The module's key for the variant. Joins weights.parquet.variant_key.
transcript
str
required
Ensembl transcript accession (ENST…), unversioned as the source publishes it. Which transcripts appear is the source's choice: Open Targets lists one canonical transcript per nearby gene, not every transcript.
rsid
str | None
optional
dbSNP id of the variant as the module knows it. Descriptive, outside the fact hash.
gene_id
str | None
optional
Ensembl gene accession (ENSG…) the transcript belongs to.
gene
str | None
optional
HGNC symbol of that gene, as the source labels it. A label; gene_id is the fact.
consequence_so_ids
str | None
optional
Sequence Ontology accessions of the predicted consequence on this transcript (SO_0001583), comma-separated in the source's order when it gives several (a splice-region variant inside an intron is two). Verbatim: accessions, not labels.
impact
str | None
optional
one of: HIGH, LOW, MODERATE, MODIFIER
VEP's impact class for this transcript: HIGH|MODERATE|LOW|MODIFIER. The class of the consequence on THIS transcript, never the most severe across the variant's transcripts.
mane_select
str | None
optional
The RefSeq accession (NM_…) when this transcript is the gene's MANE Select, else null. How a consumer that wants one row per gene picks it without a ranking policy.
canonical
bool | None
optional
Whether the source flags this transcript Ensembl-canonical. True on every row of Open Targets 26.09, which is how its transcript set was chosen; recorded rather than implied.
dataset
str
required
Which release this row came from, e.g. 'opentargets_26.09' (its VEP annotation).
source
str | None
optional
The licensed data source: opentargets|manual (open). Joins sources.csv.source.
fetched_at
str | None
optional
ISO-8601 UTC timestamp when a pass wrote this row; the release is dataset.
Generated from the row model at build time — reference.authoring_reference(), the same answer describe_table gives an authoring tool. Nothing on this page is hand-kept.