Last modified by emacasali on 2023/01/11 16:05
From version 11.1
edited by emacasali
on 2023/01/11 16:05
Change comment: There is no comment for this version
To version 7.1
edited by emacasali
on 2023/01/11 11:07
Change comment: Uploaded new attachment "image-20230111110707-1.png", version {1}

Summary

Details

Page properties
Content
... ... @@ -20,38 +20,11 @@
20 20  
21 21  Here we report on a fast and solid workflow which starts from our DF-matrix method to analyse how the protein globally behaves in the presence of a ligand. Machine Learning (ML) trains a Convolutional Neural Network (CNN) model directly on the pixel images of DF: train is preformed using a known ligand and the different behaviour of the protein is evaluated in the presence and in absence of it.
22 22  
23 -[[image:image-20230111110707-1.png]]
24 -
25 25  With the so trained model further predictions can be performed using different ligands.
26 26  
27 -= How to use the script =
25 += Who has access? =
28 28  
29 -• __Requisites__
30 -
31 - - Python 3.0 (or newer version)
32 -
33 - - Numpy
34 -
35 - - Tensorflow
36 -
37 - - Pandas
38 -
39 - - Sklearn
40 -
41 - - cv2
42 -
43 - - Matplotlib
44 -
45 -• __Usage__
46 -
47 -- CNN-training-script.py constitutes the main code of the tool: here different models of CNN can be customized, by changing also activation function and classification mode. In its final part it operates also a test using unseen data and save the trained model as a .h5 file.
48 -
49 -The first operation that is required by the user regards the very initial prepartion of the DF-images from the DF-matrices [see the following link for the DF preparation [[https:~~/~~/wiki.ebrains.eu/bin/view/Collabs/distance-fluctuations>>https://wiki.ebrains.eu/bin/view/Collabs/distance-fluctuations]]]. This can be done using the gnuplot.in file and the exectute-DF.sh file, which renames the .png accroding with the nanoseconds used to extract the image.
50 -
51 -The images required for the training of the model has to be selected and classified by the user between the two states of interest and by using the random-selector files to divide them between test, trainig and validation datasets. Here we usually preformed a random separation between test (20%), train (64%) and validation (16%) sets using the last 200ns of the equilibrated dynamics.
52 -
53 -
54 -- CNN-external-data-test.py is a script which aims to use the trained model (.h5 file) and test it on data belonging to different proteins from the ones used during the build of the model.
27 +Describe the audience of this collab.
55 55  )))
56 56  
57 57  
Collaboratory.Apps.Collab.Code.CollabClass[0]
Public
... ... @@ -1,1 +1,1 @@
1 -Yes
1 +No
XWiki.XWikiRights[3]
Allow/Deny
... ... @@ -1,1 +1,0 @@
1 -Allow
Levels
... ... @@ -1,1 +1,0 @@
1 -view
Users
... ... @@ -1,1 +1,0 @@
1 -XWiki.XWikiGuest
XWiki.XWikiRights[4]
Allow/Deny
... ... @@ -1,1 +1,0 @@
1 -Allow
Groups
... ... @@ -1,1 +1,0 @@
1 -XWiki.XWikiAllGroup
Levels
... ... @@ -1,1 +1,0 @@
1 -view