---

name: genomics-and-trees description: "Guide pyCirclize biological circular plots from local GenBank, GFF, BED/cytoband, and Newick or Bio.Phylo inputs, including sequence features, GC summaries, and TreeViz annotation." disable-model-invocation: true metadata: disco-role: operating license: MIT

Genomics and trees

Use this route when the task is to turn biological coordinates or a phylogenetic tree into a pyCirclize circular plot. Prefer local files, file-like objects, or in-memory Biopython objects. Keep the input parser, coordinate convention, sequence-id mapping, and plot track aligned before rendering.

Route by input

• GenBank: use pycirclize.parser.Genbank for sequences plus qualified SeqFeature objects. Use get_seqid2size() to create Circos sectors and get_seqid2features() or extract_features() for feature tracks.
• GFF: use pycirclize.parser.Gff. Select target_seqid for a one-seqid view, or use get_seqid2features() to retain a mapping for all seqids.
• BED or cytoband: use Circos.initialize_from_bed() for chromosome sectors. Then call add_cytoband_tracks() with a local UCSC-style BED-like cytoband file. Use data-parsers for generic tables and circular-composition for lifecycle/composition questions.
• Newick/Bio.Phylo: use Circos.initialize_from_tree() or attach track.tree() to an existing sector. The returned TreeViz owns node styling, highlights, markers, and confidence labels.

For generic Track drawing, Sector/Track radius conventions, and ordinary annotations, link to plot-primitives. For matrix/table preparation, link to data-parsers. For figure creation, shared axes, links, and final export, link to circular-composition.

Safe operating sequence

1. Identify whether coordinates are 0-based half-open (BED) or 1-based closed (GFF); GenBank becomes Biopython 0-based feature locations.
2. Parse locally and inspect names, seqid_list, records, and get_seqid2size() before constructing sectors.
3. Construct sectors from the same identifiers and lengths used by the feature mapping. Check strand and qualifier availability before styling.
4. Add dedicated tracks for forward/reverse features, GC content/skew, or cytobands. Keep all plotted x coordinates inside their sector range.
5. For trees, inspect tv.leaf_labels, tv.innode_labels, and tv.all_node_labels before querying nodes. Apply styling before plotfig()/savefig().
6. Render with Matplotlib's Agg backend for batch work and verify a non-empty output file. Use the bundled deterministic smoke test for a minimal check: scripts/genomics_tree_smoke.py.

Data-source boundary

load_example_tree_file() and load_example_image_file() resolve packaged local example data. load_prokaryote_example_file(), load_eukaryote_example_dataset(), and fetch_genbank_by_accid() are network-backed helpers that may download and cache data; never call them implicitly. Ask for an explicit dataset/cache policy, or replace them with user-provided local files. The optional ipympl tooltip extra is not required for static plots or PNG export.

Detailed signatures, coordinate rules, parser properties, and TreeViz methods are in api-reference.md. Input conventions and local/network boundaries are in data-formats.md. Runnable recipes are in workflows.md, and recovery for malformed or mismatched biological data is in troubleshooting.md.