Sub Cellular Localization
From Protein Prediction 2 Winter Semester 2014
Revision as of 13:30, 6 January 2015 by Ppwikiuser (talk | contribs)
Sub Cellular Localization in Cell
Contents
Objective
"To visualize biological cells and highlight by a user selected sub-cellular compartments in a way that they stand out from the un-selected ones"
GUI mockups
Case 1
One protein, one localization:
Case 2
Multiple proteins, multiple localizations:
Requirements
- User experience
- Interactivity
- Easy identification of the number of proteins in cell compartment
- Good visualization of data
- Functionality
- Parsing the user input “txt” file
- Mapping the file content into visualization
- Features
- Good color scheme to highlight :
- Number of proteins in a compartment
- Score of each protein (confidence)
- Good color scheme to highlight :
Application design
- Expected technical difficulties
- Wrong file input format
- Correct identification of localization
- Libraries
DHTML, JavaScript Draw Line, Circle, Ellipse (Oval), Polyline, Polygon, RectangleImage map generator- D3 Data-Driven Documents - http://d3js.org/
Data
- Remarks about input format
- Text file must contain proteins of only one type of cell
- Text file format :
- Input data file should be either a ".txt" or ".csv" file.
- File’s first line should contain cell type (i.e: eukaryota, archea, bacteria).
- File's second line should contain Score (i.e: 0-100). Note: Minimum score should be zero.
- File's third line should contain columns description (i.e: Protein Id, Score, Localization).
- User’s file can have more than 3 columns but additional columns will not be executed.
Roadmap
Week number | Completed tasks | Status | |
---|---|---|---|
Week 1 |
|
||
Week 2 |
|
||
Week 3 |
|
||
Week 4 |
|
||
Week 5 |
|
|
Implementation
What we did
- First, all the cell compartments were identified by the paths that were marked using the GIMP image editor.
- Using the information from the user's input file, the number of proteins present in each compartment were determined.
- Each compartment was highlighted using a localization color scale, which was obtained by converting the number of proteins present in each localization into a percentage and matched to a color.
- A table to display all the proteins present in each cell compartment and their scores.
- Upon 'mouseover' over a compartment, a tooltip was displayed, which shows the proteins present in that cell compartment, and the score (confidence) of the protein.
- The proteins displayed in the tooltip were made clickable so that the cell image could be updated to reflect the scores of the protein in all the cell compartments.
- Finally, the cell compartments were highlighted using a score color scale, which was obtained by mapping the score of a protein in each cell compartment to a color.
Screenshots
- Marking cell compartments using GIMP image editor
- Parsing input file
- Highlight cell's compartments following number of proteins and score
- Showing popup, in case one protein and more than one protein in input file
- Output
GitHub Link
The link to the GitHub account is : GitHub Sub-cellular localization in cell