From 863b34d0e38c8b4af2d9fe05801643f94081c16e Mon Sep 17 00:00:00 2001 From: Jiahui Chen Date: Tue, 17 Nov 2026 12:19:58 -0700 Subject: [PATCH] M1: state contigs and genome length at the top of genome (#71) Both routes already read the values but kept them only under genome.quality, while the full report looked for a top-level genome.contigs and so never showed it. genome.contigs and genome.genome_length are now top-level integers on the CGA and API routes, declared in the schema, and the report falls back to the quality block for runs written before this. Co-Authored-By: Claude Opus 5.5 --- s2f/common/schema.py | 2 ++ s2f/full_report/render.py | 15 ++++++++++----- s2f/m1_genome/collect.py | 11 +++++++++++ s2f/m1_genome/parse.py | 12 ++++++++++++ tests/test_m1_bvbrc_api.py | 22 ++++++++++++++++++++++ 5 files changed, 57 insertions(+), 5 deletions(-) diff --git a/s2f/common/schema.py b/s2f/common/schema.py index ab747d5..ea70ea3 100644 --- a/s2f/common/schema.py +++ b/s2f/common/schema.py @@ -228,6 +228,8 @@ "closest_genomes": {"type": "array"}, "tree_newick": {"type": ["string", "null"]}, "cga_job_id": {"type": ["string", "null"]}, + "contigs": {"type": ["integer", "null"], "description": "number of contigs (#71)"}, + "genome_length": {"type": ["integer", "null"], "description": "total length in bp (#71)"}, }, "additionalProperties": True, }, diff --git a/s2f/full_report/render.py b/s2f/full_report/render.py index 045e8ef..6dc6710 100644 --- a/s2f/full_report/render.py +++ b/s2f/full_report/render.py @@ -230,9 +230,9 @@ def tab_organism(report: dict[str, Any]) -> str: ("Genus", genome.get("genus")), ("Taxon", genome.get("taxon_id")), ("Genetic code", taxonomy.get("genetic_code")), - ("Length", f"{quality.get('genome_length')} bp" - if quality.get("genome_length") else None), - ("Contigs", genome.get("contigs")), + ("Length", f"{_genome_length(genome)} bp" + if _genome_length(genome) else None), + ("Contigs", genome.get("contigs") or quality.get("contigs")), ("Quality", quality.get("genome_quality")), ("Proteins", len(_rows(report.get("proteins"))))]: if value not in (None, "", "None bp"): @@ -734,8 +734,8 @@ def build_page(report: dict[str, Any], m2_run: dict[str, Any] | None = None) -> for key, value in [("Genome", genome.get("genome_id")), ("Taxon", genome.get("taxon_id")), ("Route", (genome.get("annotation_route") or "").upper()), - ("Length", f"{quality.get('genome_length')} bp" - if quality.get("genome_length") else None), + ("Length", f"{_genome_length(genome)} bp" + if _genome_length(genome) else None), ("Quality", quality.get("genome_quality")), ("Proteins", tallies["genes"] or None), ("Selected", tallies["targets"] or None), @@ -832,3 +832,8 @@ def render_run(run_dir: str | Path) -> str: except (json.JSONDecodeError, OSError): m2_run = {} return build_page(report, m2_run) + + +def _genome_length(genome: dict) -> object: + """Top-level since #71; runs written before that only carry it under quality.""" + return genome.get("genome_length") or _mapping(genome.get("quality")).get("genome_length") diff --git a/s2f/m1_genome/collect.py b/s2f/m1_genome/collect.py index d64e3ba..ecbae2d 100644 --- a/s2f/m1_genome/collect.py +++ b/s2f/m1_genome/collect.py @@ -679,6 +679,9 @@ def genome_section(genome: dict[str, Any], bundle: dict[str, Any], "cga_job_id": None, # --- API-route provenance and metadata (no CGA equivalent) --- "annotation_route": "api", + # Same keys as the CGA route (#71). Here they describe the reference genome. + "contigs": _int_or_none(genome.get("contigs")), + "genome_length": _int_or_none(genome.get("genome_length")), "annotation_source": bundle.get("annotation", "PATRIC"), "resolution": resolution, "bvbrc_url": GENOME_URL.format(gid=genome.get("genome_id")), @@ -895,3 +898,11 @@ def taxonomy_tree_text(lineage: list[dict[str, Any]], genome: dict[str, Any], lines.append(f" - {neighbor.get('genome_name')} [{flag}] " f"({neighbor.get('species')})") return "\n".join(lines) + + +def _int_or_none(value: Any) -> int | None: + """A count from a JSON record, or None; BV-BRC sometimes sends numbers as strings.""" + try: + return int(value) if value not in (None, "") else None + except (TypeError, ValueError): + return None diff --git a/s2f/m1_genome/parse.py b/s2f/m1_genome/parse.py index 3946e20..fd5b01a 100644 --- a/s2f/m1_genome/parse.py +++ b/s2f/m1_genome/parse.py @@ -329,6 +329,10 @@ def genome_section(run: CgaRun, taxon_call: dict[str, Any] | None = None) -> dic # fallback stays for runs written before this, but it is no longer the only way # to identify a CGA run. "annotation_route": "cga", + # Top-level so a report can quote the genome's size without digging into the + # quality block (#71). Same keys and types on the API route. + "contigs": _int_or_none(run.genome.get("contigs")), + "genome_length": _int_or_none(run.genome.get("genome_length")), "quality": { "genome_quality": run.quality, "genome_quality_flags": run.quality_flags, @@ -412,3 +416,11 @@ def write_m1_dir(run: CgaRun, m1_dir: str | Path) -> dict[str, int]: spec_rows += 1 return {"proteins_faa": written, "feature_rows": rows, "specialty_rows": spec_rows} + + +def _int_or_none(value: Any) -> int | None: + """A count from a JSON record, or None; BV-BRC sometimes sends numbers as strings.""" + try: + return int(value) if value not in (None, "") else None + except (TypeError, ValueError): + return None diff --git a/tests/test_m1_bvbrc_api.py b/tests/test_m1_bvbrc_api.py index 9b3fe78..656989a 100644 --- a/tests/test_m1_bvbrc_api.py +++ b/tests/test_m1_bvbrc_api.py @@ -443,3 +443,25 @@ def test_both_m1_routes_declare_their_own_annotation_route(api_bundle, kp_payloa "both routes report the same annotation_route, which defeats the point of the " "field: a consumer cannot tell a submitted assembly from a reference genome." ) + + +def test_both_routes_state_the_genome_size_at_the_top_level(api_bundle, kp_payload) -> None: + """#71: `genome.contigs` and `genome.genome_length`, same keys and types on both routes.""" + from s2f.common.schema import validate + + api = collect.genome_section(kp_payload["genome"], api_bundle, "explicit genome_id") + cga = parse.genome_section( + parse.load_cga(Path(__file__).resolve().parents[1] / "fixtures/m1/cga_sample")) + + assert (api["contigs"], api["genome_length"]) == (7, 5682322) + assert (cga["contigs"], cga["genome_length"]) == (1, 580076) + for section in (api, cga): + validate("genome", section) + + +def test_the_full_report_shows_contigs_and_length() -> None: + from s2f.full_report.render import _genome_length + + assert _genome_length({"genome_length": 580076}) == 580076 + # Runs written before #71 only carry it under quality. + assert _genome_length({"quality": {"genome_length": 580076}}) == 580076