GSE153279 Processing Pipeline
OTHER
code_examples
13 steps
Publication
Identification of the global miR-130a targetome reveals a role for TBL1XR1 in hematopoietic stem cell self-renewal and t(8;21) AML.Cell reports (2022) — PMID 35263585
Dataset
GSE153279Definition of a Small Core Transcriptional Circuit Regulated by AML1-ETO [ChIP-seq, Cut&Run]
Warning: Pipeline descriptions and code snippets may be inferred or AI-generated. Use them only as a starting point to guide analysis, and validate before use.
Processing Steps
Generate Jupyter Notebook-
1
Library strategy: Cut&Run
$ Bash example
# Install Cromwell (Java-based workflow engine) if not already installed. # Download the latest stable release from https://github.com/broadinstitute/cromwell/releases # For example: # wget https://github.com/broadinstitute/cromwell/releases/download/86/cromwell-86.jar # 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 a placeholder input JSON file for Cut&Run analysis. # Replace 'your_sample_R1.fastq.gz', 'your_sample_R2.fastq.gz' with actual paths to your FASTQ files. # Replace 'your_output_directory' with your desired output location. # The reference genome (hg38) and associated files are placeholders using ENCODE's GCS paths. # For actual analysis, ensure these files are accessible or provide local paths. cat << EOF > inputs.json { "chip.fastq_replicates": [ [ "your_sample_R1.fastq.gz", "your_sample_R2.fastq.gz" ] ], "chip.genome_tsv": "gs://encode-pipeline-genome-data/vERCC/GRCh38_vERCC/GRCh38_vERCC.tsv", "chip.chrom_sizes": "gs://encode-pipeline-genome-data/vERCC/GRCh38_vERCC/GRCh38_vERCC.chrom.sizes", "chip.blacklist": "gs://encode-pipeline-genome-data/vERCC/GRCh38_vERCC/GRCh38_vERCC.blacklist.bed", "chip.paired_end": true, "chip.pipeline_type": "cutnrun", "chip.aligner": "bwa", "chip.peak_caller": "macs2", "chip.output_directory": "your_output_directory" } EOF # Execute the ENCODE ChIP-seq pipeline for Cut&Run using Cromwell. # Ensure you have the 'chip_seq.wdl' file from the cloned repository in the current directory # and 'cromwell-X.Y.Z.jar' (e.g., cromwell-86.jar) is accessible. java -jar cromwell-86.jar run chip_seq.wdl -i inputs.json -
2
For Cut&Run:
$ Bash example
# This pipeline is designed for ChIP-seq but is adaptable for Cut&Run data analysis. # Install Cromwell (Java-based workflow management system) if not already installed. # Ensure you have Java 8 or higher installed. # For example, download the latest stable release from Cromwell's GitHub releases page: # wget https://github.com/broadinstitute/cromwell/releases/download/X.Y.Z/cromwell-X.Y.Z.jar # mv cromwell-X.Y.Z.jar /usr/local/bin/cromwell.jar # Clone the ENCODE ChIP-seq pipeline repository git clone https://github.com/ENCODE-DCC/chip-seq-pipeline2.git cd chip-seq-pipeline2 # (Optional) Checkout a specific version of the pipeline, e.g., v2.0.0 # git checkout v2.0.0 # --- Prepare input.json for Cut&Run data --- # The ENCODE ChIP-seq pipeline uses WDL (Workflow Description Language) and requires an input JSON file. # Replace placeholders with actual paths to your data and reference files. # For human (hg38) reference, you would need a genome_tsv file containing paths to: # - Reference FASTA # - Chromosome sizes file # - Blacklist BED file # - Bowtie2 index # - BWA index # - MACS2 genome size parameter (e.g., "hs" for human) # Example hg38.tsv can be found in the pipeline's 'genome_data' directory or generated using provided scripts. cat << EOF > cutnrun_input.json { "chip.fastqs_rep1_R1": ["/path/to/your/cutnrun_experiment_rep1_R1.fastq.gz"], "chip.fastqs_rep1_R2": ["/path/to/your/cutnrun_experiment_rep1_R2.fastq.gz"], "chip.fastqs_ctl_rep1_R1": ["/path/to/your/cutnrun_control_rep1_R1.fastq.gz"], "chip.fastqs_ctl_rep1_R2": ["/path/to/your/cutnrun_control_rep1_R2.fastq.gz"], "chip.paired_end": true, "chip.genome_tsv": "/path/to/your/genome_data/hg38.tsv", "chip.output_prefix": "cutnrun_hg38_experiment", "chip.aligner": "bowtie2", "chip.peak_caller": "macs2", "chip.macs2_gsize": "hs", "chip.blacklist_bed": "/path/to/your/blacklist/hg38-blacklist.bed", "chip.enable_idr": false, # Set to true if you have multiple replicates for IDR analysis "chip.fraglen_range": "0 1000" # Typical fragment size range for Cut&Run } EOF # --- Execute the pipeline using Cromwell --- # Replace '/path/to/cromwell.jar' with the actual path to your Cromwell JAR file. # You might need to configure Cromwell for a specific backend (e.g., local, SGE, SLURM) # via a Cromwell configuration file (e.g., cromwell.conf). java -jar /path/to/cromwell.jar run chip.wdl -i cutnrun_input.json -
3
Adaptor trimming with Trimmomatic-0.32.
$ Bash example
bash # Install Trimmomatic if not already installed # conda install -c bioconda trimmomatic=0.32 # Define input and output file names INPUT_R1="input_R1.fastq.gz" INPUT_R2="input_R2.fastq.gz" OUTPUT_R1_PAIRED="output_R1_paired.fastq.gz" OUTPUT_R1_UNPAIRED="output_R1_unpaired.fastq.gz" OUTPUT_R2_PAIRED="output_R2_paired.fastq.gz" OUTPUT_R2_UNPAIRED="output_R2_unpaired.fastq.gz" # Define adaptor file path (adjust path as necessary for your Trimmomatic installation) # A common adaptor file for Illumina paired-end data is TruSeq3-PE.fa, often found in the 'adapters' directory of Trimmomatic. # For Trimmomatic 0.32, this file would typically be located within the Trimmomatic installation directory. ADAPTOR_FILE="/path/to/Trimmomatic-0.32/adapters/TruSeq3-PE.fa" # Run Trimmomatic for paired-end adaptor trimming # Adjust '/path/to/Trimmomatic-0.32/trimmomatic-0.32.jar' to the actual location of your Trimmomatic JAR file. java -jar /path/to/Trimmomatic-0.32/trimmomatic-0.32.jar PE \ -phred33 \ "${INPUT_R1}" "${INPUT_R2}" \ "${OUTPUT_R1_PAIRED}" "${OUTPUT_R1_UNPAIRED}" \ "${OUTPUT_R2_PAIRED}" "${OUTPUT_R2_UNPAIRED}" \ ILLUMINACLIP:"${ADAPTOR_FILE}":2:30:10 \ LEADING:3 \ TRAILING:3 \ SLIDINGWINDOW:4:15 \ MINLEN:36 -
4
Alignment to hg19+scer genome with bowtie2.
$ Bash example
# Install Bowtie2 (if not already installed) # conda install -c bioconda bowtie2=2.3.4.3 # Define variables INPUT_FASTQ="input.fastq.gz" # Placeholder for your input FASTQ file OUTPUT_SAM="output.sam" OUTPUT_BAM="output.bam" OUTPUT_SORTED_BAM="output.sorted.bam" # Reference genome index: hg19+scer (Saccharomyces cerevisiae) combined index # This is a custom index that needs to be pre-built using bowtie2-build from a combined FASTA file # containing both human hg19 and S. cerevisiae reference sequences. # Example path for the index prefix (replace with your actual path): GENOME_INDEX_PREFIX="/path/to/genome_indices/hg19_scer" # Perform alignment using Bowtie2 # --local: local alignment mode (soft-clipping allowed, common for eCLIP) # -p 8: use 8 threads for parallel processing # -x: specify the index prefix for the reference genome # -U: specify single-end FASTQ input file # -S: specify SAM output file bowtie2 --local -p 8 -x "${GENOME_INDEX_PREFIX}" -U "${INPUT_FASTQ}" -S "${OUTPUT_SAM}" # Convert SAM to BAM, sort, and index (common post-alignment steps in eCLIP pipelines) # Install Samtools (if not already installed) # conda install -c bioconda samtools samtools view -bS "${OUTPUT_SAM}" > "${OUTPUT_BAM}" samtools sort "${OUTPUT_BAM}" -o "${OUTPUT_SORTED_BAM}" samtools index "${OUTPUT_SORTED_BAM}" -
5
High quality reads (F4q10) were filtered using samtools.
$ Bash example
# Install samtools if not already installed # conda install -c bioconda samtools=1.10 # Filter high quality reads (F4q10) # -b: output BAM format # -F 4: exclude reads where the FLAG bit 4 is set (unmapped reads) # -q 10: include reads with mapping quality >= 10 samtools view -b -F 4 -q 10 -o filtered.bam input.bam
-
6
Peaks called with MACS2 (narrowPeak, q=0.001).
$ Bash example
# Install MACS2 (if not already installed) # conda install -c bioconda macs2=2.2.7.1 # Define input files and parameters TREATMENT_BAM="treatment.bam" # Replace with your actual treatment BAM file CONTROL_BAM="control.bam" # Replace with your actual control BAM file (optional but recommended for ChIP-seq) GENOME_SIZE="hs" # Genome size (e.g., hs for human, mm for mouse, dm for drosophila, ce for C. elegans). Adjust as needed. Q_VALUE="0.001" OUTPUT_PREFIX="my_experiment_peaks" OUTPUT_DIR="macs2_output" INPUT_FORMAT="BAMPE" # Input format: BAM, BED, BAMPE (for paired-end BAM). Adjust based on your data. mkdir -p "${OUTPUT_DIR}" macs2 callpeak \ -t "${TREATMENT_BAM}" \ -c "${CONTROL_BAM}" \ -f "${INPUT_FORMAT}" \ -g "${GENOME_SIZE}" \ -q "${Q_VALUE}" \ --outdir "${OUTPUT_DIR}" \ -n "${OUTPUT_PREFIX}" \ --keep-dup all \ --call-summits # Ensures narrowPeak output and identifies peak summits -
7
For HA Cut&Run +/-dTAG, differential peaks were identified by Diffbind/DESeq2 using yeast spike-in for normalization.
$ Bash example
# Install R and DESeq2 (if not already installed) # sudo apt-get update # sudo apt-get install r-base # R -e "install.packages('BiocManager', repos='https://cloud.r-project.org')" # R -e "BiocManager::install('DESeq2')" # --- Placeholder for input files --- # These files would typically be generated by upstream steps (e.g., DiffBind for peak counts). # primary_counts.tsv: Tab-separated file with peak IDs as rows and sample names as columns, containing raw counts for the primary organism (e.g., human hg38). # yeast_counts.tsv: Tab-separated file with yeast feature IDs as rows and sample names as columns, containing raw counts for the yeast spike-in (e.g., sacCer3). # sample_metadata.tsv: Tab-separated file with sample names, condition, and any other relevant metadata. # Example content for sample_metadata.tsv: # sample_id\tcondition # sample1\tHA_CutRun_dTAG_pos # sample2\tHA_CutRun_dTAG_pos # sample3\tHA_CutRun_dTAG_neg # sample4\tHA_CutRun_dTAG_neg # Create an R script for DESeq2 analysis with spike-in normalization cat << 'EOF' > run_deseq2_spikein.R library(DESeq2) # Load primary organism counts (e.g., from DiffBind output) primary_counts <- read.csv("primary_counts.tsv", sep="\t", row.names=1) primary_counts <- as.matrix(primary_counts) # Load yeast spike-in counts yeast_counts <- read.csv("yeast_counts.tsv", sep="\t", row.names=1) yeast_counts <- as.matrix(yeast_counts) # Load sample metadata sample_metadata <- read.csv("sample_metadata.tsv", sep="\t", row.names=1) # Ensure sample order matches between counts and metadata sample_metadata <- sample_metadata[colnames(primary_counts), , drop=FALSE] yeast_counts <- yeast_counts[, colnames(primary_counts)] # Ensure yeast counts match primary samples # Calculate spike-in normalization factors # Sum counts for each sample from the yeast spike-in spikein_sums <- colSums(yeast_counts) # Calculate size factors based on spike-in sums # A common approach is to use the median ratio method on spike-in counts # Or simply scale by total spike-in counts spikein_sf <- spikein_sums / exp(mean(log(spikein_sums))) # Create DESeqDataSet object dds <- DESeqDataSetFromMatrix(countData = primary_counts, colData = sample_metadata, design = ~ condition) # Assign custom size factors based on spike-in sizeFactors(dds) <- spikein_sf # Run DESeq2 analysis dds <- DESeq(dds) # Get results for the comparison of interest (e.g., HA_CutRun_dTAG_pos vs HA_CutRun_dTAG_neg) res <- results(dds, contrast=c("condition", "HA_CutRun_dTAG_pos", "HA_CutRun_dTAG_neg")) # Order results by adjusted p-value res <- res[order(res$padj),] # Save results write.csv(as.data.frame(res), "differential_peaks_deseq2.csv") # Optional: Save normalized counts norm_counts <- counts(dds, normalized=TRUE) write.csv(as.data.frame(norm_counts), "normalized_peak_counts_deseq2.csv") # Session info for reproducibility sessionInfo() EOF # Execute the R script Rscript run_deseq2_spikein.R -
8
bigWig files were generated using deepTools and bedGraph files were generated using Homer.
$ Bash example
# Install HOMER (if not already installed) # git clone https://github.com/homer-tools/homer.git # cd homer # perl configureHomer.pl -install # Define input and output paths INPUT_BAM="input.bam" # Replace with your actual input BAM file OUTPUT_TAG_DIR="homer_tags" OUTPUT_BEDGRAPH="output.bedGraph" FRAGMENT_SIZE="150" # Adjust fragment size as appropriate for your assay (e.g., 150 for ChIP-seq, or estimated from cross-correlation analysis) # 1. Create a HOMER tag directory from the BAM file # This step processes the BAM file to create a directory of tag files, which HOMER uses for downstream analysis. makeTagDirectory "${OUTPUT_TAG_DIR}" "${INPUT_BAM}" # 2. Generate bedGraph file from the tag directory # -o: Specify the output bedGraph file name # -fsize: Fragment size (e.g., 150bp for ChIP-seq). This parameter extends reads to the estimated fragment length. # -norm 1e7: Normalize read counts to 10 million total reads (a common practice for comparing samples). # -res 1: Set the resolution to 1bp, meaning each base pair will have a coverage value. # -tbp 1: Output tags per base pair. # -bedGraph: Specify that the output format should be bedGraph. makeUCSCfile "${OUTPUT_TAG_DIR}" -o "${OUTPUT_BEDGRAPH}" -fsize "${FRAGMENT_SIZE}" -norm 1e7 -res 1 -tbp 1 -bedGraph -
9
For ChIP-seq:
$ Bash example
# This script demonstrates how to run the ENCODE ChIP-seq pipeline locally using Cromwell. # Ensure Java (JRE 8 or higher) and Cromwell are installed and accessible. # Download Cromwell JAR from https://github.com/broadinstitute/cromwell/releases # Example: wget https://github.com/broadinstitute/cromwell/releases/download/85/cromwell-85.jar # 1. Clone the ENCODE ChIP-seq pipeline repository # git clone https://github.com/ENCODE-DCC/chip-seq-pipeline2.git # cd chip-seq-pipeline2 # PIPELINE_DIR=$(pwd) # 2. Prepare input files and reference genome data # Replace these with your actual data paths and chosen reference genome (e.g., hg38, mm10). # For reference data, you can download pre-built ENCODE reference bundles or create your own. # As a placeholder, we use hg38. You would need to download the actual files. # Example for hg38 reference files (these are placeholders, actual download needed): # REF_GENOME_DIR="/path/to/your/reference_genome/hg38" # mkdir -p $REF_GENOME_DIR # Download hg38.tsv, hg38.blacklist.bed.gz, hg38.chrom.sizes, hg38.fa, bowtie2_indexes/hg38.*, bwa_indexes/hg38.* # from ENCODE DCC or similar sources into $REF_GENOME_DIR. # Example input FASTQ files (replace with your actual data) # FASTQ_REP1_R1="/path/to/your/data/sample_rep1_R1.fastq.gz" # FASTQ_REP1_R2="/path/to/your/data/sample_rep1_R2.fastq.gz" # Use an empty array [] if single-end # FASTQ_CTL_REP1_R1="/path/to/your/data/control_rep1_R1.fastq.gz" # FASTQ_CTL_REP1_R2="/path/to/your/data/control_rep1_R2.fastq.gz" # Use an empty array [] if single-end # 3. Create an input JSON file for the WDL workflow # This is a minimal example. Adjust parameters based on your experiment (e.g., paired_end, pipeline_type, peak_caller). # Refer to the pipeline's documentation for all available parameters: # https://github.com/ENCODE-DCC/chip-seq-pipeline2/blob/master/chip.wdl # https://github.com/ENCODE-DCC/chip-seq-pipeline2/blob/master/docs/inputs.md cat << EOF > chip_inputs.json { "chip.genome_tsv": "${REF_GENOME_DIR}/hg38.tsv", "chip.fastqs_rep1_R1": ["${FASTQ_REP1_R1}"], "chip.fastqs_rep1_R2": ["${FASTQ_REP1_R2}"], "chip.fastqs_ctl_rep1_R1": ["${FASTQ_CTL_REP1_R1}"], "chip.fastqs_ctl_rep1_R2": ["${FASTQ_CTL_REP1_R2}"], "chip.paired_end": true, "chip.pipeline_type": "histone", # or "tf" for transcription factor "chip.peak_caller": "macs2", # or "spp" "chip.blacklist_bed": "${REF_GENOME_DIR}/hg38.blacklist.bed.gz", "chip.chrsz": "${REF_GENOME_DIR}/hg38.chrom.sizes", "chip.ref_fasta": "${REF_GENOME_DIR}/hg38.fa", "chip.bowtie2_idx": "${REF_GENOME_DIR}/bowtie2_indexes/hg38", "chip.bwa_idx": "${REF_GENOME_DIR}/bwa_indexes/hg38", "chip.output_prefix": "my_chipseq_experiment" } EOF # 4. Run the ChIP-seq pipeline using Cromwell # Replace /path/to/cromwell-*.jar with the actual path to your Cromwell JAR file. # Replace /path/to/chip-seq-pipeline2/chip.wdl with the actual path to the WDL file. java -jar /path/to/cromwell-*.jar run /path/to/chip-seq-pipeline2/chip.wdl -i chip_inputs.json -
10
Adaptor trimming with Trimmomatic-0.32.
Trimmomatic v0.32$ Bash example
# Install Trimmomatic (if not already installed) # For example, using conda: # conda install -c bioconda trimmomatic # Define input and output file names (placeholders) INPUT_R1="sample_R1.fastq.gz" INPUT_R2="sample_R2.fastq.gz" OUTPUT_R1_PAIRED="sample_R1_trimmed_paired.fastq.gz" OUTPUT_R1_UNPAIRED="sample_R1_trimmed_unpaired.fastq.gz" OUTPUT_R2_PAIRED="sample_R2_trimmed_paired.fastq.gz" OUTPUT_R2_UNPAIRED="sample_R2_trimmed_unpaired.fastq.gz" # Define the path to Trimmomatic JAR and adaptor files # If installed via conda, you might find them like this: # TRIMMOMATIC_JAR=$(find "$(conda env list | grep '*' | awk '{print $NF}')" -name "trimmomatic-*.jar" | head -n 1) # ADAPTOR_FILE="$(dirname "${TRIMMOMATIC_JAR}")/adapters/TruSeq3-PE.fa" # Or, manually specify if not using conda or if paths differ: TRIMMOMATIC_JAR="/path/to/trimmomatic-0.32.jar" # Adjust this path to your Trimmomatic JAR file ADAPTOR_FILE="/path/to/adapters/TruSeq3-PE.fa" # Adjust this path to the appropriate adaptor file (e.g., TruSeq3-PE.fa, NexteraPE-PE.fa, etc.) # Execute Trimmomatic for paired-end adaptor trimming and quality filtering # Parameters: # PE: Paired-end mode # -phred33: Phred score encoding (common for Illumina data) # ILLUMINACLIP: Adaptor trimming (adaptor_file:seed_mismatches:palindrome_clip_threshold:simple_clip_threshold) # - 2: Maximum mismatch count that will be tolerated when aligning the adapter to a read. # - 30: Specifies the 'palindrome clip threshold'. A score of 30 is a good default. # - 10: Specifies the 'simple clip threshold'. A score of 10 is a good default. # LEADING: Remove leading low quality bases (below quality 3) # TRAILING: Remove trailing low quality bases (below quality 3) # SLIDINGWINDOW: Scan with a 4-base window, cut if average quality below 15 # MINLEN: Drop reads shorter than 36 bases java -jar "${TRIMMOMATIC_JAR}" PE \ -phred33 \ "${INPUT_R1}" "${INPUT_R2}" \ "${OUTPUT_R1_PAIRED}" "${OUTPUT_R1_UNPAIRED}" \ "${OUTPUT_R2_PAIRED}" "${OUTPUT_R2_UNPAIRED}" \ ILLUMINACLIP:"${ADAPTOR_FILE}":2:30:10 \ LEADING:3 \ TRAILING:3 \ SLIDINGWINDOW:4:15 \ MINLEN:36 -
11
Alignment to hg19+dm3 genome with bowtie2.
$ Bash example
# Install Bowtie2 and Samtools if not already installed # conda install -c bioconda bowtie2 samtools # Define variables (replace with actual paths and filenames) READ1="input_R1.fastq.gz" READ2="input_R2.fastq.gz" OUTPUT_BAM="aligned_to_hg19_dm3.bam" BOWTIE2_INDEX_PREFIX="/path/to/your/hg19_dm3_index/hg19_dm3" # This prefix points to the Bowtie2 index files (e.g., hg19_dm3.1.bt2, hg19_dm3.2.bt2, etc.) NUM_THREADS=8 # Adjust based on available CPU cores # Perform alignment with Bowtie2 using parameters common for eCLIP (very-sensitive-local) # and pipe the SAM output to Samtools for conversion to BAM format. bowtie2 \ --very-sensitive-local \ -p ${NUM_THREADS} \ -x ${BOWTIE2_INDEX_PREFIX} \ -1 ${READ1} \ -2 ${READ2} \ | samtools view -bS - \ > ${OUTPUT_BAM} # Optional: Sort and index the resulting BAM file # samtools sort -o ${OUTPUT_BAM%.bam}.sorted.bam ${OUTPUT_BAM} # samtools index ${OUTPUT_BAM%.bam}.sorted.bam -
12
High quality reads (F4q10) were filtered using samtools.
$ Bash example
# Install samtools (if not already installed) # conda install -c bioconda samtools # Filter high quality reads (F4q10) using samtools view # -F 4: Exclude reads where the 0x4 flag is set (i.e., exclude unmapped reads) # -q 10: Include only reads with a mapping quality of 10 or higher # -b: Output in BAM format samtools view -F 4 -q 10 -b input.bam > filtered.bam
-
13
Normalized bigWig files were generated with deepTools.
deepTools v3.5.1 (Inferred with models/gemini-2.5-flash)$ Bash example
# Install deepTools (if not already installed) # conda create -n deeptools_env deeptools=3.5.1 # conda activate deeptools_env # Placeholder variables INPUT_BAM="sample.bam" OUTPUT_BIGWIG="sample_normalized.bw" BIN_SIZE=10 # Bin size in base pairs NUMBER_OF_PROCESSORS="auto" # Use all available processors # Generate normalized bigWig file using CPM (Counts Per Million mapped reads) normalization bamCoverage \ --bam "${INPUT_BAM}" \ --outFileName "${OUTPUT_BIGWIG}" \ --normalizeUsing CPM \ --binSize "${BIN_SIZE}" \ --numberOfProcessors "${NUMBER_OF_PROCESSORS}" \ --verbose
Raw Source Text
Library strategy: Cut&Run For Cut&Run: Adaptor trimming with Trimmomatic-0.32. Alignment to hg19+scer genome with bowtie2. High quality reads (F4q10) were filtered using samtools. Peaks called with MACS2 (narrowPeak, q=0.001). For HA Cut&Run +/-dTAG, differential peaks were identified by Diffbind/DESeq2 using yeast spike-in for normalization. bigWig files were generated using deepTools and bedGraph files were generated using Homer. For ChIP-seq: Adaptor trimming with Trimmomatic-0.32. Alignment to hg19+dm3 genome with bowtie2. High quality reads (F4q10) were filtered using samtools. Normalized bigWig files were generated with deepTools. Genome_build: hg19 (GRCh37) Supplementary_files_format_and_content: bigWig, bedGraph files Supplementary_files_format_and_content: 042420_HA_AE_Diffbind-peakinfo_q0.001.csv: Diffbind peak info Supplementary_files_format_and_content: 042420_HA-AE_cutandrun_diffbind-deseq2_q0.001.csv: Differential peak DESeq2 Supplementary_files_format_and_content: New_ZnF_q0.001_peaks.txt: MACS2 peaks Supplementary_files_format_and_content: CD34_HA_merged_noNeg_nomodel_extsize300_q0.001_peaks.txt: MACS2 peaks