Pilot to work with dynamic visualization by using python
Switch branches/tags
Nothing to show
Clone or download
Permalink
Failed to load latest commit information.
demos Added demos Sep 14, 2017
docs - Jul 12, 2015
pynteractive Added api keys to external libraries for managing maps Sep 15, 2017
webfiles Added api keys to external libraries for managing maps Sep 15, 2017
README.md - Jul 15, 2015
setup.py - Nov 30, 2015

README.md

Pynteractive: Interactive plots in python

Pynterative is a python library aimed to create interative visualizations provided by several HTML5 JavaScript libraries like vis.js, d3.js and many more.

Its main feature is the real time communication python<->webbrowser via websockets, that allows us realtime interaction from python to the browser and from the browser to python. It means that the code you write in python affects the visualization in real time and mouse/keyboard events happening in the web browser can be connected to python callbacks creating a bidirectional communication.

We wanted to create a very simple API that, although reduces flexibility, it is extremely easy to learn and start creating web GUI's.

Pynteractive IS NOT a framework to develop:

  • Web applications: the graphic interfaces are supposed to be local. That is the main idea, not dealing with web at all, being able to use cool JS frameworks without messing with any web technology.
  • Production software: it is intended to help developers/data scientists plotting data very quickly. You've got data, you need to plot it, you need to play with it, that is what pynteractive is aimed for, do not expect more.

Pynteractive is in constant development, all feedback/suggestions/bugreports are very wellcome. It was motivated by the need to use HTML5 plots in biomedical research, (bioinformatics specifically), however different charts and plots are being introduced as they are requested/needed.

Demos

De Bruijn Graph Gene assembly using Pynteractive Mixing maps and phylogenetics

Current features

  • Networks: graphs and digraphs provided by the vis.js framework
  • Trees: Using different engines (d3.js, vis.js)
  • Maps: provided by leaflet (openstreet maps)
  • Phylogenetic trees + Annotation
  • Charts: using NVD3.js

## TODO

  • More phylogenetic tree layouts
  • Integrating Circos and making it interactive
  • Extra plots extracted from D3.js
  • Interactive scatter plots with point selection

Compatibility

It requires an HTML5 browser and currently it has been tested in Google Chome, Safari and Mozilla Firefox.

Installation

There are two ways of installing pynteractive:

  • Cloning the project
$ git clone https://github.com/coelias/Pynteractive.git
Cloning into 'Pynteractive'...
remote: Counting objects: 658, done.
remote: Compressing objects: 100% (98/98), done.
remote: Total 658 (delta 50), reused 0 (delta 0), pack-reused 560
Receiving objects: 100% (658/658), 136.28 KiB | 0 bytes/s, done.
Resolving deltas: 100% (381/381), done.
Checking connectivity... done.

$ cd Pynteractive
$ python setup.py install  ### (as root)
$ pip install pynteractive

Documentation

Pynteractive API

Examples

####Graph example (click to watch video)

Graph example

from pynteractive import *
a=Graph(directed=True)
a.view()

# adding nodes
a.addNode()
a.addNode(10)
a.addNode(11,color='red')
a.addNode('john',shape='star',color='green')

# adding edges
a.addEdge(1,10)
a.addEdge(1,'john',style='dash-line')
a.addEdge(1,11,width='4')
a.addEdge(10,11,width='2',label='hello world')
a.delNode('john')

# creating a circular graph
a.addNode('a',shape='dot',radius=30)
a.addNode('b',shape='dot',radius=20,color='red')
a.addNode('c',shape='dot',radius=10,color='green')
a.addNode('d',shape='dot',radius=20,color='blue')
a.addNode('e',shape='dot',radius=30,color='#FFFF00')
a.addNode('f',shape='dot',radius=35,color='#bbb')
a.addEdge('a','b')
a.addEdge('b','c',style='arrow-center')
a.addEdge('c','d')
a.addEdge('d','e',style='arrow-center')
a.addEdge('e','f')
a.addEdge('f','a',style='arrow-center')

# cleaning the view
a.clear()

# Generating a randon graph (40 nodes, 50 edges
a.random(40,50)

# Logging into the log area
a.log('Hello world')
a.log('<h1>header</h1>')
a.log('<code>code</code>')

# Defining callbacks to perform actions after events in the GUI
def Multiply_by2(node):
    a.log(str( int(node)*2)+"<br>")
a.setDoubleClick(Multiply_by2)

def sumall(nodes):
    a.log(str( sum([int(i) for i in nodes]))+"<br>")
a.addAction('Add node values',sumall)

####Chart example (click to watch video)

Chart example

from pynteractive import *
b=Chart()
b.view()
b.addSeries('Squirrels killed by humans',
    [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16],
    [200,240,300,313,317,380,360,320,503,460,510,600,550,500,460,490]) 
b.addSeries('People killed by squirrels',
    [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16],
    [1,20,40,60,100,110,105,150,100,80,150,180,182,210,300,491])
b.addSeries('Sharks killed by squirrels',
    [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16],
    [10,20,25,40,12,33,22,47,90,70,80,46,30,6,18,25])

####Map example Map example

from pynteractive import *
m=Map()
m.view()
m.addNode("ID1",place="london")
m.addNode("ID2",place="ox3 8pp")
m.addEdge("ID1","ID2")
m.addNode("ID3",place="08030",country="es")
m.addNode('ID4',place='paris',radius=30,color='blue')
m.addEdge('ID3','ID4',width=5,color='green')

####Phylogenetic tree example Phylo example

from pynteractive import *
import random

a=PhyloTree()
a.setData('(109:0.5593,((53:1.8461,((185:1.5264,((((198:0.7133,(14:1.7812,(70:0.9566,((178:1.4691,((45:0.7471,(194:0.6693,((44:0.5577,((13:0.8088,((191:0.5682,((61:0.5975,(88:3.2495,(9:1.0308,(96:0.8686,(((115:0.8278,(123:0.8377,(1:0.5639,179:1.0903):0.5946):1.4461):0.7937,(16:1.1170,154:0.6546):0.6107):0.5630,(23:0.6146,(64:0.6340,81:0.6373):1.7269):1.4345):0.9564):0.8100):0.6957):0.5944):1.0511,((56:1.9748,160:0.5840):0.9831,((66:0.6673,(42:0.7604,121:0.9784):0.8448):0.7313,(7:0.6708,162:0.7519):0.8166):1.0403):1.1924):0.9697):0.7165,(117:0.7042,(184:1.4130,(50:0.9100,(101:0.8581,(37:0.6301,200:0.8667):0.6302):1.1244):1.4690):1.9430):0.8444):0.5968):0.6728,(30:1.1661,33:1.1261):0.8310):2.2959):0.7608,(190:0.6007,196:0.9444):0.6424):0.6270):0.8282):0.9730,(24:1.4291,118:0.8477):1.4567):0.6991):0.8921,((133:0.7706,(28:0.5830,(134:0.6095,175:0.6930):0.7241):0.8666):0.9387,(111:1.0094,(136:0.5893,138:1.1745):1.5885):0.7924):0.5738):0.8773):0.5963):0.6160):1.0136,((65:1.1500,97:1.0232):1.3978,((57:0.5738,187:1.6658):0.6071,((19:0.7025,(174:1.3017,(104:1.2256,(114:0.8844,132:0.9376):0.6582):0.7183):1.1257):0.5755,(95:0.9269,((142:0.9090,(39:0.6623,(72:0.6360,164:0.9143):0.7035):4.5670):0.6282,((25:0.9429,(146:0.6349,(137:2.3101,((86:0.5907,(151:0.6346,171:0.6288):1.2030):1.2696,(32:0.5801,161:0.5600):1.3039):1.5206):0.5787):0.6616):0.7571,(43:2.0269,(131:0.6637,(12:1.7647,(156:1.1010,(172:0.8216,(153:0.8397,(((11:1.1326,188:0.7363):1.2086,(35:0.6256,(189:0.5714,((159:2.1437,(120:0.7276,(79:0.8632,98:0.8285):0.6579):0.8806):0.7212,(63:0.8597,85:0.7613):1.1212):0.5833):0.9273):0.6866):0.8127,(8:1.4761,(176:0.6388,(173:0.6236,(38:0.7009,177:0.6063):0.7085):1.4928):1.0702):0.5824):1.1679):1.5157):0.6207):1.2862):0.9861):0.9956):0.8808):0.7404):0.7407):1.4742):2.3546):0.6926):1.6898):1.0044,(68:3.0620,181:0.8448):0.7053):0.6960,(36:0.9977,76:0.6317):0.8789):0.6850):0.6060,(130:1.8879,(27:0.9512,(163:0.7594,197:1.3480):2.2115):0.7601):1.8072):1.2875):0.8267,(193:1.5216,(4:0.6166,(112:0.7144,147:0.9374):0.7245):0.6220):0.6376):1.6023,((82:2.8243,((92:1.1603,148:1.2875):0.6096,(18:2.2320,110:0.9872):1.3756):0.6613):0.8081,(3:0.6139,(94:0.8184,(80:2.5720,(149:1.2856,((59:0.7261,((139:1.2208,(62:0.6156,87:0.5686):0.8219):0.5661,(31:0.6006,55:0.7081):0.8803):0.5895):4.4796,(29:1.0162,((150:0.8578,(6:0.7863,106:0.5613):0.6144):0.9523,(51:2.2633,(199:0.8779,(126:0.5832,(99:1.2345,(((47:0.5868,52:0.5764):0.5797,((89:2.4215,((34:1.3367,168:1.2975):0.9865,(40:1.4390,(119:0.6238,(186:0.6974,(69:0.5586,((107:0.6821,(58:1.3881,116:0.9332):1.2708):0.9753,((10:1.1451,54:1.2720):1.1650,((141:1.0443,(74:1.8388,(21:1.1221,83:0.8512):1.5698):0.8246):0.5762,(129:0.6327,(75:0.8794,(48:0.8473,125:1.0579):0.8496):0.8923):0.6014):1.3489):1.8732):0.8591):0.5691):1.2968):0.9153):1.0208):0.6514):0.6898,(166:1.4354,(22:0.7493,(2:1.1837,(183:0.9209,(5:1.1304,(93:1.1079,144:0.6024):0.5676):1.7392):1.0896):0.6673):0.5778):0.5953):0.7470):0.5946,(((26:1.2338,165:0.7093):1.2264,((195:1.5322,(158:1.4151,(170:0.6407,(67:0.9098,(49:0.9280,143:0.7676):1.3389):0.5612):0.6873):0.9481):0.7016,(90:1.0637,(84:0.8715,(105:0.7043,157:0.6207):0.6683):1.8207):0.6640):1.0868):0.9417,(100:1.0056,(127:1.5781,((17:0.9746,(77:0.8027,180:1.4619):0.6277):1.1663,(140:1.7778,(169:0.9060,(128:0.6329,(145:0.5933,((122:0.6385,(155:0.9812,((71:0.8869,103:0.8851):1.1801,(167:0.8196,(102:0.7040,(113:0.9985,(78:1.1891,(15:1.1411,182:0.5817):0.6458):0.5856):0.8109):0.5834):0.7143):0.6831):0.7380):0.9687,((108:0.7798,152:2.0513):0.8877,(46:1.3488,((73:0.6343,(60:1.0697,(20:0.7480,192:0.9474):0.9843):1.4541):0.8802,(41:0.8659,(135:0.6945,(91:1.0905,124:0.6244):0.8901):0.9333):1.0731):0.8375):0.6176):0.9982):0.9024):0.7677):1.7023):1.5721):2.7324):0.5982):1.0601):0.6174):0.9339):0.6215):1.2051):0.7775):1.9509):0.6680):0.7644):0.6658):2.0353):1.0737):1.4884):0.8140):0.6216):0.7000);')
a.view()
    
t1=a.addTrack("first track",'magenta')
t2=a.addGradientTrack("Second track",'blue',0,100)

for i in range(60):
    a.addTrackFeature(t1,str(random.randint(1,199)))

for i in range(1,199):
    a.addTrackFeature(t2,str(i),value=random.randint(0,100))

b1=a.addBar('first bar','lightblue',1,50)
b2=a.addBar('second bar','mediumpurple',1,50)

for i in range(1,201):
    a.addTrackBar(b1,str(i),random.randint(1,51))
    a.addTrackBar(b2,str(i),random.randint(1,51))

a.markClade(['144','93'],'red')
a.markClade(['116','107','58',],'blue')
a.markClade(['39','72','164',],'green')

raw_input()

Real World Examples

De Bruijn Graph Gene assembly using Pynteractive

Being able to debug graphically gene assembly while coding was very helpful. This tool motivated pynteractive development.

De Bruijn Graph Gene assembly using Pynteractive

Mixing maps and phylogenetics

In our work it's very important to relate samples geographically in order to detect potential outbreaks and study bacterial transmission. (Dataset shown is randomly generated) Source code

Mixing maps and phylogenetics

Authors:

  • Carlos del Ojo
  • Oriol Mazariegos