columns name:
-
'relation', 'display_relation',
-
'x_index', 'x_id', 'x_type', 'x_name', 'x_source',
-
'y_index', 'y_id', 'y_type', 'y_name', 'y_source'.
types and numbers of knowledge (based on x_index):
| types | numbers |
|---|---|
| drug | 2805696 |
| gene/protein | 2631229 |
| anatomy | 1566154 |
| disease | 341244 |
| effect/phenotype | 257096 |
| biological_process | 252202 |
| molecular_function | 96723 |
| cellular_component | 93102 |
| pathway | 47716 |
| exposure | 9336 |
for more details, see in kg_split.ipynb Script.
SPO file and outputs of kg_split.ipynb Script
Spo stands for Subject Predicate Object triples, which should be separated by '\t' and without index. For convenience, it should be save in the folder './kgs'
kg_split.ipynb Script offers 2 spo files outputs.
-
CheXpert_KG.spo include knowledge about the labels in MIMIC CXR and CheXpert dataset and anatomical knowledge along with medical effect/phenotype as well. there are 38251 entries in total
-
kg_anatomy3kAndAtelectasis.spo is only used for debug and performs terriblly during the experiment. It includes knowledge about Atelectasis, and relation of drug_effect, of disease_phenotype_positive, and of phenotype_phenotype. there are 3136 entries in total.