ChIP- seq seminar “Omics data analysis” - 2026

1. ChIP-seq seminar

“Omics data analysis”-2026

2. Practice summary

1. Peak calling. Report number of peaks - 1 point
2. QC. FRiP. Run plotCoverage with your bam and peak le. Provide plot and your observations - 1 point
3. QC. Run multiBamSummary and plotCorrelation with bams. Provide correlation plot and your
observations - 1 point
4. Create bigwig. Upload bigwig AND peak le to IGV. Report screenshots and your observations - 1 point
5. Run idr. Upload both replicates' peaks and resulting peak le to IGV (3 les in total). Provide screenshot
and your observations - 1 point
6. Bigwig around genes. Use computeMatrix and plotHeatmap to plot your signal (bigwig) around genes
(bed). Provide plot and your observations - 2 points
15
2 . 1
2

3. Data

• The data are taken from Wang et al., 2021 paper
• V is for Vegetative, S is for Streaming, M is for Mound stages
2 . 3

4. Your samples

15 . 3

5. Prepare environment

Add the path to conda executable to your PATH environment variable
export PATH="/home/i.zhegalova/anaconda3/bin/:$PATH"
source activate chipseq_add
All bam les are located at
/home/i.zhegalova/chipseq_dicty/bams as well as .bai
indexes for them.
NO NEED to copy all bam les into your folder, just specify the path to
these les OR use symlink
ln -s < path to bam file > < your bam link.bam >
! Note. Some steps can be time consumable, run them inside screen
15
2 . 4
5

6. Task 1. Peak calling. MACS2

• MACS2 is the software of choice in most cases for ChIP-seq
• macs2 requires an INPUT.bam le to build a peak model. The input
les are also located in
/home/i.zhegalova/chipseq_dicty/bams
• macs2 requires genome size, option -g 3.16E7 provide dicty's
genome size
macs2 callpeak -t EXPERIMENT.bam \
-c INPUT.bam \
-f BAM -g 3.16E7 -q 0.05 \
-n OUTPUT_NAME \
--outdir YOUR_FOLDER
? run command and report nal number of peaks (1 point)
2 . 6

7. Task 2. QC. FRiP

• FRiP = Fraction of Reads In
Peak
• ratio of those reads that map in
the peak of interest over the total
reads.
• a minimum FRIP score of 0.3
is suggested by ENCODE
2 . 7

8. Task 2. QC. FRiP

use your peak les to compute FRiP for your le
(in this command, use full path to your .bam file)
plotEnrichment -b EXPERIMENT.bam INPUT.bam --BED PEAKS.narrowPeak \
-o enr.png --smartLabels
? provide FRiP plot and your explanations (1 point)
2 . 8

9. Task 3. QC. Correlation

15 . 8

10. Task 3. QC. Correlation

• with option bins multiBamSummary computes the read coverage
per xed size genomic bin
• Analyze AT LEAST two replicates of your experiment (you are welcome
to study more). To do this, collaborate with your fellow student who has
the second replicate of the same experiment.
multiBamSummary bins -b EXPERIMENT_R1.bam EXPERIMENT_R2.bam \
-o OUTPUT.tab
2 . 10

11. Task 3. QC. Correlation

multiBamSummary generates a speci c le as output and you need to
convert it to a human-readable picture with the following command
plotCorrelation -in OUTPUT.tab \
--corMethod spearman --skipZeros \
--plotTitle "My Awseome Plot" --whatToPlot heatmap \
--colorMap RdYlBu --plotNumbers \
-o heatmap_SpearmanCorr_readCounts.png
? Are biological replicates good enough? Add a picture and your
observations (1 point)
2 . 11

12. Task 4. bigwig and peaks at IGV

• --binSize 10 option change bin size to 10 to allow a more smooth signal.
• RPGC (per bin) - number of reads per bin / scaling factor for 1x
average coverage.
• This option used instead of RPKM normalization. The scaling factor
used is the inverse of the sequencing depth computed for the sample to
match the 1x coverage.
• This option requires --effectiveGenomeSize
bamCoverage --bam EXPERIMENT.bam -o OUTPUT.bw --binSize 10 --normalizeUsing RPGC
--effectiveGenomeSize 31600000
bamCoverage --bam CONTROL.bam -o OUTPUT.bw --binSize 10 --normalizeUsing RPGC
--effectiveGenomeSize 31600000
2 . 12

13. Task 4. bigwig and peaks at IGV

• go to IGV.org -> IGV in web browser
• you will need a genome le found at
/home/i.zhegalova/chipseq_dicty/genome/dicty.fa and corresponding .fai,
download them
• upload dicty genome to IGV (Genome -> Local File -> select BOTH dicty.fa
and dicty.fa.fai with Ctrl button)
• upload your .bigwig (experiment and input) and .narrowpeak le to IGV
(Tracks -> Local File)
You can also upload the gene annotation. The file is located at
/home/m.molodova/genome/Dictyostelium_discoideum.dicty_2.7.60.gff3
2 . 13

14. Task 4. bigwig and peaks at IGV

• Create the comparable scales:
• click on gear to the right of your track and
unmark autoscale for both bigwigs
• click set data range and ll in with the same
numbers for both bigwigs
(If using the IGV Desktop Application, right-click on the track)
2 . 14

15. Task 4. bigwig and peaks at IGV

? provide your observations on peak calling quality and a supportive IGV
screenshot (1 point)
2 . 15

16. Irreproducible Discovery Rate

• Compares a pair of ranked lists
of peaks.
• If two replicas measure the same
cell state, the most signi cant
peaks are expected to have high
inter-replica consistency.
• Peaks of low signi cance will
have poor consistency (noise!).
2 . 16

17. Task 5. IDR

use replicate peaks from your classmate to run IDR on both of your les to
generate reasonable track
idr --samples EXPERIMENT_R1.narrowPeak EXPERIMENT_R2.narrowPeak
--idr-threshold 0.05 --output-file-type narrowPeak \
--output-file IDR.narrowPeak
\
upload both replicates' peaks as well as resulting peak le (3 les in
total). Provide screenshot and your observations - 1 point
2 . 17

18. Task 6. Bigwig around genes

15 . 17

19. Task 6. Bigwig around genes

• First, generate a normalized BigWig file by dividing the ChIP-seq coverage by
the input control signal.
bamCompare -b1 EXPERIMENT.bam -b2 CONTROL.bam \
--outFileName NORMALIZED.bw --outFileFormat=bigwig \
--operation ratio --normalizeUsing BPM --scaleFactorsMethod None
• Compute the normalized signal around genes. Gene coordinates are located
at /home/i.zhegalova/chipseq_dicty/genome/genes.bed
Run this command in screen/ tmux
computeMatrix scale-regions \
-S NORMALIZED.bw \
-R genes.bed \
--beforeRegionStartLength 3000 \
--afterRegionStartLength 3000 \
-o MATRIX.mat.gz \
--smartLabels
2 . 19

20. Task 6. Bigwig around genes

plotProfile -m MATRIX.mat.gz \
-out PLOT.png \
--plotTitle "my fabilous plot" \
--averageType median \
--plotType se
Provide plot and your observations - 2 points
2 . 20