update

README.md

@@ -1,5 +1,5 @@
 ---
-title: Molecular properties
+title: Protein properties
 emoji: 💡
 colorFrom: green
 colorTo: blue

app.py

@@ -4,71 +4,60 @@ import pathlib
 import gradio as gr
 import numpy as np
 import pandas as pd
-from gt4sd.properties.molecules import MOLECULE_PROPERTY_PREDICTOR_FACTORY
+from gt4sd.properties.proteins import PROTEIN_PROPERTY_PREDICTOR_FACTORY
 
 from utils import draw_grid_predict
 
 logger = logging.getLogger(__name__)
 logger.addHandler(logging.NullHandler())
 
-REMOVE = ["docking", "docking_tdc", "molecule_one", "askcos", "plogp"]
-REMOVE.extend(["similarity_seed", "activity_against_target", "organtox"])
 
-MODEL_PROP_DESCRIPTION = {
-    "Tox21": "NR-AR, NR-AR-LBD, NR-AhR, NR-Aromatase, NR-ER, NR-ER-LBD, NR-PPAR-gamma, SR-ARE, SR-ATAD5, SR-HSE, SR-MMP, SR-p53",
-    "Sider": "Hepatobiliary disorders,Metabolism and nutrition disorders,Product issues,Eye disorders,Investigations,Musculoskeletal disorders,Gastrointestinal disorders,Social circumstances,Immune system disorders,Reproductive system and breast disorders,Bening & malignant,General disorders,Endocrine disorders,Surgical & medical procedures,Vascular disorders,Blood & lymphatic disorders,Skin & subcutaneous disorders,Congenital & genetic disorders,Infections,Respiratory & thoracic disorders,Psychiatric disorders,Renal & urinary disorders,Pregnancy conditions,Ear disorders,Cardiac disorders,Nervous system disorders,Injury & procedural complications",
-    "Clintox": "FDA approval, Clinical trial failure",
-}
+AMIDE_FNS = ["protein_weight", "charge", "charge_density", "isoelectric_point"]
+PH_FNS = ["charge", "charge_density", "isoelectric_point"]
 
 
-def main(property: str, smiles: str, smiles_file: str):
+def main(property: str, seq: str, seq_file: str, amide: bool, ph: float):
 
-    algo, config = MOLECULE_PROPERTY_PREDICTOR_FACTORY[property.lower()]
-    kwargs = (
-        {"algorithm_version": "v0"} if property in MODEL_PROP_DESCRIPTION.keys() else {}
-    )
+    prop_name = property.lower()
+    algo, config = PROTEIN_PROPERTY_PREDICTOR_FACTORY[prop_name]
+
+    # Pass hyperparameters if applicable
+    kwargs = {}
+    if prop_name in AMIDE_FNS:
+        kwargs["amide"] = amide
+    if prop_name in PH_FNS:
+        kwargs["ph"] = ph
     model = algo(config(**kwargs))
-    if smiles is not None and smiles_file is not None:
-        raise ValueError("Pass either smiles or smiles_file, not both.")
-    elif smiles is not None:
-        smiles = [smiles]
-    elif smiles_file is not None:
-        smiles = pd.read_csv(smiles_file.name, header=None, sep="\t")[0].tolist()
-    props = np.array(list(map(model, smiles))).round(2)
+
+    # Read and parse data
+    if seq is not None and seq_file is not None:
+        raise ValueError("Pass either smiles or seq_file, not both.")
+    elif seq is not None:
+        seqs = [seq]
+    elif seq_file is not None:
+        seqs = pd.read_csv(seq_file.name, header=None, sep="\t")[0].tolist()
+    props = np.array(list(map(model, seqs))).round(2)
 
     # Expand to 2D array if needed
     if len(props.shape) == 1:
         props = np.expand_dims(np.array(props), -1)
 
-    if property in MODEL_PROP_DESCRIPTION.keys():
-        property_names = MODEL_PROP_DESCRIPTION[property].split(",")
-    else:
-        property_names = [property]
-
-    return draw_grid_predict(
-        smiles, props, property_names=property_names, domain="Molecules"
-    )
+    return draw_grid_predict(seqs, props, property_names=[property], domain="Proteins")
 
 
 if __name__ == "__main__":
 
     # Preparation (retrieve all available algorithms)
-    properties = list(MOLECULE_PROPERTY_PREDICTOR_FACTORY.keys())[::-1]
-    for prop in REMOVE:
-        prop_to_idx = dict(zip(properties, range(len(properties))))
-        properties.pop(prop_to_idx[prop])
+    properties = list(PROTEIN_PROPERTY_PREDICTOR_FACTORY.keys())[::-1]
     properties = list(map(lambda x: x.capitalize(), properties))
 
     # Load metadata
     metadata_root = pathlib.Path(__file__).parent.joinpath("model_cards")
 
     examples = [
-        ["Qed", None, metadata_root.joinpath("examples.smi")],
-        [
-            "Esol",
-            "CN1CCN(CCCOc2ccc(N3C(=O)C(=Cc4ccc(Oc5ccc([N+](=O)[O-])cc5)cc4)SC3=S)cc2)CC1",
-            None,
-        ],
+        ["Aliphaticity", None, metadata_root.joinpath("examples.smi"), False, 7],
+        ["Isoelectric_point", "KFLIYQMECSTMIFGL", None, False, 7],
+        ["Charge", "KFLIYQMECSTMIFGL", None, True, 12],
     ]
 
     with open(metadata_root.joinpath("article.md"), "r") as f:
@@ -78,18 +67,15 @@ if __name__ == "__main__":
 
     demo = gr.Interface(
         fn=main,
-        title="Molecular properties",
+        title="Protein properties",
         inputs=[
-            gr.Dropdown(properties, label="Property", value="qed"),
+            gr.Dropdown(properties, label="Property", value="Instability"),
             gr.Textbox(
-                label="Single SMILES",
-                placeholder="CC(C#C)N(C)C(=O)NC1=CC=C(Cl)C=C1",
-                lines=1,
-            ),
-            gr.File(
-                file_types=[".smi"],
-                label="Multiple SMILES (tab-separated, `.smi` file)",
+                label="Single Protein sequence", placeholder="KFLIYQMECSTMIFGL", lines=1
             ),
+            gr.File(file_types=[".smi"], label="One AAS per line"),
+            gr.Radio(choices=[True, False], label="Amide", value=True),
+            gr.Slider(minimum=0, maximum=14, value=7, label="pH", description="Blub"),
         ],
         outputs=gr.HTML(label="Output"),
         article=article,

model_cards/description.md

@@ -2,6 +2,6 @@
 
 <img align="right" src="https://raw.githubusercontent.com/GT4SD/gt4sd-core/main/docs/_static/gt4sd_logo.png" alt="logo" width="120" >
 
-### Molecular property prediction
+### Protein property prediction
 
-This is the GT4SD web-app for prediction of various molecular properties. For **examples** and **documentation** of the supported properties, please see below. Please note that this API does not expose **all** properties that are supported in GT4SD (a list of the non-supported ones can be found at the bottom).
+This is the GT4SD web-app for prediction of various protein (or peptide) properties. For **examples** and **documentation** of the supported properties, please see below. Please note that this API does not expose **all** properties that are supported in GT4SD (a list of the non-supported ones can be found at the bottom).

model_cards/examples.smi

@@ -1,13 +1,14 @@
-Cc1cc2c(c3oc(CCCC#N)cc13)C(=O)c1c(O)cccc1C2=O
-C=CCN1C(=O)C(=NNC(=S)NC2OC(COC(C)=O)C(OC(C)=O)C(OC(C)=O)C2OC(C)=O)c2ccccc21
-O=C1C(=Cc2ccc(F)cc2)CCOc2c1ccc1ccccc21
-CC(C)CNc1cc(NCC(C)C)nc(NCC(C)C)n1
-CN1CCN(CCCOc2ccc(N3C(=O)C(=Cc4ccc(Oc5ccc([N+](=O)[O-])cc5)cc4)SC3=S)cc2)CC1
-COc1ccc2ccccc2c1C1CC1NC(C)=O
-Cc1ccc(-n2c(=O)[nH]cc(C(=O)Nc3ccc4c(c3)OCCO4)c2=O)cc1
-Cc1ccc(NCc2nnc(SCC(=O)NCCc3ccccc3)n2C)cc1
-CCCNC(=O)c1ccc2c(c1)N=C(C)c1c(C)ccc(C)c1S2
-COc1ccc(Cn2ccn(CC(=O)Nc3cc(C)ccc3C)c(=O)c2=O)cc1
-Cn1nccc1C(=O)NN=Cc1c(O)ccc2ccccc12
-CCOC(=O)Nc1cc(N)c2c(n1)NC(C)C(c1ccccc1)=N2
-Cn1nc(N)c2ncc(C(Cl)(Cl)Cl)nc21
+
+NEGVKAAW	
+FPRPWLHGL	
+HPVGEADYFEY	
+TPGPGVRYPL	
+EEYLKAWTF	
+RMFPNAPYL	
+GPGMKARVL	
+RLRPGGKKK
+VMAPRTLIL
+ARMILMTHF
+FLYNLLTRV
+SLYNTVATL
+ILKEPVHGV

utils.py

@@ -42,6 +42,8 @@ def draw_grid_predict(
             logger.warning(f"Could not draw sequence {seq}")
 
     result = pd.DataFrame({"SMILES": smiles})
+    if domain == "Proteins":
+        result["Seqs"] = sequences
     for i, name in enumerate(property_names):
         result[name] = properties[:, i]
     n_cols = min(3, len(result))