README file for Python source code supporting the paper "MNBDR: A Module Network Based Method for Drug Repositioning".
- Python = 3.X
- Sklearn
- Numpy
Numpy and Sklearn can be installed using pip.
PPI_Clusetr --dense clusters in the SRTING PPI network, geneId = "Entrez".
Example:
5693 56929 10199 55313
522 6390 9551 6391
51542 9827 25782 636
29883 485 23019 10605
PPI_Clusetr_network --the edge of dense clusters.
Example:
0 1 1 1
1 0 1 1
0 1 0 0
1 0 1 1
Drug_gene --Since Disease Data and drug data often come from different databases, gene's order and number
don't usually match. Therefore, we provide a list of drug genes to facilitate the extraction of drug expression information.
Example:
5693
56929
10199
55313
29883
Disease --tab delimited, Expression data of disease, the column are Gene_Entrez, Normal group, Disease group,respective (In the test cases, we provide data for breast cancer patients). Example:
1 0.10158704 0.161963237
29974 0.01379371 0.013093553
2 242.8467729 115.1052494
144568 0.197790261 2.093597589
53947 6.994426346 6.398611842
AVE_DMSO_MCF7_ctl_vehicle_ALL h --Expression data of Cell Lines after drug stimulation (Control group) (In the test cases, we provide data for breast cancer Cell Lines). Example:
0.07991 0.06652 -0.06172 0.04316
AVE_MCF7_Min --Expression data of Cell Lines after drug stimulation (treatment group) (In the test cases, we provide data for breast cancer Cell Lines). Example:
lamotrigine -0.31464 0.2192 -0.0083 -0.05618
glimepiride 0.8858 -0.26607 -0.13603 -0.16573
RUN
In this test case, we used breast cancer as an example and provided a small data set of drug responses.
If you run "test.py" , you can get a scores for breast cancer and drugs.
Example:
Disease Drug Score
BRCA pentobarbital 46.37012
BRCA SN-38 45.97034
BRCA epirubicin 44.385
BRCA daunorubicin 40.46897
Custom dataset
You can also apply our method to your Dataset by replacing the data in DATA.
For example, you can provide disease data in our format, then you can get scores for new diseases and drugs.
Similarly, you can also replace drug data or module network data.
If you have any questions,contact the maintainer of this package: HG_Chen [email protected]