7: Bioconductor Core & GenomicRanges

Chapter 7: Bioconductor Core: S4 Classes & GenomicRanges

The GenomicRanges package represents the foundational data structure of modern computational genomics. By encapsulating chromosomes, interval coordinates, and strandedness inside validated S4 classes, GRanges provides fast interval arithmetic and overlap queries across millions of features.

1. Constructing GRanges Objects

library(GenomicRanges)

# Define gene models
genes_gr <- GRanges(
    seqnames = c("chr1", "chr1", "chr2", "chr17"),
    ranges   = IRanges(start = c(10000, 50000, 25000, 7668402),
                       end   = c(15000, 58000, 32000, 7687550)),
    strand   = c("+", "-", "+", "-"),
    gene_id  = c("GENE_A", "GENE_B", "GENE_C", "TP53")
)
print(genes_gr)

2. Genomic Arithmetic: Promoters & Flanks

Extracting transcription start site (TSS) promoter regions (e.g. 2000bp upstream, 200bp downstream):

promoters_gr <- promoters(genes_gr, upstream = 2000, downstream = 200)
print(promoters_gr)

3. Finding Interval Overlaps (ChIP-Seq Peaks vs Promoters)

Identifying which transcription factor peaks overlap promoter regions:

overlaps <- findOverlaps(tf_peaks_gr, promoters_gr)

# Extract matching peak-gene pairs
query_idx <- queryHits(overlaps)
subj_idx  <- subjectHits(overlaps)
cat("Overlapping Hits:\n")
print(data.frame(Peak = tf_peaks_gr$peak_id[query_idx], Gene = promoters_gr$gene_id[subj_idx]))