Quantitative analysis of transcription start site selection inSaccharomyces cerevisiaereveals control by DNA sequence, RNA Polymerase II activity, and NTP levels

Abstract

DNA sequence at Transcription Start Sites (TSSs) is a key determinant of initiation by RNA Polymerase II (Pol II). To function as a TSS, an initiation compatible sequence must be specified by a promoter in an appropriate chromatin context. We report the development of a method for quantitative analysis of transcription initiation by Pol II that involves construction of DNA libraries of barcoded promoter variants, production of RNA transcripts, and analysis of transcript 5’ ends and transcript yields (Pol II MAssively Systematic Transcript End Readout, “Pol II MASTER”). Using Pol II MASTER, we measure the efficiency of transcription initiation during “promoter scanning” by Saccharomyces cerevisiae Pol II for ~1 million unique TSS sequences. Furthermore, we employ Pol II MASTER to determine how Pol II activity, known to widely alter TSS selection in vivo, alters TSS efficiencies across our promoter variants. Pol II MASTER recapitulates known critical qualities of Saccharomyces cerevisiae TSS −8, −1, and +1 positions while demonstrating that surrounding sequences modulate initiation efficiency over a wide range. We discover functional interactions between neighboring sequence positions, indicating that adjacent positions likely function together. We demonstrate that initiation efficiencies are altered for +1 A TSSs relative to +1 G TSSs when Pol II activity is perturbed through genetic means. Pol II MASTER provides data for predictive models of TSS initiation efficiency at genomic promoters.

Combining TSS-MPRA and sensitive TSS profile dissimilarity scoring to study the sequence determinants of transcription initiation

“…Together with a 10-nt A-rich tract, this limited the effective insert size of the promoter portion of each synthetic 200-nt oligonucleotide to 118 nucleotides ( Supplementary Figure S1B ). In contrast, the MASTER method ( 26 , 32 ) uses randomized oligonucleotides (7–10 nt random sequence at the TSS of a minimal promoter and 15–20 nt random sequence downstream of the TSS) ( Supplementary Figure S1B ). This provides full coverage of all possible sequence variants but is only feasible for short stretches of random sequence (at 100x RNA-seq coverage, the diversity of e.g.…”