You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<figcaption>Fig.1 - Overview of how DeLeMa Detect was built</figcaption>
31
31
</figure>
32
32
33
-
### ```app.py```
33
+
### ```app.py```
34
34
35
35
The main python script uses the Flask micro web framework to create a local host where the model is deployed. It contains auxiliary functions ```model_predict()```, ```index()``` and ```upload()```
36
36
37
-
1.```model_predict``` calls the `model.h5` file stored in ```./model/```. We created many models and tested each one's accuracy, size and processing power. Since the size of a few models were greater than 50MB (a soft limit setup by Github) we have uploaded them on [Google Drive](https://drive.google.com/drive/folders/11ULc4FWlB3VScfZIR4y3o8KJgljHZPFe?usp=sharing). Based on the model one wants, it can be downloaded placed in ```./model/```
37
+
1.```model_predict``` calls the `model.h5` file stored in ```./model/```. We created many models and tested each one's accuracy, size and processing power. Since the size of a few models were greater than 50MB (a soft limit setup by Github) we have uploaded them on [Google Drive](https://drive.google.com/drive/folders/11ULc4FWlB3VScfZIR4y3o8KJgljHZPFe?usp=sharing). Based on the model one wants, it can be downloaded and placed in ```./model/```
38
38
39
39
Although the dimensions of the uploaded image can be anything, each model takes a particular input image size which is controlled and preprocessed in ```app.py```, namely :
40
40
| Sr no | Model |```target_size```|
@@ -117,21 +117,13 @@ On clicking the `Predict` Button, the Image is sent to the Model. The Model Prep
117
117
118
118
<br><br>
119
119
120
-
The result for an Image that has been Identified as Uninfected.
<figcaption>Results for a Parasitized ir infected Blood Smear Image</figcaption>
123
+
<figcaption>Results for a Parasitized/Infected Blood Smear Image</figcaption>
129
124
</figure>
130
125
<br><br>
131
126
132
-
The result for an Image that has been Identified as Uninfected
133
-
<br><br>
134
-
135
127
### On Mobile
136
128
137
129
The [Web App](https://delema-detect-igem-iiserpune.herokuapp.com/) works on mobile too. The procedure is the same. Upload the image and tap on Predict.
@@ -171,7 +163,7 @@ Activate the Virtual Environment by running the following.
171
163
172
164
<br><br>
173
165
174
-
# Sample BLood Smear Images
166
+
##Sample Blood Smear Images
175
167
176
168
We have added upto 20 blood smear images for testing purposes at ```./uploads/```. There are two directories called ```Parasitized``` and ```Uninfected```, each containing upto 10 images. These can be downloaded and tested by running the application locally or on our Heroku platform.
Copy file name to clipboardExpand all lines: Peptides_against_Malaria/README.md
+16-14Lines changed: 16 additions & 14 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,6 +1,6 @@
1
1
-[Overview](#overview)
2
2
-[Analysis of Molecular Dynamics Simulations](#analysis-of-molecular-dynamics-simulations)
3
-
-[How does the distance between the centroid of the Protein and peptide evolve over time?](#how-does-the-distance-between-the-centroid-of-the-protein-and-peptide-evolve-over-time)
3
+
-[How does the distance between the centroid of the protein and peptide evolve over time?](#how-does-the-distance-between-the-centroid-of-the-protein-and-peptide-evolve-over-time)
-[Hydrogen Bond Analysis](#hydrogen-bond-analysis)
6
6
-[Computing the Hydrogen Bond Profile](#computing-the-hydrogen-bond-profile)
@@ -45,18 +45,18 @@ To perform these calculations, we wrote the script ```PDB_centroid_analyser```.
45
45
</figure>
46
46
47
47
Here :
48
-
- Atom no --> group (1)
49
-
- Chain name --> group (3)
50
-
- X co-ordinates --> group (5) + group (6)
51
-
- Y co-ordinates --> group (8) + group (9)
52
-
- Z co-ordinates --> group (11) + group (12)
53
-
- Atom --> group(14)
48
+
- Atom no --> group (1)
49
+
- Chain name --> group (3)
50
+
- X co-ordinates --> group (5) + group (6)
51
+
- Y co-ordinates --> group (8) + group (9)
52
+
- Z co-ordinates --> group (11) + group (12)
53
+
- Atom --> group(14)
54
54
55
55
After reading these lines, the function ```get_centroid()``` obtains the centroid (x, y, z) of a given chain. We perform calculations in numpy arrays using the standard centroid formula:
A wrapper function, ```main()``` calls other multiple functions to determine the distance between the centroid of two chains using `euclidean_distance()`. This is done for all the .pdb files submitted in the text file. *Special care* has to be taken to ensure that all .pdb files are in the official prescribed format, otherwise, the regex syntax will not be able to identify the chain, atom or element. Finally, the script writes a .csv file in the same directory which includes the following fields:
59
+
A wrapper function, ```main()``` calls other multiple functions to determine the distance between the centroid of two chains using `euclidean_distance()`. This is done for all the .pdb files submitted in the text file. *Special care* has to be taken to ensure that all .pdb files are in the official prescribed format; otherwise, the regex syntax will not be able to identify the chain, atom or element. Finally, the script writes a .csv file in the same directory which includes the following fields:
60
60
61
61
1. Filename
62
62
2. Centroid of Chain1 "{chain-name given}"
@@ -65,13 +65,15 @@ A wrapper function, ```main()``` calls other multiple functions to determine the
65
65
66
66
--> This .csv file serves as a good checkpoint for performing further analysis.
67
67
68
-
The script can be run on the command line/terminal and also imported into other scripts if particular functions seem interesting. The command-line version takes two arguments:
68
+
The script can be run on the command line/terminal and also imported into other scripts if some particular function seems interesting. The command-line version takes two arguments:
The other functions included in PDB_centroid_analyser are :
77
79
1.`get_centorid()` --> returns the centroid of a particular chain from a .pdb file
@@ -83,7 +85,7 @@ The other functions included in PDB_centroid_analyser are :
83
85
84
86
To understand effective binding, we analyze the number of intermolecular Hydrogen bonds between the protein and the peptide. This entails the analysis of hbonds over each snapshot of the MD simulation to determine which residues and specifically which atom in these residues of the peptide inhibitor form hydrogen bonds with the protein. We also aim to determine the relative abundance of Hydrogen bonds formed over the entire simulation period.
85
87
86
-
To perform these calculations, I wrote a script that functions on build over [UCSF Chimera](https://www.cgl.ucsf.edu/chimera/).
88
+
To perform these calculations, we wrote a script that functions and builds over [UCSF Chimera](https://www.cgl.ucsf.edu/chimera/).
87
89
88
90
89
91
### Computing the Hydrogen Bond Profile
@@ -100,13 +102,13 @@ The script takes two arguments :
100
102
1. A text file containing .pdb files that are to be analyzed (must be in the same directory as the script and .pdb files)
101
103
2. A .cmd file containing chimera commands
102
104
103
-
**Note**: It is assumed that the chimera executable file is stored at ```~/chimera/bin/chimera```. If not the path to the executable has to be specified on Line 11.
105
+
**Note**: It is assumed that the chimera executable file is stored at ```~/chimera/bin/chimera```. If not the path to the executable has to be specified on Line 11.
104
106
105
107
The commands to be run in Chimera is submitted in a .cmd file, with each line containing the command based on chimera documentation.
106
108
107
109
##### Working
108
110
109
-
The script loops over each line of the text line (basically a file name), reads that particular file and processes it based on the instructions given in the .cmd file. After processing, the last 8 lines of the output (required information is stored here) is written to an external file. We then use the Linux stream editor ('sed') to replace the filename to be read next in the .cmd file, and a similar process continues until all files in the .txt file are processed.
111
+
The script loops over each line of the text file (basically a file name), reads that particular file and processes it based on the instructions given in the .cmd file. After processing, the last 8 lines of the output (required information is stored here) is written to an external file. We then use the Linux stream editor ('sed') to replace the filename to be read next in the .cmd file, and a similar process continues until all files in the .txt file are processed.
110
112
111
113
112
114
#### ```PDB_Hbond_analyser.py ```
@@ -116,7 +118,7 @@ The script takes two arguments :
116
118
2. The name of the output .csv file where the output data is written
117
119
118
120
##### Working
119
-
The python script contains a host of preprocessing functions that read data from the output files of ```Hbond-analyzer.sh, ```i.e. files ending with ```hbond_info.txt```. The script processes each line in the .txt file using grep expressions and returns a dataframe containing information on the Hydrogen bonds (Donor atom, Acceptor atom, Donor-Hydrogen, Distance D-A, Distance D-H-A, No of Hydrogen bonds). This is used for further Data Analysis and Interpretation.
121
+
The python script contains a host of preprocessing functions that read data from the output files of ```Hbond-analyzer.sh```, i.e. files ending with ```hbond_info.txt```. The script processes each line in the .txt file using grep expressions and returns a dataframe containing information on the Hydrogen bonds (Donor atom, Acceptor atom, Donor-Hydrogen, Distance D-A, Distance D-H-A, No of Hydrogen bonds). This is used for further Data Analysis and Interpretation.
0 commit comments