Cut&RUN ACSS2 on liver tissue

Pre-processed CUT&RUN files for ACLY and GCN5 were normalized to effective genome size.For each file, background signal was calculated in heterochromatin regions using negativeATAC-seq signal mask, and subsequently removed from the overall coverage. CUT&RUNcoverage heatmaps and profiles around the TSS of expressed genes were produced usingcomputeMatrix, plotHeatmap functions from deepTools v3.5.2 package51. The list of expressedgenes was inferred from the mouse primary hepatocyte RNA-seq data using variancestabilizing transformation of the expression matrix and selecting genes with values > 0 for allreplicates. Peak calling on each replicate was performed using MACS252 v.2.2.7.1, and theconsensus overlapping peaks between all replicates were considered as reproducible for thecorresponding dataset. Number of overlapping peaks between conditions was calculated in Rwith subsetByOverlaps (GenomicRanges v.1.54.1)57, and nearest genes were annotated topeaks using biomaRt.For the correlation of ChIP peaks with gene expression, all expressed genes derived fromRNA-seq data were divided into 4 equal groups based on their expression levels (quartiles (Q)1 to 4). Promoter regions of those genes (± 1kb TSS) were extracted using the Rsubreadv.2.16.0)59 package, and the CUT&RUN signal was quantified with multiBigwigSummary fromdeepTools package