diff --git a/cellxgene_schema_cli/tests/test_schema_compliance.py b/cellxgene_schema_cli/tests/test_schema_compliance.py index ceaa373f..58e8c722 100644 --- a/cellxgene_schema_cli/tests/test_schema_compliance.py +++ b/cellxgene_schema_cli/tests/test_schema_compliance.py @@ -3352,3 +3352,44 @@ def test_sex_ontology_term_id__invalid(self, validator_with_roundworm_adata): "'sex_ontology_term_id' MUST be 'PATO:0000384' for male, 'PATO:0001340' for hermaphrodite, or 'unknown'." ) assert error_message in validator.errors + + +class TestMultiSpecies: + """ + Tests to verify our support for human / mouse is not impacted by support for additional species + """ + + @pytest.mark.parametrize( + "cell_type_ontology_term_id", + [ + "UBERON:0000001", # Wrong ontology + "ZFA:0000003", # Valid for zebrafish, not valid for human or mouse data + "FBbt:00049192", # Valid for fruit fly, not valid for human or mouse data + "WBbt:0008611", # Valid for roundworm, not valid for human or mouse data + "na", # Allowed for other organisms, not allowed if organism is fruit fly + ], + ) + def test_cell_type_ontology_term_id__invalid(self, validator_with_adata, cell_type_ontology_term_id): + validator = validator_with_adata + obs = validator.adata.obs + obs.loc[obs.index[0], "cell_type_ontology_term_id"] = cell_type_ontology_term_id + validator.validate_adata() + assert len(validator.errors) > 0 + + @pytest.mark.parametrize( + "tissue_ontology_term_id", + [ + "CL:0000001", # Wrong ontology + "ZFA:0001262", # Valid for zebrafish, not valid for human or mouse data + "FBbt:00007337", # Valid for fruit fly, not valid for human or mouse data + "WBbt:0006749", # Valid for roundworm, not valid for human or mouse data + "na", + "unknown", + ], + ) + def test_tissue_ontology_term_id__invalid(self, validator_with_adata, tissue_ontology_term_id): + validator = validator_with_adata + obs = validator.adata.obs + obs.loc[obs.index[0], "tissue_ontology_term_id"] = tissue_ontology_term_id + validator.validate_adata() + assert len(validator.errors) > 0 diff --git a/schema/drafts/5.3.0.md b/schema/drafts/5.3.0.md index 93d8178b..5ac33c79 100644 --- a/schema/drafts/5.3.0.md +++ b/schema/drafts/5.3.0.md @@ -259,29 +259,29 @@ ENSEMBL identifiers are required for genes and [External RNA Controls Consortium The following gene annotation dependencies are *pinned* for this version of the schema. ~~For multi-organism experiments, cells from any Metazoan organism are allowed as long as orthologs from the following organism annotations are used.~~ -| Organism | Source | Required version | Download | +| NCBITaxon | Source | Required version | Download | |:--|:--|:--|:--| -| NCBITaxon:6239
for Caenorhabditis elegans | [ENSEMBL](https://www.ensembl.org/Caenorhabditis_elegans/Info/Index) | WBcel235 (GCA_000002985.3)
Ensembl 113 | [Caenorhabditis_elegans.WBcel235.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/caenorhabditis_elegans/Caenorhabditis_elegans.WBcel235.113.gtf.gz) | +| NCBITaxon:6239
for Caenorhabditis elegans | [ENSEMBL](https://www.ensembl.org/Caenorhabditis_elegans/Info/Index) | WBcel235 (GCA_000002985.3)
Ensembl 113 | [Caenorhabditis_elegans.WBcel235.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/caenorhabditis_elegans/Caenorhabditis_elegans.WBcel235.113.gtf.gz) | | NCBITaxon:9483
for Callithrix jacchus | [ENSEMBL](https://www.ensembl.org/Callithrix_jacchus/Info/Index) | mCalJac1.pat.X
(GCA_011100555.1)
Ensembl 113 | [Callithrix_jacchus.mCalJac1.pat.X.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/callithrix_jacchus/Callithrix_jacchus.mCalJac1.pat.X.113.gtf.gz) | | NCBITaxon:7955
for Danio rerio | [ENSEMBL](https://www.ensembl.org/Danio_rerio/Info/Index) | GRCz11 (GCA_000002035.4)
Ensembl 113 | [Danio_rerio.GRCz11.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/danio_rerio/Danio_rerio.GRCz11.113.gtf.gz) | | NCBITaxon:7227
for Drosophila melanogaster| [ENSEMBL](https://www.ensembl.org/Drosophila_melanogaster/Info/Index) | BDGP6.46 (GCA_000001215.4)
Ensembl 113 | [Drosophila_melanogaster.BDGP6.46.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/drosophila_melanogaster/Drosophila_melanogaster.BDGP6.46.113.gtf.gz) | | NCBITaxon:9595
for Gorilla gorilla gorilla | [ENSEMBL](https://www.ensembl.org/Gorilla_gorilla/Info/Index) | gorGor4
(GCA_000151905.3)
Ensembl 113 | [Gorilla_gorilla.gorGor4.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/gorilla_gorilla/Gorilla_gorilla.gorGor4.113.gtf.gz) | | NCBITaxon:9606
for Homo sapiens | [GENCODE](https://www.gencodegenes.org/human/) | Human reference GRCh38.p14
(GENCODE v44/Ensembl 110) | [gencode.v44.primary_assembly.annotation.gtf](https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_44/gencode.v44.primary_assembly.annotation.gtf.gz) | -| NCBITaxon:10090
for Mus musculus | [GENCODE](https://www.gencodegenes.org/mouse/) | Mouse reference GRCm39
(GENCODE vM33/Ensembl 110) | [gencode.vM33.primary_assembly.annotation.gtf](https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M33/gencode.vM33.primary_assembly.annotation.gtf.gz) | | NCBITaxon:9541
for Macaca fascicularis | [ENSEMBL](https://www.ensembl.org/Macaca_fascicularis/Info/Index) | Macaca_fascicularis_6.0
(GCA_011100615.1)
Ensembl 113 | [Macaca_fascicularis.Macaca_fascicularis_6.0.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/macaca_fascicularis/Macaca_fascicularis.Macaca_fascicularis_6.0.113.gtf.gz) | | NCBITaxon:9544
for Macaca mulatta | [ENSEMBL](https://www.ensembl.org/Macaca_mulatta/Info/Index) | Mmul_10
(GCA_003339765.3)
Ensembl 113 | [Macaca_mulatta.Mmul_10.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/macaca_mulatta/Macaca_mulatta.Mmul_10.113.gtf.gz) | -| NCBITaxon:30608
for Microcebus murinus | [ENSEMBL](https://www.ensembl.org/Microcebus_murinus/Info/Index) | Mmur_3.0
(GCA_000165445.3)
Ensembl 113| [Microcebus_murinus.Mmur_3.0.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/microcebus_murinus/Microcebus_murinus.Mmur_3.0.113.gtf.gz) | +| NCBITaxon:30608
for Microcebus murinus | [ENSEMBL](https://www.ensembl.org/Microcebus_murinus/Info/Index) | Mmur_3.0
(GCA_000165445.3)
Ensembl 113| [Microcebus_murinus.Mmur_3.0.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/microcebus_murinus/Microcebus_murinus.Mmur_3.0.113.gtf.gz) | +| NCBITaxon:10090
for Mus musculus | [GENCODE](https://www.gencodegenes.org/mouse/) | Mouse reference GRCm39
(GENCODE vM33/Ensembl 110) | [gencode.vM33.primary_assembly.annotation.gtf](https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M33/gencode.vM33.primary_assembly.annotation.gtf.gz) | | NCBITaxon:9986
for Oryctolagus cuniculus | [ENSEMBL](https://www.ensembl.org/Oryctolagus_cuniculus/Info/Index) | OryCun2.0
(GCA_000003625.1)
Ensembl 113 | [Oryctolagus_cuniculus.OryCun2.0.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/oryctolagus_cuniculus/Oryctolagus_cuniculus.OryCun2.0.113.gtf.gz) | | NCBITaxon:9598
for Pan troglodytes | [ENSEMBL](https://www.ensembl.org/Pan_troglodytes/Info/Index) | Pan_tro_3.0
(GCA_000001515.5)
Ensembl 113 | [Pan_troglodytes.Pan_tro_3.0.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/pan_troglodytes/Pan_troglodytes.Pan_tro_3.0.113.gtf.gz) | | NCBITaxon:10116
for Rattus norvegicus | [ENSEMBL](https://www.ensembl.org/Rattus_norvegicus/Info/Index) | mRatBN7.2
(GCA_015227675.2)
Ensembl 113| [Rattus_norvegicus.mRatBN7.2.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/rattus_norvegicus/Rattus_norvegicus.mRatBN7.2.113.gtf.gz) | | NCBITaxon:2697049
for SARS-CoV-2 | [ENSEMBL](https://covid-19.ensembl.org/index.html) | SARS-CoV-2 reference (ASM985889v3) | [Sars\_cov\_2.ASM985889v3.101.gtf](https://ftp.ensemblgenomes.org/pub/viruses/gtf/sars_cov_2/Sars_cov_2.ASM985889v3.101.gtf.gz) | | NCBITaxon:9823
for Sus scrofa | [ENSEMBL](https://www.ensembl.org/Sus_scrofa/Info/Index) | Sscrofa11.1
(GCA_000003025.6)
Ensembl 113 | [Sus_scrofa.Sscrofa11.1.113.gtf](https://ftp.ensembl.org/pub/release-113/gtf/sus_scrofa/Sus_scrofa.Sscrofa11.1.113.gtf.gz) | -| | [ThermoFisher ERCC Spike-Ins] | ThermoFisher ERCC RNA Spike-In Control Mixes (Cat # 4456740, 4456739) | [cms_095047.txt] | +| NCBITaxon:32630
for synthetic construct | [ThermoFisher ERCC
Spike-Ins] | ThermoFisher ERCC RNA Spike-In Control Mixes (Cat # 4456740, 4456739) | [cms_095047.txt] | ||||| [RNA Spike-In Control Mixes]: https://www.thermofisher.com/document-connect/document-connect.html?url=https%3A%2F%2Fassets.thermofisher.com%2FTFS-Assets%2FLSG%2Fmanuals%2Fcms_086340.pdf&title=VXNlciBHdWlkZTogRVJDQyBSTkEgU3Bpa2UtSW4gQ29udHJvbCBNaXhlcyAoRW5nbGlzaCAp -[ThermoFisher ERCC Spike-Ins]: https://www.thermofisher.com/order/catalog/product/4456740#/4456740 +[ThermoFisher ERCC
Spike-Ins]: https://www.thermofisher.com/order/catalog/product/4456740#/4456740 [cms_095047.txt]: https://assets.thermofisher.com/TFS-Assets/LSG/manuals/cms_095047.txt