haneulpark commited on
Commit
20dbc54
β€’
1 Parent(s): 6b03f5e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +158 -3
README.md CHANGED
@@ -1,3 +1,158 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ task_categories:
4
+ - text-classification
5
+ language:
6
+ - en
7
+ tags:
8
+ - chemistry
9
+ size_categories:
10
+ - 1K<n<10K
11
+ pretty_name: Blood-Brain Barrier Database
12
+ dataset_summary: >-
13
+ Curation of 50 published resources of categorical and numeric measurements of Blood-Brain Barrier penetration.
14
+ citation: >-
15
+ COPY AND PASTE WHAT YOU GOT FROM THE BIBTEX WEBSITE
16
+
17
+ config_names:
18
+ - B3DB_classification
19
+ - B3DB_classification_extended
20
+ - B3DB_regression
21
+ - B3DB_regression_extended
22
+ configs:
23
+ - config_name: B3DB_classification
24
+ data_files:
25
+ - split: test
26
+ path: B3DB_classification/test.csv
27
+ - split: train
28
+ path: B3DB_classification/train.csv
29
+ - config_name: B3DB_classification_extended
30
+ data_files:
31
+ - split: test
32
+ path: B3DB_classification_extended/test.csv
33
+ - split: train
34
+ path: B3DB_classification_extended/train.csv
35
+ - config_name: B3DB_regression
36
+ data_files:
37
+ - split: test
38
+ path: B3DB_regression/test.csv
39
+ - split: train
40
+ path: B3DB_regression/train.csv
41
+ - config_name: B3DB_regression_extended
42
+ data_files:
43
+ - split: test
44
+ path: B3DB_regression_extended/test.csv
45
+ - split: train
46
+ path: B3DB_regression_extended/train.csv
47
+ dataset_info:
48
+ - config_name: B3DB_regression_extended
49
+ features:
50
+ - name: "NO."
51
+ dtype: int64
52
+ - name: "compound_name"
53
+ dtype: object
54
+ - name: "IUPAC_name"
55
+ dtype: object
56
+ - name: "SMILES"
57
+ dtype: object
58
+ - name: "CID"
59
+ dtype: object
60
+ - name: "logBB"
61
+ dtype: float64
62
+ - name: "Inchi"
63
+ dtype: object
64
+ - name: "reference"
65
+ dtype: object
66
+ - name: "smiles_result"
67
+ dtype: object
68
+ - name: "group"
69
+ dtype: object
70
+ - name: "comments"
71
+ dtype: float64
72
+ - name: "ClusterNo"
73
+ dtype: int64
74
+ - name: "MolCount"
75
+ dtype: int64
76
+ splits:
77
+ - name: train
78
+ num_bytes: 82808
79
+ num_examples: 795
80
+ - name: test
81
+ num_bytes: 27480
82
+ num_examples: 263
83
+ ---
84
+
85
+ # Blood-Brain Barrier Database
86
+
87
+
88
+ ## Quickstart Usage
89
+
90
+ ### Load a dataset in python
91
+ Each subset can be loaded into python using the Huggingface [datasets](https://huggingface.co/docs/datasets/index) library.
92
+ First, from the command line install the `datasets` library
93
+
94
+ $ pip install datasets
95
+
96
+ then, from within python load the datasets library
97
+
98
+ >>> import datasets
99
+
100
+ and load one of the `B3DB` datasets, e.g.,
101
+
102
+
103
+ and inspecting the loaded dataset
104
+
105
+
106
+ ### Use a dataset to train a model
107
+ One way to use the dataset is through the [MolFlux](https://exscientia.github.io/molflux/) package developed by Exscientia.
108
+ First, from the command line, install `MolFlux` library with `catboost` and `rdkit` support
109
+
110
+ pip install 'molflux[catboost,rdkit]'
111
+
112
+ then load, featurize, split, fit, and evaluate the a catboost model
113
+
114
+ import json
115
+ from datasets import load_dataset
116
+ from molflux.datasets import featurise_dataset
117
+ from molflux.features import load_from_dicts as load_representations_from_dicts
118
+ from molflux.splits import load_from_dict as load_split_from_dict
119
+ from molflux.modelzoo import load_from_dict as load_model_from_dict
120
+ from molflux.metrics import load_suite
121
+
122
+ split_dataset = load_dataset('maomlab/B3DB', name = 'B3DB_classification')
123
+
124
+ split_featurised_dataset = featurise_dataset(
125
+ split_dataset,
126
+ column = "SMILES",
127
+ representations = load_representations_from_dicts([{"name": "morgan"}, {"name": "maccs_rdkit"}]))
128
+
129
+ model = load_model_from_dict({
130
+ "name": "cat_boost_classifier",
131
+ "config": {
132
+ "x_features": ['SMILES::morgan', 'SMILES::maccs_rdkit'],
133
+ "y_features": ['BBB+/BBB-']}})
134
+
135
+ model.train(split_featurised_dataset["train"])
136
+ preds = model.predict(split_featurised_dataset["test"])
137
+
138
+ classification_suite = load_suite("classification")
139
+
140
+ scores = classification_suite.compute(
141
+ references=split_featurised_dataset["test"]['BBB+/BBB-'],
142
+ predictions=preds["cat_boost_classifier::BBB+/BBB-"])
143
+
144
+ >>> B3DB_classification = datasets.load_dataset("maomlab/B3DB", name = "B3DB_classification")
145
+ Downloading readme: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 635500.61%/s]
146
+ Downloading data: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 82808/82808 [00:00<00:00, 524655476.79%/s]
147
+ Downloading data: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 27480/27480 [00:00<00:00, 195686712.94%/s]
148
+ Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 795/795 [00:00<00:00, 5218265.54 examples/s]
149
+ Generating train split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 263/263 [00:00<00:00, 1835444.18 examples/s]## About the B3DB
150
+
151
+ ### Features of B3DB
152
+
153
+
154
+ ### Data splits
155
+ The original B3DB dataset does not define splits, so here we have used the `Realistic Split` method
156
+ described in [(Martin et al., 2018)](https://doi.org/10.1021/acs.jcim.7b00166).
157
+
158
+ ###Citation