diff --git a/data/uspto_500k/meta.yaml b/data/uspto_500k/meta.yaml
new file mode 100644
index 000000000..0d1cfb0f1
--- /dev/null
+++ b/data/uspto_500k/meta.yaml
@@ -0,0 +1,41 @@
+---
+name: uspto_500k
+description: United States Patent and Trademark Office reaction dataset with yields.
+targets:
+    - id: yield
+      description: reaction yields
+      units: '%'
+      type: continuous
+      names:
+          - reaction yield
+          - yield
+      uris:
+          - http://purl.allotrope.org/ontologies/quality#AFQ_0000227
+identifiers:
+    - id: reaction_SMILES
+      type: RXNSMILES
+      description: reaction SMILES
+license: CC0
+links:
+    - url: https://doi.org/10.17863/CAM.16293
+      description: corresponding publication
+    - url: https://github.com/reymond-group/drfp/blob/main/data/uspto_yields_below.csv
+      description: data source
+    - url: https://github.com/reymond-group/drfp/blob/main/data/uspto_yields_above.csv
+      description: data source
+    - url: https://tdcommons.ai/single_pred_tasks/yields/#uspto
+      description: other source
+num_points: 853638
+bibtex:
+    - |-
+      @article{https://doi.org/10.17863/cam.16293,
+      doi = {10.17863/CAM.16293},
+      url = {https://www.repository.cam.ac.uk/handle/1810/244727},
+      year = {2012},
+      publisher = {Apollo - University of Cambridge Repository},
+      keywords = {Name to structure,  OPSIN,  Chemical text mining,  Text mining,
+      Patent reaction extraction,  Reaction mining,  Patents},
+      language = {en},
+      author = {Lowe,  Daniel Mark},
+      title = {Extraction of chemical structures and reactions from the literature},
+      copyright = {All Rights Reserved}
diff --git a/data/uspto_500k/transform.py b/data/uspto_500k/transform.py
new file mode 100644
index 000000000..f85e4a13d
--- /dev/null
+++ b/data/uspto_500k/transform.py
@@ -0,0 +1,135 @@
+import pandas as pd
+import yaml
+from tdc.single_pred import Yields
+
+
+def get_and_transform_data():
+    # get raw data
+    data = Yields(name="USPTO_Yields")
+    splits = data.get_split()
+    df_train = splits["train"]
+    df_valid = splits["valid"]
+    df_test = splits["test"]
+    df_train["split"] = "train"
+    df_valid["split"] = "valid"
+    df_test["split"] = "test"
+    df = pd.concat([df_train, df_valid, df_test], axis=0)
+
+    df["catalyst"] = df.Reaction.apply(lambda x: x["catalyst"])
+    df["reactant"] = df.Reaction.apply(lambda x: x["reactant"])
+    df["product"] = df.Reaction.apply(lambda x: x["product"])
+    df = df.drop("Reaction", axis=1)
+
+    fn_data_original = "data_original.csv"
+    df.to_csv(fn_data_original, index=False)
+    del df
+
+    # create dataframe
+    df = pd.read_csv(
+        fn_data_original, delimiter=","
+    )  # not necessary but ensure we can load the saved data
+
+    # check if fields are the same
+    fields_orig = df.columns.tolist()
+    assert fields_orig == [
+        "Reaction_ID",
+        "Y",
+        "split",
+        "catalyst",
+        "reactant",
+        "product",
+    ]
+    fields_clean = ["Reaction_ID", "yield", "split", "catalyst", "reactant", "product"]
+    # overwrite column names = fields
+    df.columns = fields_clean
+    assert df.columns.tolist() == fields_clean
+
+    # remove leading and trailing white space characters
+    assert not df.duplicated().sum()
+
+    # save to csv
+    fn_data_csv = "data_clean.csv"
+    df.to_csv(fn_data_csv, index=False)
+
+    # create meta yaml
+    meta = {
+        "name": "uspto_500k",
+        "description": """United States Patent and Trademark Office reaction dataset with yields.""",
+        "targets": [
+            {
+                "id": "yield",
+                "description": "reaction yields",
+                "units": "%",
+                "type": "continuous",
+                "names": [
+                    "reaction yield",
+                    "yield",
+                ],
+                "uris": [
+                    "http://purl.allotrope.org/ontologies/quality#AFQ_0000227",
+                ],
+            },
+        ],
+        "identifiers": [
+            {
+                "id": "reaction_SMILES",
+                "type": "RXNSMILES",
+                "description": "reaction SMILES",
+            },
+        ],
+        "license": "CC0",  # license under which the original dataset was published
+        "links": [  # list of relevant links (original dataset, other uses, etc.)
+            {
+                "url": "https://doi.org/10.17863/CAM.16293",
+                "description": "corresponding publication",
+            },
+            {
+                "url": "https://github.com/reymond-group/drfp/blob/main/data/uspto_yields_below.csv",
+                "description": "data source",
+            },
+            {
+                "url": "https://github.com/reymond-group/drfp/blob/main/data/uspto_yields_above.csv",
+                "description": "data source",
+            },
+            {
+                "url": "https://tdcommons.ai/single_pred_tasks/yields/#uspto",
+                "description": "other source",
+            },
+        ],
+        "num_points": len(df),  # number of datapoints in this dataset
+        "bibtex": [
+            """@article{https://doi.org/10.17863/cam.16293,
+doi = {10.17863/CAM.16293},
+url = {https://www.repository.cam.ac.uk/handle/1810/244727},
+year = {2012},
+publisher = {Apollo - University of Cambridge Repository},
+keywords = {Name to structure,  OPSIN,  Chemical text mining,  Text mining,
+Patent reaction extraction,  Reaction mining,  Patents},
+language = {en},
+author = {Lowe,  Daniel Mark},
+title = {Extraction of chemical structures and reactions from the literature},
+copyright = {All Rights Reserved}""",
+        ],
+    }
+
+    def str_presenter(dumper, data):
+        """configures yaml for dumping multiline strings
+        Ref: https://stackoverflow.com/questions/8640959/how-can-i-control-what-scalar-form-pyyaml-uses-for-my-data
+        """
+        if data.count("\n") > 0:  # check for multiline string
+            return dumper.represent_scalar("tag:yaml.org,2002:str", data, style="|")
+        return dumper.represent_scalar("tag:yaml.org,2002:str", data)
+
+    yaml.add_representer(str, str_presenter)
+    yaml.representer.SafeRepresenter.add_representer(
+        str, str_presenter
+    )  # to use with safe_dum
+    fn_meta = "meta.yaml"
+    with open(fn_meta, "w") as f:
+        yaml.dump(meta, f, sort_keys=False)
+
+    print(f"Finished processing {meta['name']} dataset!")
+
+
+if __name__ == "__main__":
+    get_and_transform_data()