We should enforce some additional information on groups.
If disease is set, healthy must be set for the group! I see the due to the parent inheritance many incorrect/unclear encodings in the tables:
E.g. Here, where the disease groups get a partial healthy inheritance from the all group which is incorrect.
study | name | parent | group_count | measurement | choice | count
-- | -- | -- | -- | -- | -- | --
Klotz1975 | all | | 54 | healthy | Y | 33
Klotz1975 | all | | 54 | species | homo sapiens | 54
Klotz1975 | all | | 54 | sex | M | 44
Klotz1975 | all | | 54 | sex | F | 10
Klotz1975 | control | all | 33 | healthy | Y | 33
Klotz1975 | alco_cirr | all | 9 | disease | alcoholic liver cirrhosis | 9
Klotz1975 | viral_hep | all | 8 | disease | viral hepatitis | 8
Klotz1975 | chronic_hep | all | 4 | disease | hepatitis | 4
Enforce some sets of characteristicas:
If disease set, healthy required:
disease -> healthy
Similar dependencies will occur with other characteristica.
Similar dependencies in the amount fields:
smoking amount (cigarettes) -> smoking (some of this informaiton can de deduced, but we should enforce that this is encoded.
We should enforce some additional information on groups. If disease is set, healthy must be set for the group! I see the due to the parent inheritance many incorrect/unclear encodings in the tables: E.g. Here, where the disease groups get a partial healthy inheritance from the all group which is incorrect.
Enforce some sets of characteristicas: If disease set, healthy required:
disease -> healthy
Similar dependencies will occur with other characteristica.