Skip to content
bedrock.bio

open_targets.association_by_datatype_direct

import bedrock_bio as bb

rel = bb.load_table('open_targets.association_by_datatype_direct')

df = rel.limit(5).pl()  # Polars
df = rel.limit(5).df()  # Pandas
df.head()
diseaseId targetId aggregationType aggregationValue associationScore evidenceCount timeseries currentNovelty version
0 EFO_0000180 ENSG00000070831 datatypeId literature 0.023403 4 [{'year': 2022, 'associationScore': 0.02264215... 0.000747 26.09
1 EFO_0000180 ENSG00000091831 datatypeId literature 0.030415 1 [{'year': 1996, 'associationScore': 0.0, 'nove... 0.003626 26.09
2 EFO_0000180 ENSG00000092201 datatypeId literature 0.012166 1 [{'year': 2021, 'associationScore': 0.0, 'nove... 0.001450 26.09
3 EFO_0000180 ENSG00000127955 datatypeId literature 0.060830 1 [{'year': 2014, 'associationScore': 0.0, 'nove... 0.000003 26.09
4 EFO_0000180 ENSG00000136271 datatypeId literature 0.030415 1 [{'year': 2024, 'associationScore': 0.0, 'nove... 0.029868 26.09

Description

Section titled “Description”

Data type-specific metrics for direct target-disease associations and its development over time implying association novelty.

Dataset contains association scores calculated for each data type indepedently for the aggregation of all evidence supporting any given target-disease.

This dataset only contains direct associations, therefore no evidence is propagated across the disease or phenotype space using the Experimental Factor Ontology [Target-Disease]

Partitions

Section titled “Partitions”

Not partitioned.

Schema

Section titled “Schema”
Column Type Description
diseaseId TEXT Open Targets disease identifier
targetId TEXT Unique identifier for the target [bioregistry:ensembl]
aggregationType TEXT Evidence column used to aggregate data: datatypeId
aggregationValue TEXT Datasource identifier of the group: eg. genetics_literature
associationScore DOUBLE Current disease/target association score of evidence from the given datatype
evidenceCount BIGINT Total number of evidence supporting the disease/target association for a given datatype
timeseries STRUCT(year INT, associationScore DOUBLE, novelty DOUBLE, yearlyEvidenceCount BIGINT)[] Timeseries describing the evolution of association
currentNovelty DOUBLE Novelty of the disease/target association
version TEXT Open Targets Platform release version (major.minor), e.g. 26.09. Added during ingest.