API Reference

Signatures and short descriptions for every public name, grouped by module, with examples that run on 2.0. Use Ctrl-F / the search box to jump around. Everything you need day to day is one import away:

import genomeblocks as gb

Top-level name	What it is	Page
`gb.Loci`, `gb.Locus`, `gb.Genome`	Intervals as a table; one interval; chromosome names ↔ codes.	loci, locus
`gb.Genes`	Genes / transcripts / features, three linked tables.	genes
`gb.Pairs`	BEDPE as a table (anchors `a` and `b`).	bedpe
`gb.Architecture`	A CRE interaction graph: vertices = a `Loci`, edges = a table.	architecture
`gb.Atlas`	A sparse bins x tracks index for enrichment.	atlas
`gb.as_loci`	Anything interval-like → `Loci`; what every function calls on its inputs.	interop
`gb.read_fasta`, `gb.load_motifs`	A FASTA as `{chrom: seq}`; a motif file as a `Library`.	backends
`gb.backends()`, `gb.use_backend`	List the engines; pick one for a block of code.	backends
`gb.tmm`, `gb.compare_heatmap`, `gb.plot_motif_heatmap`	Signal normalisation and the two cross-set plots.	signal
`gb.browser`, `gb.View`, `gb.igv_html`, `gb.coverage`	The matplotlib browser, the one-file HTML viewer, igv.js pages, BAM coverage.	browser

Names are imported on first use, so import genomeblocks stays light.

Modules

Module	Public names	Page
`genomeblocks.locus`	`Locus`, `parse_region`, `parse_regions` (and `loci.LocusView`, the row view)	locus
`genomeblocks.genome`	`Genome` (`from_sizes`, `from_fasta`, `encode` / `decode`, `rank`, `sizes`), `read_sizes`	loci, Concepts
`genomeblocks.loci`	`Loci`, `LocusView`, `STRANDS`, `SCODE`	loci
`genomeblocks.genes`	`Genes`, `GeneView`, `tss_base`, `LABELS`, `KINDS`	genes
`genomeblocks.bedpe`	`Pairs`, `read_bedpe`, `as_pairs`, `read_pairs_chunks`, `PAIRS_FORMAT_COLUMNS`, `count_pairs`, `count_pairs_2d`, `pair_2d_block`, `pair_2d_to_frame`	bedpe
`genomeblocks.architecture`	`Architecture`, `Props`	architecture
`genomeblocks.architecture_draw`	`draw`	architecture
`genomeblocks.signal`	`signal`, `tmm`	signal
`genomeblocks.signal_draw`	`group_mask`, `plot_heatmap`, `plot_profiles`, `compare_heatmap`	signal
`genomeblocks.motifs_draw`	`plot_motif_heatmap`, `plot_archetype`, `plot_archetypes`, `plot_cluster_members`, `plot_dendrogram`	signal
`genomeblocks.browserview`	`browser` (re-exported as `gb.browser`)	browser
`genomeblocks.bam`	`coverage`, `pileup_counts`, `reference_seq`, `BASES`	browser
`genomeblocks.igv`	`igv_html`, `IGV_VERSION`, `IGV_CDN`	browser
`genomeblocks.view`	`View`, `view_html`	browser
`genomeblocks.motifs`	`scan_motifs`, `scan_motifs_matrix`, `scan_motifs_matrix_masked`, `scan_motifs_profile`, `bootstrap_enrichment`, `compare_motifs`, `compare_motifs_to_ref`, `pwm_distance_matrix`, `cluster_motifs`, `archetype`, `archetype_from_names`, `build_archetypes`, `write_meme` (re-exports `Library`, `load_motifs`)	motifs
`genomeblocks.atlas`	`Atlas`	atlas
`genomeblocks.se`	`knee`, `stitch_peaks`, `call_se`, `nearest_gene_within`	atlas
`genomeblocks.hichip`	`shortrange_ends`, `write_bed`, `fragments`, `coverage`, `to_bigwig`, `to_bedgraph`, `macs3`, `shortrange_track`	atlas
`genomeblocks.interop`	`as_loci`, `frame`, `loci_from_frame` / `loci_from_pyranges` / `loci_from_bedtool` / `loci_from_anndata`, `loci_to_pandas` / `loci_to_polars` / `loci_to_arrow` / `loci_to_bioframe` / `loci_to_pyranges` / `loci_to_bedtool` / `loci_to_cgranges` / `loci_to_anndata`, `cube_to_xarray` / `cube_to_anndata` / `cube_to_pandas`, `sequences`, `windows`, `revcomp`, `to_seqrecords`, `write_fasta`, `liftover`	interop
`genomeblocks.backends`	`families`, `installed`, `resolve`, `use_backend`, `backends`, `unsupported`, `AUTO`	backends
`genomeblocks.backends.intervals`	`overlap_pairs`, `overlap_any`, `nearest`, `merge`, `point_rows`	backends
`genomeblocks.backends.bigwig`	`open_bigwig`, `handle_backend`, `PyBigToolsHandle`, `PyBigWigHandle`, `PythonHandle`	backends
`genomeblocks.backends.motifs`	`Library`, `load_motifs`, `logodds_matrix`, `threshold_from_pvalue`, `write_meme`, `Block`, `FORMATS`, `PSEUDOCOUNT`	backends
`genomeblocks.backends.fasta`	`read_fasta`, `open_fasta`, `IndexedSource`, `PysamSource`, `PyfaidxSource`, `MemorySource`, `BiopythonSource`	backends
`genomeblocks.backends.tables`	`sniff`, `read_columns`	backends
`genomeblocks.backends.graph`	`to_scipy`, `native`, `components`, `pagerank`, `layout`, `layout_edges`	backends

Conventions

Coordinates are 0-based, half-open in every table ([start, end); a GTF’s 1-based start becomes start - 1 once, while parsing). A gene’s TSS is the 1-bp interval [t, t + 1) with t = start on + and t = end - 1 on -. See Concepts.
Rows are the join key. Anything computed from a Loci — a signal cube, a motif matrix, labels, Architecture vertex columns — has one row per locus in the table’s order.
Inputs go through as_loci: every function that takes intervals takes a Loci, a path, a region string, a Locus, a pandas / polars / pyarrow / bioframe / pyranges / pybedtools frame, a dict of columns, an AnnData or a list of regions.
Outputs: every container has shape, columns, head(), tail(), describe(), to_pandas() / to_polars() / to_arrow(), _repr_html_() and the Arrow C stream, dataframe-interchange and narwhals protocols, so polars, pyarrow, DuckDB, seaborn, plotly and altair take it as is.
Backends: backend= on a call or with gb.use_backend(family=name): picks the engine; a requested engine that is missing raises ImportError with the install command, never a silent switch. There is no global Genome and no global backend setting.
Signatures on these pages are copied from the source; keyword-only arguments follow the *.

API Reference

Modules

Conventions

Table of contents