API Reference

logoplot(pfm[, background]; kwargs...)

Render a sequence logo from a position frequency matrix.

Arguments

• pfm: position frequency matrix (4 × N for DNA/RNA, 20 × N for protein), each column summing to 1.
• background: optional background frequencies (length 4 for nucleotides, 20 for protein). Defaults to a uniform background.

Keyword arguments

• protein::Bool=false: treat the PFM as a 20-row amino-acid matrix.
• rna::Bool=false: use A, C, G, U glyphs instead of A, C, G, T.
• tight::Bool=false: use tight plot limits (drops padding around the logo).
• xaxis, yaxis: toggle axis display.
• alpha, beta: glyph transparency and width scaling.
• uniform_color::Bool=false: use a single color for all glyphs.
• scale_by_frequency::Bool=false: scale letters by frequency only (stack to full height) instead of by information content.
• dpi, _margin_, pos, xrotation, and additional styling parameters are also accepted.

See also logoplot_with_highlight and save_logoplot.

logoplot_with_highlight(pfm, background, highlighted_regions; kwargs...)

Plot a sequence logo with specific regions highlighted.

Arguments

• pfm: Position frequency matrix
• background: Background frequencies
• highlighted_regions: Vector of UnitRange{Int} specifying positions to highlight
• scale_by_frequency::Bool=false: If true, scale letters by frequency only (stack to full height)
• Additional keyword arguments for customization

logoplot_with_highlight(pfm, highlighted_regions; kwargs...)

Plot a sequence logo with highlighted regions using default background frequencies.

save_logoplot(pfm, background, save_name; kwargs...)

Save a sequence logo plot to a file.

Arguments

• pfm::Matrix{Real}: Position frequency matrix (rows = characters, columns = positions)
• background::Vector{Real}: Background probabilities for each character
• save_name::String: File path to save the plot
• scale_by_frequency::Bool=false: If true, scale letters by frequency only (stack to full height)

Requirements

• pfm must be a probability matrix (each column sums to 1)
• background must be a vector of probabilities that sum to 1
• For nucleotides: background length = 4; for proteins: length = 20

Example

pfm = [0.02 1.0 0.98 0.0; 0.98 0.0 0.02 0.19; ...]
background = [0.25, 0.25, 0.25, 0.25]
save_logoplot(pfm, background, "logo.png"; dpi=65)

save_logoplot(pfm, save_name; kwargs...)

Save a sequence logo plot using default background frequencies.

This is equivalent to save_logoplot(pfm, background, save_name; kwargs...) where background is set to uniform frequencies for the sequence type.

reduce_entropy!(pfm; factor=10)

Sharpen each column of a PFM toward its dominant symbol, in place.

Each column is raised element-wise to the power factor and then renormalized to sum to 1. Larger factor values make the dominant symbol more overwhelming (lower entropy); factor=1 leaves the column unchanged. Useful for cleaning up noisy protein PFMs before plotting.

filter_counts_by_reference(counts, ref; tol=1e-9)

Keep only columns where dot(col, ref_col) ≠ sum(col). When equal, the column has single nonzero entry matching reference → remove it. Returns indices of columns to keep.

count_fragments(count_matrices, reference_pfms, starting_indices; tol=1e-9)
count_fragments(count_matrices, reference_pfms; tol=1e-9)

Filter count matrices by reference and report the resulting contiguous fragments.

Returns (total_fragments::Int, span::String), where span is a human-readable description of the fragments in global coordinates. Without starting_indices, each matrix is assumed to start at position 1.

Span examples:

• single column at 45 -> "45"
• columns 45-48 -> "45-48"
• two fragments at 45 and 51-55 -> "(45, 51-55)"

apply_count_filter(count_matrices, starting_indices, reference_pfms; tol=1e-9)

Filter count matrices by reference, returning new matrices with only non-matching columns. Updates starting indices to reflect new fragment positions.

logoplot_with_rect_gaps(count_matrices, starting_indices, total_length; kwargs...)

Plot several logo fragments positioned along a track of length total_length, joined by rectangular connector glyphs across the gaps between them.

count_matrices is a vector of per-fragment count/frequency matrices and starting_indices gives each fragment's starting position on the track. When reference_pfms (a vector of column-wise one-hot BitMatrix) is supplied and filter_by_reference=true, columns that match the reference are dropped via apply_count_filter before plotting.

Notable keyword arguments

• reference_pfms=nothing, filter_by_reference=true, filter_tolerance=1e-9: reference-based column filtering.
• ref_match_color, ref_mismatch_color: colors for reference-matching / mismatching letters.
• rna, protein, uniform_color, dpi, xrotation, height_top, arrow_shape_scale_ratio, basic_fcn: styling of the logos and connectors.

See also save_logo_with_rect_gaps.

save_logo_with_rect_gaps(count_matrices, starting_indices, total_length, save_name; kwargs...)

Render a gapped multi-fragment logo (see logoplot_with_rect_gaps) and save it to save_name. The output format is inferred from the file extension. Accepts the same keyword arguments as logoplot_with_rect_gaps, including reference-based column filtering.

get_rectangle_basic(; line_scale=1.0, right=true, x_offset=0.0)

Return the basic rectangular arrow-shaft shape used as the connector glyph between logo fragments in the gap/spacer plots. line_scale scales its length, right orients the arrow head, and x_offset shifts it horizontally. Used as the default basic_fcn for logoplot_with_rect_gaps and save_logo_with_rect_gaps.