Calculate Resolution and Depth for Pairs Data

Functions to analyze genomic interaction resolution and coverage depth.

Load a pairs file into C memory for fast repeated analysis. The cache persists until garbage collected or explicitly cleared.

Usage

calculateResolutionDepth(pairs, bin_size)

calculateGenomeCoverage(pairs, bin_size, min_contacts = 1000)

findOptimalResolution(
  pairs,
  min_bin = 1000,
  max_bin = 5e+06,
  target_coverage = 0.8,
  min_contacts = 1000
)

readPairsCache(pairs_file)

calcResDepthChunked(pairs_file = NULL, bin_size, cache = NULL)

calcGenomeCovChunked(
  pairs_file = NULL,
  bin_size,
  min_contacts = 1000,
  cache = NULL
)

findOptResChunked(
  pairs_file = NULL,
  min_bin = 1000,
  max_bin = 5e+06,
  target_coverage = 0.8,
  min_contacts = 1000,
  cache = NULL,
  return_cache = FALSE
)

Arguments

pairs: Data frame with columns: read_name, chrom1, pos1, chrom2, pos2. For large files, use pairs_file instead.
bin_size: Integer. Bin size in base pairs.
min_contacts: Integer. Minimum contacts per bin (default: 1000).
min_bin: Integer. Minimum bin size for search (default: 1000).
max_bin: Integer. Maximum bin size for search (default: 5000000).
target_coverage: Numeric (0-1). Target coverage fraction (default: 0.8).
pairs_file: Character path to pairs file.
cache: External pointer from readPairsCache(). Faster for repeated operations.
return_cache: Logical. Return list with bin size and cache pointer (default: FALSE).

Value

calculateResolutionDepth(): Tibble with chrom, bin, count.
calculateGenomeCoverage(): Coverage fraction (bins ≥ min_contacts).
findOptimalResolution(): Optimal bin size (integer).
calcResDepthChunked(): Same, reads file in chunks.
calcGenomeCovChunked(): Coverage from file.
findOptResChunked(): Optimal resolution for large files.

An external pointer to the cached data.

Details

Analyzes Hi-C pairs to determine optimal resolution. For large files, use _chunked variants or cache with readPairsCache() for speed.

Examples

if (FALSE) { # \dontrun{
pairs <- read.table("contact_matrix.txt")
depth_10kb <- calculateResolutionDepth(pairs, bin_size = 10000)
coverage <- calculateGenomeCoverage(pairs, bin_size = 10000)
opt_bin <- findOptimalResolution(pairs, target_coverage = 0.8)

# Large file with cache
cache <- readPairsCache("contact_matrix.txt")
opt_bin <- findOptResChunked(cache = cache, target_coverage = 0.8)
depth <- calcResDepthChunked(cache = cache, bin_size = opt_bin)
} # }

if (FALSE) { # \dontrun{
# Create cache once
cache <- readPairsCache("data.pairs.gz")

# Reuse cache for multiple operations
depth1 <- calcResDepthChunked(cache = cache, bin_size = 10000)
depth2 <- calcResDepthChunked(cache = cache, bin_size = 50000)
coverage <- calcGenomeCovChunked(cache = cache, bin_size = 10000)

# Cache is automatically freed when R session ends
# Or explicitly remove it:
rm(cache)
gc()
} # }