GSE120104 Processing Pipeline

Publication

Cell (2019) — PMID 31251911

Dataset

Pervasive Chromatin-RNA Binding Protein Interactions Enable RNA-based Regulation of Transcription [ChIP-Seq]

1. 1

   ChIP-seq data were processed in accordance with ENCODE uniform transcription factor ChIP-seq pipeline (https://www.encodeproject.org/chip-seq/transcription_factor/).

   ChIP-seq vlatest stable release (Inferred with models/gemini-2.5-flash) GitHub

   $ Bash example

   # Clone the ENCODE ChIP-seq pipeline repository (if not already cloned)
   # git clone https://github.com/ENCODE-DCC/chip-seq-pipeline2.git
   # cd chip-seq-pipeline2
   
   # Create an input JSON file for a paired-end Transcription Factor ChIP-seq experiment (hg38).
   # Replace '/path/to/...' with your actual FASTQ file paths.
   # Reference genome files are sourced from ENCODE Project (GRCh38_gencode_v34).
   cat << EOF > input.json
   {
     "chip.fastqs_rep1_R1": ["/path/to/rep1_R1.fastq.gz"],
     "chip.fastqs_rep1_R2": ["/path/to/rep1_R2.fastq.gz"],
     "chip.fastqs_ctl_R1": ["/path/to/control_R1.fastq.gz"],
     "chip.fastqs_ctl_R2": ["/path/to/control_R2.fastq.gz"],
     "chip.paired_end": true,
     "chip.pipeline_type": "tf",
     "chip.genome_tsv": "https://www.encodeproject.org/files/GRCh38_gencode_v34/@@download/GRCh38_gencode_v34.tsv",
     "chip.blacklist_bed": "https://www.encodeproject.org/files/ENCFF356LFX/@@download/ENCFF356LFX.bed.gz",
     "chip.chrsz": "https://www.encodeproject.org/files/GRCh38_gencode_v34/@@download/GRCh38_gencode_v34.chrsz.tsv",
     "chip.ref_fa": "https://www.encodeproject.org/files/GRCh38_gencode_v34/@@download/GRCh38_gencode_v34.fa.gz",
     "chip.bowtie2_idx": "https://www.encodeproject.org/files/GRCh38_gencode_v34/@@download/GRCh38_gencode_v34.bowtie2_idx.tar",
     "chip.bwa_idx": "https://www.encodeproject.org/files/GRCh38_gencode_v34/@@download/GRCh38_gencode_v34.bwa_idx.tar",
     "chip.output_prefix": "my_tf_chipseq_hg38"
   }
   EOF
   
   # Execute the ENCODE ChIP-seq pipeline using Cromwell.
   # This command assumes 'cromwell.jar' is in your PATH or specified with its full path,
   # and that you are running this from the root directory of the cloned 'chip-seq-pipeline2' repository.
   # Replace 'cromwell.jar' with the actual path to your Cromwell JAR file if not in PATH.
   # The 'chip.wdl' file is the main workflow definition.
   java -jar cromwell.jar run chip.wdl -i input.json

   Copy View on GitHub
2. 2

   Briefly, ChIP-seq reads were aligned to the GRCh37 human genome using BWA

   BWA v0.7.17 GitHub

   $ Bash example

   # Install BWA (if not already installed)
   # conda install -c bioconda bwa
   
   # Install Samtools (if not already installed)
   # conda install -c bioconda samtools
   
   # Define variables
   REF_GENOME="/path/to/GRCh37/GRCh37.fa" # Placeholder for GRCh37 human genome reference
   READS_R1="input_reads_R1.fastq.gz" # Placeholder for ChIP-seq forward reads
   READS_R2="input_reads_R2.fastq.gz" # Placeholder for ChIP-seq reverse reads (if paired-end)
   OUTPUT_BAM="aligned_reads.bam"
   OUTPUT_SORTED_BAM="aligned_reads.sorted.bam"
   
   # 1. Index the reference genome (if not already indexed)
   # This step only needs to be run once per reference genome
   # bwa index ${REF_GENOME}
   
   # 2. Align reads to the GRCh37 human genome using BWA-MEM
   # Assuming paired-end reads for ChIP-seq, adjust for single-end if necessary
   bwa mem -M -t 8 ${REF_GENOME} ${READS_R1} ${READS_R2} | samtools view -bS - > ${OUTPUT_BAM}
   
   # 3. Sort the resulting BAM file
   samtools sort ${OUTPUT_BAM} -o ${OUTPUT_SORTED_BAM}
   
   # 4. Index the sorted BAM file
   samtools index ${OUTPUT_SORTED_BAM}

   Copy View on GitHub
3. 3

   PCR duplicates and reads of low quality were filtered

   samtools (Inferred with models/gemini-2.5-flash) v1.9 (Inferred with models/gemini-2.5-flash) GitHub

   $ Bash example

   # Install samtools if not already installed
   # conda install -c bioconda samtools
   
   # Filter out reads marked as PCR duplicates (FLAG 1024) and reads with mapping quality less than 10
   samtools view -F 1024 -q 10 -b input.bam > filtered.bam

   Copy View on GitHub
4. 4

   Peak-calling was performed with SPP and IDR with 0.02 IDR threshhold

   IDR v0.02

   $ Bash example

   # Install IDR (example using pip, check official documentation for best practice)
   # pip install idr
   
   # IDR (Irreproducible Discovery Rate) analysis
   # This step takes peak files (e.g., from SPP for two replicates) and identifies reproducible peaks.
   # The '0.02 IDR threshhold' refers to the --output-threshold parameter, meaning peaks with an IDR score <= 0.02 are considered reproducible.
   idr --samples rep1_spp_peaks.narrowPeak rep2_spp_peaks.narrowPeak \
       --input-file-type narrowPeak \
       --rank-by signal.value \
       --output-file idr_reproducible_peaks.narrowPeak \
       --plot \
       --log-file idr_log.txt \
       --output-threshold 0.02

   Copy
5. 5

   Signal tracks were generated with MACS2 and bigWig applications from UCSC genome browser (http://hgdownload.soe.ucsc.edu/admin/exe/linux.x86_64/)

   MACS2 vMACS2: 2.2.7.1 (Inferred with models/gemini-2.5-flash); UCSC bedGraphToBigWig: Not explicitly versioned, typically latest available at time of download. GitHub

   $ Bash example

   # Placeholder variables
   TREATMENT_BAM="path/to/treatment.bam"
   CONTROL_BAM="path/to/control.bam" # Optional, but common for MACS2
   OUTPUT_PREFIX="sample_name"
   GENOME_ASSEMBLY="hg38" # Latest assembly as placeholder
   OUTPUT_DIR="macs2_output"
   CHROM_SIZES_FILE="${GENOME_ASSEMBLY}.chrom.sizes"
   
   # Create output directory if it doesn't exist
   mkdir -p "${OUTPUT_DIR}"
   
   # --- MACS2 Installation (commented out) ---
   # conda install -c bioconda macs2
   
   # --- UCSC Tools Installation (commented out) ---
   # Download bedGraphToBigWig from UCSC Genome Browser utilities
   # wget http://hgdownload.soe.ucsc.edu/admin/exe/linux.x86_64/bedGraphToBigWig
   # chmod +x bedGraphToBigWig
   # mv bedGraphToBigWig /usr/local/bin/ # Or add to PATH
   
   # --- Generate/Download chrom.sizes file ---
   # This file is required by bedGraphToBigWig. Download from UCSC Genome Browser for the specified assembly.
   wget -O "${CHROM_SIZES_FILE}" "http://hgdownload.soe.ucsc.edu/goldenPath/${GENOME_ASSEMBLY}/bigZips/${GENOME_ASSEMBLY}.chrom.sizes"
   
   # --- MACS2 Execution ---
   # Generate signal tracks (bedGraph files) using MACS2 callpeak with -B and -SPMR options.
   # -t: treatment file (BAM format)
   # -c: control file (BAM format, optional but recommended for ChIP-seq/eCLIP)
   # -f BAM: input file format (BAM for single-end or paired-end if not using BAMPE)
   # -g hs: genome size (e.g., hs for human, mm for mouse). Use a more precise value if known.
   # -n: name string for output files
   # -B: store fragment pileup and control lambda values in bedGraph files
   # -SPMR: calculate signal per million reads (SPMR)
   # --outdir: output directory
   macs2 callpeak -t "${TREATMENT_BAM}" \
                  -c "${CONTROL_BAM}" \
                  -f BAM \
                  -g hs \
                  -n "${OUTPUT_PREFIX}" \
                  -B \
                  -SPMR \
                  --outdir "${OUTPUT_DIR}"
   
   # --- Convert MACS2 bedGraph outputs to bigWig using UCSC bedGraphToBigWig ---
   # MACS2 generates several bedGraph files with -B.
   # Common signal tracks are:
   # 1. ${OUTPUT_PREFIX}_treat_pileup.bdg (raw signal pileup)
   # 2. ${OUTPUT_PREFIX}_control_lambda.bdg (background lambda values)
   # Often, fold enrichment (FE) or -log10(pvalue) signal is desired, which requires an additional 'macs2 bdgcmp' step.
   
   # Convert treatment pileup bedGraph to bigWig
   bedGraphToBigWig "${OUTPUT_DIR}/${OUTPUT_PREFIX}_treat_pileup.bdg" \
                    "${CHROM_SIZES_FILE}" \
                    "${OUTPUT_DIR}/${OUTPUT_PREFIX}_treat_pileup.bw"
   
   # Example: Generate and convert fold enrichment signal if desired
   # macs2 bdgcmp -t "${OUTPUT_DIR}/${OUTPUT_PREFIX}_treat_pileup.bdg" \
   #              -c "${OUTPUT_DIR}/${OUTPUT_PREFIX}_control_lambda.bdg" \
   #              -m FE \
   #              -o "${OUTPUT_DIR}/${OUTPUT_PREFIX}_FE.bdg"
   # bedGraphToBigWig "${OUTPUT_DIR}/${OUTPUT_PREFIX}_FE.bdg" \
   #                  "${CHROM_SIZES_FILE}" \
   #                  "${OUTPUT_DIR}/${OUTPUT_PREFIX}_FE.bw"

   Copy View on GitHub

Tools Used