Cut&Run and heatmap

Question

[qudrat.nii]

Hello guys,

I am doing cut&run analysis for some transcription factors and I have three replicates of BAM files and three replicates of MACS2 output. For downstream studies should I merge three replicates of bam file to plot heatmap? In case replicates are not highly correlated what should I do?

[ATpoint]

Do not delete questions that received answers, it's plain disrespectful towards the users who ivnested time into your question. After all, biostars is a knowledge repository so the Q/A might serve others in the future. It's all anonymous here if you set your account accordingly, so don't worry about the irrational fear of people pointing fingers at you, it's not a realistic issue.

[sogand]

Hi all,

I am new to cut&run or any peak-related analysis and appreciate any guidance on my issue here: I have analyzed peakcalling .txt files that are ready for plotting and visualization and I want to use deeptools with linux command line for meta-profile heatmaps and other possible tools, since my files are all .txt and contain peak information with 19 columns as below:

PeakID (cmd=annotatePeaks.pl CHA-1_peaks.narrowPeak hg38) | Chr | Start | End | Strand | Peak Score | Focus Ratio/Region Size | Annotation | Detailed Annotation | Distance to TSS | Nearest PromoterID | Entrez ID | Nearest Unigene | Nearest Refseq | Nearest Ensembl | Gene Name | Gene Alias | Gene Description | Gene Type

can somebody help me with the common steps and workflow to convert .txt files to bed and bigwig files to do further analysis like DE and also visualize by using the deeptools and IGV?

my thought is to : 1-Create a BED file from .txt files by the command below: awk 'BEGIN{OFS="\t"} NR>1 {print $2, $3, $4, $6}' yourfile.txt > peaks.bed

2- Sort the BED File sort -k1,1 -k2,2n peaks.bed > sorted_peaks.bed

3- Create a BedGraph bedtools genomecov -i sorted_peaks.bed -g hg38.chrom.sizes -bg > peaks.bedGraph

4- Convert BedGraph to BigWig bedGraphToBigWig peaks.bedGraph hg38.chrom.sizes peaks.bw

Then I can go further using computeMatrix, plotProfile, plotHeatmap, for further visualization steps ... Do you think this is reasonable??

I really appreciate any guidance.

Best Sogand

[gglim]

I'll take a look at these peaks in IGV with the signal files and set a threshold for filtering.

Answer 1

[rpolicastro]

deepTools is great for general plot making for CUT&RUN, ChIP-seq, and the like. If you have any other plots you want to make in mind you should include an example so we can give more specific advice.