kexinhuang commited on
Commit
7e115d6
1 Parent(s): af38ee7

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +51 -0
README.md ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ metrics:
5
+ - accuracy
6
+ - AUC ROC
7
+ - precision
8
+ - recall
9
+ tags:
10
+ - biology
11
+ - chemistry
12
+ library_name: tdc
13
+ license: bsd-2-clause
14
+ ---
15
+
16
+ ## Dataset description
17
+
18
+ As a membrane separating circulating blood and brain extracellular fluid, the blood-brain barrier (BBB) is the protection layer that blocks most foreign drugs. Thus the ability of a drug to penetrate the barrier to deliver to the site of action forms a crucial challenge in development of drugs for central nervous system.
19
+
20
+ ## Task description
21
+
22
+ Binary classification. Given a drug SMILES string, predict the activity of BBB.
23
+
24
+ ## Dataset statistics
25
+
26
+ Total: 1,975 drugs
27
+
28
+ ## Dataset split:
29
+ Random split on 70% training, 10% validation, and 20% testing
30
+ To load the dataset in TDC, type
31
+
32
+
33
+ ```python
34
+ from tdc.single_pred import ADME
35
+ data = ADME(name = 'BBB_Martins')
36
+ ```
37
+ ## Model description
38
+ CNN is applying Convolutional Neural Network on SMILES string fingerprint. Model is tuned with 100 runs using Ax platform.
39
+ To load the pre-trained model, type
40
+
41
+ ```python
42
+ from tdc import tdc_hf_interface
43
+ tdc_hf = tdc_hf_interface("BBB_Martins-CNN")
44
+ # load deeppurpose model from this repo
45
+ dp_model = tdc_hf_herg.load_deeppurpose('./data')
46
+ tdc_hf.predict_deeppurpose(dp_model, ['YOUR SMILES STRING'])
47
+ ```
48
+
49
+ ## References:
50
+
51
+ [1] Martins, Ines Filipa, et al. “A Bayesian approach to in silico blood-brain barrier penetration modeling.” Journal of chemical information and modeling 52.6 (2012): 1686-1697.