1. Public SRA & GEO Ingestion Architecture

Public repositories such as the NCBI Gene Expression Omnibus (GEO) and the Sequence Read Archive (SRA) store thousands of high-quality ChIP-seq runs. For our Huntington's Disease study, we retrieve matched biological replicates: HD patients vs. healthy controls, alongside their corresponding Input (sheared chromatin without antibody) libraries.

# Create structured project directories
mkdir -p hd_chipseq/{00_raw_fastq,01_trimmed,02_aligned,03_filtered,04_peaks,05_diffbind,06_qc_reports,07_visualization}
cd hd_chipseq

# Download SRA datasets using multi-threaded fasterq-dump
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041280 # HD H3K4me3 Rep 1
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041281 # HD H3K4me3 Rep 2
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041282 # Control H3K4me3 Rep 1
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041283 # Control H3K4me3 Rep 2
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041284 # HD Input
fasterq-dump --split-files --threads 8 --outdir 00_raw_fastq/ SRR2041285 # Control Input

# Compress raw FASTQ files to conserve disk space
pigz -p 8 00_raw_fastq/*.fastq

2. Raw Read Quality Assessment with FastQC

Prior to read trimming or alignment, we run FastQC to inspect base call Phred scores ($Q$), per-base sequence content, GC distribution, and over-represented adapter sequences:

fastqc -t 8 00_raw_fastq/*.fastq.gz -o 06_qc_reports/

3. Automated Trimming & Quality Filtering with fastp

ChIP-seq libraries frequently suffer from Illumina adapter read-through when DNA fragments are shorter than the sequencing cycle count. Additionally, NextSeq and NovaSeq two-color chemistries produce spurious poly-G tails when no fluorescent signal is detected. We use fastp to perform adapter detection, poly-G clipping, and quality filtering:

for sample in HD_Rep1 HD_Rep2 Control_Rep1 Control_Rep2 HD_Input Control_Input; do
    echo "Processing sample: ${sample}"
    fastp \
        -i 00_raw_fastq/${sample}.fastq.gz \
        -o 01_trimmed/${sample}.trimmed.fastq.gz \
        --html 06_qc_reports/${sample}_fastp.html \
        --json 06_qc_reports/${sample}_fastp.json \
        --trim_poly_g \
        --cut_front --cut_tail --cut_mean_quality 20 \
        --length_required 35 \
        --thread 6
done

# Aggregate all QC statistics into a single interactive HTML report
multiqc 06_qc_reports/ -o 06_qc_reports/multiqc_raw_report/