profiles were 609! queried in 20-nt, 10-nt or 1-nt bins. The shaded areas display 12.5-87.5% fractions of the data 610! in each queried window. To generate an average profile of gene bodies with different lengths, 611! 1/500th of the gene body length was set to the bin size, after filtering out short genes where 612! the bin would have been less than 1 nt. To identify genes with decreased Pol II density in 5'-and increased density in 3'-region, genes 616! were first divided into three distinct regions: 1) 5'-coding region comprising 1000 nt 617! downstream of the mid coordinate between Pol II pause sites of divergent transcription, 2) 618! gene body, measured from +1000 nt from the mid of the pause sites to -1000 nt from the CPS, 619! and 3) downstream, +100 nt to +6000 nt, of the CPS. PRO-seq reads in each region were 620! measured, after which the read count in the preconditioned 60-minute heat shock sample was 621! deduced from the respective read count in the unconditioned 60-minute heat shock sample.