feat: implement benchmark field 2 (#128)

Co-authored-by: Kevin Maik Jablonka <kevin.jablonka@epfl.ch> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
OpenBioML · Mar 23, 2023 · b2ae8dc · b2ae8dc
1 parent ff03074
commit b2ae8dc
Show file tree

Hide file tree

Showing 27 changed files with 158 additions and 24 deletions.
diff --git a/data/caco2_wang/meta.yaml b/data/caco2_wang/meta.yaml
@@ -20,7 +20,10 @@ targets:
       uris:
           - http://www.bioassayontology.org/bao#BAO_0010008
           - http://purl.obolibrary.org/obo/MI_2162
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 identifiers:
     - id: SMILES
       type: SMILES

diff --git a/data/caco2_wang/transform.py b/data/caco2_wang/transform.py
@@ -70,7 +70,13 @@ def get_and_transform_data():
                 ],
             },
         ],
-        "split_col": "split",  # name of the column that contains the split information
+        "benchmarks": [
+            {
+                "name": "TDC",  # unique benchmark name
+                "link": "https://tdcommons.ai/",  # benchmark URL
+                "split_column": "split",  # name of the column that contains the split information
+            },
+        ],
         "identifiers": [
             {
                 "id": "SMILES",  # column name

diff --git a/data/cav3_t-type_calcium_channels_butkiewicz/meta.yaml b/data/cav3_t-type_calcium_channels_butkiewicz/meta.yaml
@@ -37,7 +37,10 @@ links:
       description: corresponding publication
     - url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5962024/
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 100875
 url: https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al
 bibtex:

diff --git a/data/cav3_t-type_calcium_channels_butkiewicz/transform.py b/data/cav3_t-type_calcium_channels_butkiewicz/transform.py
@@ -81,7 +81,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",  # name of the column that contains the split information
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            },
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al",
         "bibtex": [

diff --git a/data/choline_transporter_butkiewicz/meta.yaml b/data/choline_transporter_butkiewicz/meta.yaml
@@ -28,7 +28,10 @@ targets:
           - 588401
           - 493222
           - 602208
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 identifiers:
     - id: SMILES
       type: SMILES

diff --git a/data/choline_transporter_butkiewicz/transform.py b/data/choline_transporter_butkiewicz/transform.py
@@ -67,7 +67,13 @@ def get_and_transform_data():
                 "pubchem_aids": [488975, 493221, 504840, 588401, 493222, 602208],
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "identifiers": [
             {
                 "id": "SMILES",  # column name

diff --git a/data/clintox/meta.yaml b/data/clintox/meta.yaml
@@ -17,7 +17,10 @@ targets:
           - http://purl.bioontology.org/ontology/MESH/Q000633
           - https://ncit.nci.nih.gov/ncitbrowser/ConceptReport.jsp?dictionary=NCI_Thesaurus&ns=ncit&code=C27990
           - https://ncit.nci.nih.gov/ncitbrowser/ConceptReport.jsp?dictionary=NCI_Thesaurus&ns=ncit&code=C27955
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 identifiers:
     - id: SMILES
       type: SMILES

diff --git a/data/clintox/transform.py b/data/clintox/transform.py
@@ -58,7 +58,13 @@ def get_and_transform_data():
                 ],
             },
         ],
-        "split_col": "split",  # name of the column that contains the split information
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "identifiers": [
             {
                 "id": "SMILES",  # column name

diff --git a/data/kcnq2_potassium_channel_butkiewicz/meta.yaml b/data/kcnq2_potassium_channel_butkiewicz/meta.yaml
@@ -35,7 +35,10 @@ links:
       description: corresponding publication
     - url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5962024/
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 302405
 url: https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al
 bibtex:

diff --git a/data/kcnq2_potassium_channel_butkiewicz/transform.py b/data/kcnq2_potassium_channel_butkiewicz/transform.py
@@ -87,7 +87,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al",
         "bibtex": [

diff --git a/data/ld50_zhu/meta.yaml b/data/ld50_zhu/meta.yaml
@@ -29,7 +29,10 @@ license: CC BY 4.0
 links:
     - url: https://doi.org/10.1021/tx900189p
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 7385
 url: https://tdcommons.ai/single_pred_tasks/tox/#acute-toxicity-ld50
 bibtex:

diff --git a/data/ld50_zhu/transform.py b/data/ld50_zhu/transform.py
@@ -87,7 +87,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/tox/#acute-toxicity-ld50",
         "bibtex": [

diff --git a/data/m1_muscarinic_receptor_agonists_butkiewicz/meta.yaml b/data/m1_muscarinic_receptor_agonists_butkiewicz/meta.yaml
@@ -32,7 +32,10 @@ links:
       description: corresponding publication
     - url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5962024/
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 61833
 url: https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al
 bibtex:

diff --git a/data/m1_muscarinic_receptor_agonists_butkiewicz/transform.py b/data/m1_muscarinic_receptor_agonists_butkiewicz/transform.py
@@ -85,7 +85,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            },
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al",
         "bibtex": [

diff --git a/data/m1_muscarinic_receptor_antagonists_butkiewicz/meta.yaml b/data/m1_muscarinic_receptor_antagonists_butkiewicz/meta.yaml
@@ -32,7 +32,10 @@ links:
       description: corresponding publication
     - url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5962024/
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 61756
 url: https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al
 bibtex:

diff --git a/data/m1_muscarinic_receptor_antagonists_butkiewicz/transform.py b/data/m1_muscarinic_receptor_antagonists_butkiewicz/transform.py
@@ -75,7 +75,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",  # name of the column that contains the split information
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al",
         "bibtex": [

diff --git a/data/orexin1_receptor_butkiewicz/meta.yaml b/data/orexin1_receptor_butkiewicz/meta.yaml
@@ -29,7 +29,10 @@ identifiers:
     - id: SMILES
       type: SMILES
       description: SMILES
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 license: CC BY 4.0
 links:
     - url: https://doi.org/10.3390/molecules18010735

diff --git a/data/orexin1_receptor_butkiewicz/transform.py b/data/orexin1_receptor_butkiewicz/transform.py
@@ -61,7 +61,13 @@ def get_and_transform_data():
                 "description": "SMILES",  # description (optional, except for "Other")
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "license": "CC BY 4.0",  # license under which the original dataset was published
         "links": [  # list of relevant links (original dataset, other uses, etc.)
             {

diff --git a/data/pampa_ncats/meta.yaml b/data/pampa_ncats/meta.yaml
@@ -29,7 +29,10 @@ links:
       description: original dataset link
     - url: https://journals.sagepub.com/doi/full/10.1177/24725552211017520
       description: corresponding publication
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 2034
 bibtex:
     - |-

diff --git a/data/pampa_ncats/transform.py b/data/pampa_ncats/transform.py
@@ -72,7 +72,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "bibtex": [
             """@article{siramshetty2021validating,

diff --git a/data/potassium_ion_channel_kir2_1_butkiewicz/meta.yaml b/data/potassium_ion_channel_kir2_1_butkiewicz/meta.yaml
@@ -31,7 +31,10 @@ identifiers:
     - id: SMILES
       type: SMILES
       description: SMILES
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 license: CC BY 4.0
 links:
     - url: https://doi.org/10.3390/molecules18010735

diff --git a/data/potassium_ion_channel_kir2_1_butkiewicz/transform.py b/data/potassium_ion_channel_kir2_1_butkiewicz/transform.py
@@ -72,7 +72,13 @@ def get_and_transform_data():
                 "description": "SMILES",  # description (optional, except for "Other")
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "license": "CC BY 4.0",  # license under which the original dataset was published
         "links": [  # list of relevant links (original dataset, other uses, etc.)
             {

diff --git a/data/serine_threonine_kinase_33_butkiewicz/meta.yaml b/data/serine_threonine_kinase_33_butkiewicz/meta.yaml
@@ -24,7 +24,10 @@ identifiers:
     - id: SMILES
       type: SMILES
       description: SMILES
-split_col: split
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 license: CC BY 4.0
 links:
     - url: https://doi.org/10.3390/molecules18010735

diff --git a/data/serine_threonine_kinase_33_butkiewicz/transform.py b/data/serine_threonine_kinase_33_butkiewicz/transform.py
@@ -70,7 +70,13 @@ def get_and_transform_data():
                 "description": "SMILES",  # description (optional, except for "Other")
             },
         ],
-        "split_col": "split",
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "license": "CC BY 4.0",  # license under which the original dataset was published
         "links": [  # list of relevant links (original dataset, other uses, etc.)
             {

diff --git a/data/tyrosyl-dna_phosphodiesterase_butkiewicz/meta.yaml b/data/tyrosyl-dna_phosphodiesterase_butkiewicz/meta.yaml
@@ -29,6 +29,10 @@ links:
       description: corresponding publication
     - url: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5962024/
       description: corresponding publication
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
 num_points: 341365
 url: https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al
 bibtex:

diff --git a/data/tyrosyl-dna_phosphodiesterase_butkiewicz/transform.py b/data/tyrosyl-dna_phosphodiesterase_butkiewicz/transform.py
@@ -84,6 +84,13 @@ def get_and_transform_data():
                 "description": "corresponding publication",
             },
         ],
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            }
+        ],
         "num_points": len(df),  # number of datapoints in this dataset
         "url": "https://tdcommons.ai/single_pred_tasks/hts/#butkiewicz-et-al",
         "bibtex": [

diff --git a/src/chemnlp/data_val/model.py b/src/chemnlp/data_val/model.py
@@ -151,6 +151,19 @@ class Link(YamlModel):
     description: str
 
 
+class Benchmark(YamlModel):
+    """Benchmark information."""
+
+    """The name of the benchmark, e.g. MoleculeNet."""
+    name: str
+
+    """The link to the benchmark."""
+    link: str
+
+    """The name of the column in the dataset that indicates the fold of the data point."""
+    split_column: str
+
+
 class Dataset(YamlModel):
     name: str
     description: str
@@ -163,6 +176,8 @@ class Dataset(YamlModel):
     fields: Optional[Dict[str, TemplateField]]
     links: List[Link]
 
+    benchmarks: Optional[List[Benchmark]]
+
     @validator("num_points")
     def num_points_must_be_positive(cls, v):
         if v < 0: