Hi I have called some H3K4me3 peaks in HCT116 cells. I have cells treated with DMSO or the experimental compound. I intend to cluster these peaks by log fold change after compound treatmen and use deeptools for doing that. My approach was to use deeptools. in the reference point mode with center of the peak set as reference point. Below are my two scripts to generate the matrix and then do the clustering while plotting the data. Can someone please comment if my approach is correct?
computeMatrix reference-point --referencePoint center --beforeRegionStartLength 5000 --afterRegionStartLength 5000 --binSize 10 \ --scoreFileName HCT116_10cm4M_2926_60uM_DMSO_075per_H3K4me3_518_121919_logfc.bw \ --regionsFileName HCT116_10cm4M_DMSO_075per_H3K4me3_518_121919.macsPeaks_peaks.bed \ --outFileName HCT116_10cm4M_2926_60uM_DMSO_075per_H3K4me3_518_121919_logfc.gz \ --missingDataAsZero -p 8 \
plotHeatmap -m HCT116_10cm4M_2926_60uM_DMSO_075per_H3K4me3_518_121919_logfc.gz --kmeans 3 -out HCT116_10cm4M_2926_60uM_DMSO_075per_H3K4me3_518_121919_logfc_3C.pdf --colorList blue,white,red --samplesLabel "2926/DMSO" --regionsLabel "C1" "C2" "C3" --outFileSortedRegions HCT116_10cm4M_2926_60uM_DMSO_075per_H3K4me3_518_121919_logfc_3C.bed --heatmapWidth 10 --heatmapHeight 30 --missingDataColor "white" --yMin -1 --yMax 1 --zMin -1 --zMax 1 --xAxisLabel "" --sortRegions descend --sortUsing sum
