freitasleandro / pdac-ann Goto Github PK

The script runs a deep learning model (PDAC-ANN) to classify samples in normal or pancreatic ductal adenocarcinoma (PDAC) using gene expression information. The genes for the prediction are AHNAK Nucleoprotein 2 (AHNAK2),Laminin Subunit Beta 3 (LAMB3), Laminin Subunit Gamma 2 (LAMC2),Keratin 19 (KRT19), and S100 Calcium Binding Protein P (S100P).



This initiative was built by 
Leandro Martins de Freitas - [email protected], 
Palloma Porto Almeida - [email protected]



The computer must have python 3.6 or higher to run the script and the packages:
pandas 
pandas, numpy, sklearn, keras, tensorflow, and argparse. 


Help 

python PDAC.py -h 

help instructions to run the script.

 
Running the script 

python script.py -i input_file.csv -o output_file.csv -t 0.78 

-i input file

-o output file 

-t threshold value to be considered PDAC sample  (0.78 by default)



Test run

python PDAC.py -i test/input_file.csv -o test/output_file.csv

.

The output file contains four columns. 

First column: samples name
. 
Second column: the probability of the sample to be normal. 

Third column: the likelihood of the sample to be PDAC

Fourth column: the classification of the sample based on the threshold value.


The input file must be a csv file separated by ;
 Fist line must be  the header
. Example of an input csv file:


Samples;AHNAK2;LAMB3;LAMC2;KRT19;S100P

Sample1;0.84;0.78;0.58;0.55;0.50

Sample2;0.90;0.87;0.80;0.521;0.588

Sample3;0.923;0.908;0.86;0.53;0.68

Sample4;0.503;0.208;0.303;0.119;0.277

Sample5;0.59;0.181;0.35;0.41;0.56