DRIPc-seq, ChIP-seq, and RNA-seq read mapping, peak calling, annotation, comparison, and visualization

Sequenced paired-ends reads were subjected to a quality control pipeline using the FASTQ Toolkit V.1.0.0 software (Illumina) and then mapped to the Saccharomyces cerevisiae reference genome sacCer3 using the Rsubread V2.0.1 software package with unique=TRUE parameter⁷¹. For DRIPc-seq, mapped reads were assigned to Watson or Crick strand using SAMtools V1.10⁷². Peak calling on DRIPc-seq data was performed with chromstaR V1.12.0 software package⁴³. A multivariate analysis considering only peaks present in the two replicates (S-phase experiments) and with a value of maximum posterior in the peak cut-off of 0.99 999 was performed. Peaks smaller than 100 bp were discarded. For comparative analysis, regions covered by peaks in the two conditions that are being compared were merged and fused when closer than 200 bp distance using BEDtools V2.27.1⁷³. The differential enrichment of these regions in each condition was performed using csaw V1.20.0 software package⁷⁴. First we counted count reads in full genome using windowCounts() with bin = TRUE and width = 200 parameters. Then normalization factors were calculated using normOffsets(). After that, estimateDisp(), glmQLFit() and glmLRT() from edgeR package (v3.20.9), was used in order to calculate log2FC and p-value of the peaks. R-loop enriched regions were established selecting those peaks whose DRIPc signal fold change was higher than 1.2 X (2X for G1 data sets) and the—log 10 (p-value) was higher than 0.6 (1 for G1 data sets). After that, R-loop enriched regions in hpr1-aid and sen1-aid conditions were merged and fused again when closer than 200 bp distance using BEDtools⁷³. The differential enrichment of these regions in each condition was performed the same way as before and R-loop enriched regions were divided in “hpr1-aid specific”, “sen1-aid specific” and “common” according to the same criteria as before. In order to compare G1 and S phase conditions, since they differ in the number of replicas, we checked if R-loop-gain peaks detected in each condition overlap with one another, defining three categories: “G1 R-loop-gain peaks”, “G1-S R-loop-gain peaks”, and “S R-loop-gain peaks”. In the S-phase analysis, enriched and specific peaks were annotated to the genomic features retrieved from the Saccharomyces Genome Database⁷⁵ where they overlap or were closer than 200 bp upstream using ChIPpeakAnno V3.28.1 software package⁷⁶, allowing each peak to be annotated to more than one genomic feature.

Comparison of WT expression levels and R-loop-enriched genes. RNA-seq data from ref. ⁷⁷ was used. Samples were mapped to the Saccharomyces cerevisiae reference genome sacCer3 using the Rsubread V2.0.1 software package with unique=TRUE parameter⁷¹. RPKM normalization method was performed using windowCounts() tool from csaw software package⁷⁴.

Coverage profiling of ChIP-seq and DRIPc-seq were obtained using bamCoverage tool from deepTools V3.4.3⁷⁸. Coverage plots were represented in the 5′ to 3′ direction. A bin size of 10 and normalization by RPKM were used. DRIPc plots show the average signal of the two replicates, if applicable. In order to compare DRIPc peaks with Rad52 ChIP-seq signal, data from ref. ⁴⁶ were used.

Genome example regions were plotted using IGV V2.8.2 software⁷⁹. The background was removed from DRIPc tracks, so only regions considered as peaks were plotted. Also, WT H2AP ChIP-seq signal is subtracted from the mutants signal. In order to compare properly the samples treated or not with RNH1, we applied a scale factor to RNH-treated samples based on the ratio between the uniquely mapped reads and the total reads of each sample. Early ARS coordinates from ref. ⁸⁰ were used. Only protein-coding genes located closer than 1 kb from the ARS midpoint were considered. These genes were split into codirectional or head-on genes according to their orientation with respect to the fork’s orientation. Then, the percentage of these genes, which were R-loop enriched in each condition, was plotted.