inab / disease_perception Goto Github PK
View Code? Open in Web Editor NEWPersonalized Comorbidity Profile Generation
License: GNU Affero General Public License v3.0
Personalized Comorbidity Profile Generation
License: GNU Affero General Public License v3.0
Once visualizing relative risk interactions between subgroups of patients, be able to show only the subgroups with at least "x" patients (in the manuscript we are selecting subgroups with at least 4 patients). Once done that, show only those subgroups where there is at least one gene up- and one gene down-regulated in all the patients composing the subgroup, showing only those interactions where at least a gene is involve, this is: in those positive relative risk interactions there should be at least one gene up- or down-regulated in all the patients from both subgroups, while in the case of negative relative risk interactions we are going to show only those interactions where a gene is up-regulated in all the patients from one subgroup and down-regulated in all the patients from the other subgroup, or vice versa. Do the same with drugs. Add the possibility to see the genes and drugs involved in the interactions. Of special interest for the interactions between subgroups from different diseases.
As an example, if you select Alzheimer's disease you will see Alzheimer's disease and all its comorbid diseases. Extract the subgroups of all the selected diseases (Alzheimer's disease and their comorbidity diseases) and show their interactions.
And if possible the interactions between them, as done with the disease and patient-subgroup network
Since last time I was exploring cytoscape ecosystem a new plugin, which allows exporting the graph view to SVG, has arisen:
https://github.com/kinimesi/cytoscape-svg
Explore its features and usability
It could be useful to have a general view of the graph when zooming in
The database loader is assuming that the database is empty, so it tries creating the tables before loading the data. The transactional approach will not leave the service down meanwhile the database is being updated.
In the future we'll be able to do the same to show the drugs involved in the interactions :)
Disease, genes and drugs views are going to need an infrastructure and a GUI to search nodes by name
Currently, it is not possible to go back to the previous view, so a history management infrastructure, and a GUI interface, are needed
As there are going to be lots of new GUI elements, it should be possible to have a way to hide/show all of them.
It should be possible to search by genes and drugs, in the same way diseases are searched. It may require creating more than one destination view for the same search.
Patient subgroups interactions are interesting with this kind of search.
It would be really important that a legend indicating what people are looking at is displayed somewhere. I can imagine and assume what I'm looking at but users might miss the meaning of colors and/or shapes.
Add a button like the one "Hide intra-disease edges" to show/hide positive or negative relative risk interactions (both in the disease and patient-subgroup networks)
There is a jQuery library which allows submitting path change calls to Google Analytics
https://shamasis.net/projects/ga/
As path changes in Disease PERCEPTION are associated to searches and view changes, this library could help tracking most common kind of searches
Currently, when a graph view is saved as PNG, it uses the screen's DPI, so it is too fuzzy when it is zoomed with office tools. Add a fix for it
As there are now more than one view, a brief title describing what it is being viewed should be added to the interface
It would be good to have a visual mechanism for users move between different visualization e.g. from a specific-node view to the whole network
When the graph representation is saved, it is currently saving the viewport, and not the entire graph
In a tabular format
When patients are shown , the ranked list of most similar diseases should be shown. That view will require to compute a new table, which will be provided by @jonsv89
Add a drop-down tab that allows the selection of the disease of interest
Choose colours that can be differentiated easily, right now there are cases where blue and red arrows are hardly differentiated.
Choose different types of arrows (dashed, solid...) to make visually easy to detect the interactions that have been previously described by epidemiological networks.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.