Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -234,7 +234,7 @@ with gr.Blocks(theme=gr.themes.Ocean()) as demo:
|
|
234 |
This is a general-purpose model that generates diverse drug-like molecules without targeting a specific protein. Trained with a general [ChEMBL dataset]((https://drive.google.com/file/d/1oyybQ4oXpzrme_n0kbwc0-CFjvTFSlBG/view?usp=drive_link)
|
235 |
Molecules larger than 45 heavy atoms were excluded.
|
236 |
|
237 |
-
|
238 |
|
239 |
|
240 |
For more details, see our [paper on arXiv](https://arxiv.org/abs/2302.07868).
|
@@ -256,7 +256,7 @@ with gr.Blocks(theme=gr.themes.Ocean()) as demo:
|
|
256 |
discriminator during training.
|
257 |
|
258 |
### Structural Metrics
|
259 |
-
- **Average Length**: Normalized average number of atoms in the generated molecules, normalized by the maximum
|
260 |
- **Mean Atom Type**: Average number of distinct atom types in the generated molecules
|
261 |
- **Internal Diversity**: Diversity within the generated set (higher is more diverse)
|
262 |
|
@@ -302,7 +302,7 @@ with gr.Blocks(theme=gr.themes.Ocean()) as demo:
|
|
302 |
with gr.TabItem("Custom Input SMILES"):
|
303 |
custom_smiles = gr.Textbox(
|
304 |
label="Input SMILES (one per line, maximum 100 molecules)",
|
305 |
-
info="This space runs on a CPU, which may result in slower performance. Generating 100 molecules takes approximately 6 minutes. Therefore, we set a 100-molecule cap.\n
|
306 |
placeholder="C(C(=O)O)N\nCCO\n...",
|
307 |
lines=10
|
308 |
)
|
|
|
234 |
This is a general-purpose model that generates diverse drug-like molecules without targeting a specific protein. Trained with a general [ChEMBL dataset]((https://drive.google.com/file/d/1oyybQ4oXpzrme_n0kbwc0-CFjvTFSlBG/view?usp=drive_link)
|
235 |
Molecules larger than 45 heavy atoms were excluded.
|
236 |
|
237 |
+
- Useful for exploring chemical space, generating diverse scaffolds, and creating molecules with drug-like properties.
|
238 |
|
239 |
|
240 |
For more details, see our [paper on arXiv](https://arxiv.org/abs/2302.07868).
|
|
|
256 |
discriminator during training.
|
257 |
|
258 |
### Structural Metrics
|
259 |
+
- **Average Length**: Normalized average number of atoms in the generated molecules, normalized by the maximum number of atoms (e.g., 45 for AKT1/NoTarget, 38 for CDK2)
|
260 |
- **Mean Atom Type**: Average number of distinct atom types in the generated molecules
|
261 |
- **Internal Diversity**: Diversity within the generated set (higher is more diverse)
|
262 |
|
|
|
302 |
with gr.TabItem("Custom Input SMILES"):
|
303 |
custom_smiles = gr.Textbox(
|
304 |
label="Input SMILES (one per line, maximum 100 molecules)",
|
305 |
+
info="This space runs on a CPU, which may result in slower performance. Generating 100 molecules takes approximately 6 minutes. Therefore, we set a 100-molecule cap.\n\n Molecules larger than allowed maximum length (45 for AKT1/NoTarget and 38 for CDK2) and allowed atom types are going to be filtered.\n\n Novelty (Inference) metric will be calculated using these input smiles.",
|
306 |
placeholder="C(C(=O)O)N\nCCO\n...",
|
307 |
lines=10
|
308 |
)
|