programminglife2015 / lifetiles Goto Github PK

Visualization and comparison of DNA strains.

License: BSD 3-Clause "New" or "Revised" License

Java 98.95% CSS 1.05%

lifetiles's Introduction

LifeTiles

LifeTiles is a tool for interactive visualization of DNA sequence graphs to represent the genome architecture of organisms of interest. The application mainly consists of a Graph Viewer and a Tree Viewer but also a Sequence Viewer used for additional overview.

The Graph Viewer shows the contents of the graph in the structure of a tile lane diagram. It is able to visualize multiple DNA strands and the mutations between those DNA strands if a reference is selected.
The Tree Viewer shows the contents of the phylogenetic tree in the structure of a Sunburst diagram. This shows the relation between the different samples used in the program.
The Sequence Viewer shows all sequences names so the user has an overview of the data which is currently viewed.

The LifeTiles application was made as a group project for the Programming Life section of the TI2806 Context Project TU Delft course.

Features

View a single or multiple DNA strands as a graph to research the structure. In a unified form where every sequence segment has the same coordinate system and empty sequence segments are also visualized.
Display a phylogenetic tree based on several DNA strands so that the user can compare them and observe the timeline and the common ancestors of several functionalities.
Make interaction between the Tree Viewer and the Graph Viewer interactive by giving the user the ability to select subsets in the graph or in the tree which are synchronized.
Indicate mutation types through colours based on a reference sequence.
Filter certain types of mutations so that the user can identify specific types of mutations: SNP, insertion, deletion, inversion, translocation and duplication.
Select the level of detail of the viewport of DNA so that the user can see the relevant parts of DNA using semantic zooming.

Installation

The application can be executed by downloading or cloning the source.

To start researching your DNA sequence graph you need two identical named files in the graph format, a file with .node.graph and a file with .edge.graph as extension. For example simple_graph.node.graph and simple_graph.edge.graph. The phylogenetic tree files must be in the Newick file format. Newick files have the .nwk extension.

lifetiles's People

Contributors

Stargazers

Watchers

Forkers

alangerak jorenham programminglife5

lifetiles's Issues

Create phylogenetic tree navigation in Sunburst diagram

Make navigation possible in the tree

Semantic scrolling / mini-map

The horizontal scroll-bar now is the fully zoomed out version of the graph. On that scroll-bar, a rectangle is placed, indicating the area of the graph currently displayed in the graph view. Inspired on the scroll-bar of PyDev (http://i.stack.imgur.com/4cIgU.png).

Refactor package structure into modular mvc

For the sake of modularity, the package structure needs to be refactored. An example of this:

├── nl.tudelft.lifetiles
│   ├── graph
│   │   ├── model
│   │   ├── view
│   │   ├── controller
│   ├── tree
│   │   ├── model
│   │   ├── view
│   │   ├── controller
│   ├── strains
│   │   ├── model
│   │   ├── view
│   │   ├── controller

Visualization axis

Internal sorting of edges on destination starting position

Create multiple DNA strains navigation in TileMap (non-semantic)

Navigate in the TileMap (move left and right) and make basic (non-semantic) zooming possible

Implement filtering in the Sequence Viewer

Rename Data View(er) to Sequence View(er)

Data view(er) is too ambiguous, Sequence viewer is a more clear definition.

Initialize basic GUI: menu and panels

Init JavaFX. Draw basic screen components such as menu and panels. Make Tree visualization panel and Graph visualization panel.

Implement filtering in the Graph Viewer

Calculate relative coordinates of nodes in DNA sequence graph

Loop over the data and use inversed dijkstra, maximum distance.

Sequence Parser Bug

Parser returns wrong data, not sequences but segment coordinates, must be looked at. Add test and solve.

caught 'ClassCastException' warnings

In modena.bss of JavaFX these warnings get thrown. These need to be fixed.

Edge weight calculations

Sequence text in a tile in a graph viewer does not span the tile-width

This makes part of (long) tiles without text. Letter spacing or size needs to be adjusted.

Data viewer

Refactor package structure

The package structure has become a mess again.

nl.tudelft.lifetiles.graph.models.sequence should be merged into nl.tudelft.lifetiles.sequence.
The graph and tile classes should be merged.
Make the plural package names (i.e. models) singular.

Specific DNA strain selection

When the user clicks on a specific tile, the graph view should only display the strains containing the strains "passing through" that tile and update the sunburst diagram accordingly.

Graph view causes out of memory crash on 100_strains test data

Parsing the 100_strains data set causes a stack-overflow. This can possible be resolved by https://github.com/ProgrammingLife5/LifeTiles/issues/22 .

Is SequenceSegment.endVar overhead?

Is SequenceSegment.endVar overhead by SequenceSegment.startVar + SequenceSegment.contentVar.getLength()?

Architecture Design [Draft]

Architecture Design must be delivered 30-4-2015

Remove graph style from code and into StyleSheet

See pull request #58

Add an application icon

Create test graph: Node + Edge files and Graph instance

Parse Node and Edge files

Parse the given Node and Edges files.

Parsing the external metadata sets

Calculate sequences when parsing

Currently SequenceGenerator does this, but it has to traverse the graph. To avoid this, sequences can be generated when parsing directly (and put into the graph object).

Segment bookmarks

Save a location of the sequence inside the graph viewer to the bookmarks so that the user can find that location back easily.

Display errors in the GUI.

Currently the user isn't informed of an error, the program just terminates or silently throws an exception. This isn't desired behaviour.

Implement bucket caching

Buckets contain the references to the DNA strains at some index. Because of the strain subset selection feature, these buckets need to be cached due to performance.

Make ViewController thread-safe

nl.tudelft.lifetiles.core.controller.ViewController is not thread-safe but used in different JavaFX threads.

Make it possible for DefaultGraphParser to parse files with an absolute path.

Currently it is only able to parse files located in the build path, this is undesired.

Display phylogenetic tree in Sunburst diagram

Initialize SunburstJ library and display the standard static parsed Newick file

Insert empty nodes (insertion / deletion) in DNA sequence graph

Insert empty nodes based on relative coordinates.

Product Vision [Draft]

Product Vision must be delivered 30-4-2015

Improve the readme

Indicate mutation types based on empty tiles and relative coordinates (insertion, deletion and SNP)

Create Node and Edge datastructure

Create the Node and Edges datastructure. Find appropriate library and create specific implementation

Refactor controller structure

Currently it uses the singleton ViewController for communication between controllers. By decentralizing the controller communication, this can be refactored into a more MVC adhering structure. It will also improve performance, since ViewController is very slow.