Chapter 4 uliCUT&RUN data analysis

For analyzing the uliCUT&RUN datasets, just begin with the bdg(bedGraph) files.

Steps before getting the bed file (CUT&RUN protocol on GitHub):

  • trim reads to 25 bases (while keeping paired)

  • split reads using novocraft (also trims off barcode with -l option)

  • Align reads using Bowtie2 (align to yeast or to mouse)

  • Remove duplicate reads (created through PCR) using Picard.

  • Remove low quality reads (MAPQ <10)

  • Make Size Distribution file for mm10; Make size distribution for sacCer3, spike in

  • Make size classes (1-120 for TF, 150-500 for histones)

  • Homer analysis

  • makeTagDirectory

  • makeUCSCfile

  • make aggregation plots or heatmaps over specific anchor location

  • call peaks

  • find motifs

4.1 Experimental design

Biological question:

In this project we are only interested in samples from blastocysts, and the goal is to find out which NANOG binding regions are influenced by the depletion of BRG1.

Table 4.1: Experimental design in CUT & RUN
cellType Experiment Antibody Rep
blastocysts EGFPKD NoAb 4
blastocysts EGFPKD NANOG 4
blastocysts Brg1KD(Smarca4KD) NoAb 4
blastocysts Brg1KD(Smarca4KD) NANOG 4
blastocysts NanogkD NoAb 4
blastocysts NanogkD NANOG 4

Hainer, S. J., et al. (2019).

Hints:

  1. Fig F shows GEFP-KD with NANOG Antibody group does not affect the enrichment of NANOG, so actually these samples are used as the control group in the analysis.

  2. From EGFP-KD and Smarca4-KD conditions. I think we might tell which NSNOG binding regions are BRG1 related and which are not by analyzing the two samples to find the differential NANOG peaks.

4.2 bedGraph

BedGraph format

The score is placed in column 4, not column 5

Track lines are compulsory, and must include type=bedGraph. Currently the only optional parameters supported by Ensembl are:

  1. name: unique name to identify this track when parsing the file

  2. description: label to be displayed under the track in Region in Detail

  3. priority: integer defining the order in which to display tracks, if multiple tracks are defined.

  4. graphType: either ‘bar’ or ‘points’

zcat Brg1KD_1.bedGraph.gz | head
# track type=bedGraph name="Brg1KD_Nanog_1_1-120 Total Tags = 1.16e+05, normalized to 1.00e+07" description="Brg1KD_Nanog_1_1-120 Total Tags = 1.16e+05, normalized to 1.00e+07" color=123,110,212 visibility=full yLineOnOff=on autoScale=on yLineMark="0.0" alwaysZero=on graphType=bar maxHeightPixels=128:75:11 windowingFunction=maximum smoothingWindow=off
# chr1    3011588 3011599 41.40
# chr1    3011599 3011692 82.80
# chr1    3011692 3011703 41.40
# chr1    3048182 3048232 41.40
# chr1    3048232 3048286 82.80

bedgraph, bed2bam

Genome browser tracks were generated from mapped reads using the “makeUCSCfile” command.

Mapped reads were aligned over specific regions using the “annotatePeaks” command to make 20 bp bins over regions of interest and sum the reads within each bin. Peaks were called using parameters similar to those previously described (Skene et al., 2018) but implemented in HOMER using the “findPeaks” command with the following parameters: -style factor < or > histone -P 1 -poisson 0.01 -F 0.5 -L 2 -LP 0.001 -i noab.

Motifs were identified using the “findMotifs” command.