12  Lab 3: Processing MNase-seq data

Aims
  • Trim adaptors from paired-end fastq files
  • Map MNase-seq reads with bowtie2
  • Filter mapped reads with samtools
  • Generate a nucleosome position track
Datasets

We will process a data from the Koszul lab, generated in 2024 and published in Science.

12.1 Download MNase-seq reads from SRA

We can download the paired-end reads (R1 and R2 fastq files) directly from the internet.

  • Find the download links associated with the SRR31398330 SRR ID. You can go to SRA-explorer to easily recover links.
  • Download the two fastq files for the SRR31398330 SRR ID.
sh
curl -L ftp://ftp.sra.ebi.ac.uk/vol1/fastq/SRR313/030/SRR31398330/SRR31398330_1.fastq.gz -o MNase_20_R1.fq.gz
curl -L ftp://ftp.sra.ebi.ac.uk/vol1/fastq/SRR313/030/SRR31398330/SRR31398330_2.fastq.gz -o MNase_20_R2.fq.gz

12.2 Trim adaptor sequences

  • Why is it important to trim adaptors from fastq files?
  • Why should it be done in a paired-end fashion?
  • Proceed to fastq trimming with trim_galore.
sh
trim_galore \
    --cores 2 \
    --length 20 \
    --gzip \
    --paired \
    --output_dir ./ \
    Share/reads/MNase_20_R1.fq.gz Share/reads/MNase_20_R2.fq.gz
  • Was it necessary? How many reads were trimmed? How many bases were trimmed?
sh
head -n 40 MNase_20_R1.fq.gz_trimming_report.txt
head -n 40 MNase_20_R1.fq.gz_trimming_report.txt

12.3 Align reads to a genome reference

  • Can and should STAR be used to map MNase-seq reads to a genome reference?
  • Build S. cerevisiae genome (R64-1-1) index for bowtie2.
sh
bowtie2-build Share/genome/R64-1-1.fa genomes/R64-1-1
  • Map paired-end reads with bowtie2.
  • What does the output sam file contain, in terms of reads? In term of un/aligned reads?
sh
bowtie2 \
    --threads 2 \
    -x genomes/R64-1-1 \
    -1 MNase_20_R1_val_1.fq.gz \
    -2 MNase_20_R2_val_2.fq.gz \
    > MNase_20.sam

12.4 Filter MNase-seq reads

  • Filter mapped pairs using the following procedure:

    • Fix mates

    • Filter pairs:

      • Only keep paired reads (0x001)
      • Only keep reads mapped in proper pair (0x002)
      • Remove unmapped reads (0x004)
      • Remove reads with unmapped mate (0x008)
      • Reads mapped with a MAPQ >= 20

    • Sort and index reads

To better understand the combinations of information described by the SAM “flag”, check the Decoding SAM flags page.

sh
## Fixing mates
#      -r: "Remove unmapped reads and secondary alignments"
#      -m: "Add mate score tag"
samtools fixmate \
    -@ 2 \
    --output-fmt bam \
    -r -m \
    MNase_20.sam MNase_20.bam

## - Filter read pairs
#       -f 0x001: "Keep read paired"
#       -f 0x002: "Keep read mapped in proper pair"
#       -F 0x004: "Remove read unmapped"
#       -F 0x008: "Remove mate unmapped"
#       -q 20: "MAPQ >= 20"
#       --fast: "Use fast bam compression"
samtools view \
    -@ 2 \
    --output-fmt bam \
    -f 0x001 -f 0x002 -F 0x004 -F 0x008 -q 20 \
    --fast \
    MNase_20.bam \
    -o MNase_20_filtered.bam

## - Sorting read pairs
#| -l 9: "Use best compression "
samtools sort \
    -@ 2 \
    --output-fmt bam \
    -l 9 \
    --write-index \
    MNase_20_filtered.bam \
    -o MNase_20_filtered_sorted.bam
  • Create a sequencing depth-normalized track, filling out fragments.
sh
bamCoverage \
    --bam MNase_20_filtered_sorted.bam \
    --outFileName MNase_20_filtered_sorted.CPM.bw \
    --binSize 10 \
    --numberOfProcessors 4 \
    --normalizeUsing CPM \
    --skipNonCoveredRegions \
    --extendReads

12.5 Create a nucleosome position track

  • Create a nucleosome position track by keeping the fragments between 130 and 165bp, and extending them to 40bp aligned at their center.
sh
bamCoverage \
    --bam MNase_20_filtered_sorted.bam \
    --outFileName MNase_20_filtered_sorted.135-160bp.nuc-center.CPM.bw \
    --binSize 1 \
    --smoothLength 10 \
    --numberOfProcessors 4 \
    --normalizeUsing CPM \
    --skipNonCoveredRegions \
    --MNase

12.6 Visualize nucleosome positioning

  • Load the nucleosome tracks together with the RNA-seq tracks from the previous lab in IGV.
  • Visualize the nucleosome positioning around the HHO1 gene.
  • How do you interpret the observed nucleosome positioning?
Back to top