Introduction to Biological Network
Analysis and Visualization with
Cytoscape
Keiichiro Ono
Cytoscape Core Developer Team
UC, San Diego Trey Ideker Lab / National Resource for Network Biology
5/10/2016 The Scripps Research Institute
Lecture 1: Basics

Keiichiro Ono
Cytoscape Core Developer
since 2005
@UCSD Trey Ideker Lab
Area of Interest:
Biological Data Integration &
Visualization

Agenda
• Lecture 1 (Today):
Introduction to Biological Network
Analysis and Visualization
• What is the benefits of biological
network analysis and visualization?
• Introduction to Cytoscape
• Preview of Lecture 2: cyREST
• Lecture 2:
Reproducible Analysis & Visualization
• Introduction to Jupyter Notebook
• Create a reproducible network
visualization workflows with Python

All documents, data, and code are available here:
https://github.com/idekerlab/tsri-lecture

Why Network Analysis?

Networks?

EP300
PPARG
SMARCD3
STMN1
SMARCA4
OPTN
ATP6V1C1
PSMD1
HTT
PRNP
HNRNPUL1
CCDC88A
CLU
HSP90AB1
SMARCD3
MAP4K4
MIF4GD
USP11
MARCH6TUBB
EDF1 CHD8
Protein-Protein Interactions

Human-Curated Pathways
KEGG Pathway (TCA Cycle) visualized by Cytoscape KGMLReader

Interactomes
Human Interactome data from BioGRID visualized with Cytoscape

Social Networks

Network extracted from Panama Papers data set

Node 1 - Edge Type - Node 2

Protein 1 - Y2H - Protein 2

Networks vs Pathways

Networks Pathways

Networks Pathways
Collection of binary interactions Human-curated / detailed
Large Scale Small Scale
Generated from omics-data Constructed from literature

Networks / Pathways
= Graphs

Benefits of Network Analysis

Benefits of Network Analysis
- You have list of N genes from your
screening
- Now you want to know:
- Relationships among those
genes
- Functions
- etc.
Screening 1
PPARG
TCF7L2
RETN
IRS1
HNF1A
HNF4A
KCNJ11
GCK
LIPC
PTPN1
ABCC8
ENPP1
HNF1B

ENSG00000167780
ENSG00000255974
EBI-9992455
ENSG00000241119 ANP32B
EBI-10055098
EBI-10055672
EBI-9871829
NR3C1
EBI-10096648
STK16
SNCG
EBI-10034984
EBI-9973444
RAD50
EBI-9980932
ENSG00000070019
EBI-9871836
ENSG00000105398
EBI-9992700
ENSG00000207778
ENSG00000143819
EBI-9980935
ENSG00000125730
ENSG00000180432
ENSG00000197249
EXT2
EBI-5333164
EBI-10051521
OPTN
EBI-10050213 PPARGC1A
EBI-10039585
MAPK8
HNF4A EDF1
SFPQ
ENSG00000110245
PPARGEP300
SMARCD3
PRNP
HNRNPUL1
ATP6V1C1
HTT
EBI-10050241
EBI-10039564
ENSG00000118137
PABPC4
EBI-10050232
EBI-10051518
EBI-9871840
EBI-10096653
ENSG00000095596
MARCH6
EBI-3946155
EBI-5333185
BCAR3
IRS1
PIK3CA
LRRK2
Irs1
PIK3R1
PIK3R1
Grb2
Phb
Grin1
H1f0
Rps3
Rpn2
Ipo9
Scml2
Cand1
Eif2ak2
Ipo8
Ppfibp1
Sdpr
Poldip3
Tenm3
Rars
Ppp2r1a
Vim
Rfc3
Hsp90ab1
R q c d 1
Rplp0
Hnrnpu
Irs1
Snd1
Hspa8
Ung
Tp53bp2
GRB2
TP53BP2
YWHAB
IGF1R
DDR1
SMAD3
EBI-1108795
Ctnnd1
Ran
Ywhag
Rab6b
Ybx1
Epha2
Grin2b
Flot2
Aff3
Ptpn1
PELI1
EIF2AK2
INSR
YWHAE
NCOR1
Pik3r1
N U M B
PRKCQ
TP53BP2
PIK3R3
RAP2A
Mink1
CHEBI:39112
Ywhab
CHEBI:39079
CDK5
STMN1
PPARGUSP11
CHEBI:45783
CHEBI:18319
HSP90AB1
MAP4K4
MIF4GD
CHEBI:64310
CHEBI:49840
TUBB
CLU
CHD8
CCDC88A
SMARCA4
SMARCD3
PSMD1
MAP4K2
MINK1
Map3k1
MAP3K1
PKM
MAP3K11
CHEBI:49375
GOLIM4
ARHGAP39
GOSR1
BCOR
PTRF
PCNP
KIDINS220
TMEM216
PHB
ABCD3
1C
ATP6V0D1
FLOT2
RNF213
SMARCD2
LTN1
PIAS1
Sumo1
ATP6V0A1
WRAP53
EVC2
ACTN1
GALNT2
dhbF
B4E2V5
TCTN3
GHR
EGFR
ATP6V1B2
CDH2
TMEM17
RMDN3
PHB2 PTPN1
Cdh2
TCTN2
PDGFRB
ATF2
CHEBI:17440
mviM1
MET
MVP
MSN
MREG
COL5A1
FLOT1
ASS1
CHEBI:17283
uvrB
FOXM1
tyrB
UBE2I
glnA
Ctnnb1
PSMA3
ENSG00000138795
CTNNB1
RUNX3
PSEN1
JUP
DAXX
TCF7L2
hmwP2
hemL2
vgrG7
fadB
FBLN1
XRCC6
XRCC5
PARP1
Q99IB8-PRO_0000045599
pagA
Psen1
ENSG00000168646
ENSG00000065361
p p s C
Q99IB8-PRO_0000045596
Q8CLD5
TGIF1
YPO2975
p g i
pyrE
ENSG00000110092
ENSG00000136997

Benefits of Network Analysis
- You can see the relationships among the group
of biological entities
- Find drag targets
- Overrepresented functions AND their
connections

Gene List to Network to Biological Insight
- How?
- You need to search, integrate, and visualize
multiple data sources
This is what you will learn in this lecture

What is Cytoscape?

An Open Source Platform for Biological Network Data
Integration, Analysis and Visualization
Cytoscape

Cytoscape 3.4.0 (Latest Release)

Cytoscape
- Open Source (LGPL)
- Free for both commercial and academic use
- Developed and maintained by universities,
companies, and research institutions
- UC, San Diego
- University of Toronto
- UC, San Francisco
- ISB
- And collaborators world-wide

Cytoscape
- De-facto standard software in biological network
research community
- Large User and Developer Community
- Expandable by Apps
- This is why Cytoscape is a Platform, not a
simple desktop application

C. Elegans Interactome from BioGRID Database
?

Biological Networks
- Tell us anything by themselves
- Just a big hairball…

Module 1
Module 2

In other words…

Module 1
Need a tool to extract
meaningful biological modules

Basic Use Case

Networks
Public Interaction Databases

List of
Genes

Other Data

Network Data Analysis
Analysis
Graph Analysis
NetworkX
igraph
Cytoscape
Python
Pandas
NumPy
SciPy
Excel
Visualization
Desktop
Gephi
Cytoscape
matplotlib
Web
Cytoscape.js
sigma.js
d3
NDV3
d3.chart
Google Charts
Data Storage
Graph
Neo4j
GraphX
Document
MongoDB
Relational
MySQL
IPython
3rd Party Apps
NetworkAnalyzer

Network Data Analysis
Analysis
Visualization
Desktop
Gephi
Cytoscape
matplotlib
Web
Cytoscape.js
sigma.js
d3
NDV3
d3.chart
Google Charts
Data Storage

Network Data Analysis
Analysis
Graph Analysis
NetworkX
igraph
Cytoscape
Python
Pandas
NumPy
SciPy
Excel
Visualization
IPython
3rd Party Apps
NetworkAnalyzer

Three Basic Steps for Data Visualization
with Cytoscape

<?xml version="1.0" encoding="UTF-8"?>
<graphml xmlns="http://graphml.graphdrawing.org/xmlns"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns
http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd">
<!-- Created by igraph -->
<key id="degree" for="node" attr.name="degree" attr.type="double"/>
<key id="betweenness" for="node" attr.name="betweenness"
attr.type="double"/>
<graph id="G" edgedefault="directed">
<node id="n0">
<data key="degree">79</data>
<data key="betweenness">0</data>
</node>
<node id="n1">
<data key="degree">9</data>
<data key="betweenness">167</data>
</node>
<node id="n2">
<data key="degree">18</data>
<data key="betweenness">75</data>
</node>
<node id="n3">
<data key="degree">8</data>
<data key="betweenness">12</data>
</node>
<node id="n4">
<data key="degree">26</data>
<data key="betweenness">210</data>
</node>
<node id="n5">
<data key="degree">29</data>
<data key="betweenness">320</data>
</node>
Data Integration

Analysis

Visualization

Drawing Biological Networks
VS

Drawing Tools
You need to specify
color of each node,
width of each edge,
shape of nodes, etc.

There is one huge difference
between Cytoscape and
Illustrator…

In Cytoscape,
Your Data Controls View

Creating Visualizations in Cytoscape
Name Type
BRCA1 gene
MAP2K1 gene
C05981 compound
• Mapping from Type to Node Shape
• Mapping from Type to Node Color
C05981
BRCA1
MAP2K1
Creating mappings from data points
to Visual Properties

Network Data
Annotated
Networks
Attributes
Analyzed Data

Apps

Cytoscape Apps
- Extension programs to
add new features to
Cytoscape
- formerly called Plugins
- Large App developer/
user community
- This is why Cytoscape
is so successful in life
science community!

Example
Apps

ClueGO
Creates and visualizes a functionally grouped network of terms/pathways

ReactomeFIPlugIn
Explore Reactome pathways and search for diseases related pathways and network patterns
using the Reactome functional interaction network

KEGGScape
KEGG Pathway Importer for Cytoscape

clusterMaker2
Multi-algorithm clustering app for Cytoscape

cyREST (Now Part of The Core!)
RESTful API for Cytoscape

cyREST
RESTful API for Cytoscape

APPS.CYTOSCAPE.ORG

Overview of App Ecosystem
A travel guide to Cytoscape plugins
Rintaro Saito, Michael E Smoot, Keiichiro Ono, Johannes Ruscheinski, Peng-
Liang Wang, Samad Lotia, Alexander R Pico, Gary D Bader, Trey Ideker (2012)
Nature Methods 9 (11) p. 1069-1076

Tips for Learning
New Tools

Choose a Right Tool

Choose a Right Tool
Analysis VisualizationData Preparation

Data Visualization Tools
http://selection.datavisualization.ch/

Data Visualization Tools
http://selection.datavisualization.ch/

Data Visualization Tools
http://selection.datavisualization.ch/

Tools
• In some cases, you can get exact same result using different tools
• Example: Data preparation (cleansing / cleansing)
• But if you choose right tools, you can do it 100x faster than
others.
• ex: Re-formatting complex data sets
• Excel vs Python Script
• Some recommendations:
• R/Bioconductor, Python/Pandas, Git/GitHub/Gist

Learning Tools = Saving Your Time

Hands-on:
Introduction to
Network Visualization
with Cytoscape

Sample Data Files:
http://cl.ly/VTJs
Same data is available in lecture1/dataset

Data Visualization

- Goal: Help others to
understand your data
- Emphasize what you
want to tell
- Use color, shape,
size of objects
effectively!
- Excellent resource for
data visualization
- Tamara Munzner’s Web Site:
http://www.cs.ubc.ca/~tmm/
Data Visualization

Today’s Goal

Story:
I want to show gene expression changes over time
as a network diagram

YPL201C YPL211W YML007WYPL131W YOR327CYDR171W YCL067C
YGL208WYER074WYBL050W YLR134WYPL149W
YDR050C YMR311CYGL134W YBR112CYKL101W YNL199C
YPL222W
YLR264W
YNL098C
YLL028W
YOR039W YNL135C YPR041WYDR174W YIL074C YKL028W
YIL162W
YNL189W
YOR212W
YPR080W
YPR145W
YLL019C
YLR284CYPL031C YFR037CYML074C YPL240CYPR048W YBR274W YBR050C YML032C
YJR022WYBR248C YDR382W YER081WYIR009W YDR244W YOL016C
YER103W YGR058WYLR256WYAL003W YOR355WYIL061C YER111C YMR309C
YPL248CYBR019CYLR362W YGL035CYPR167C YML123C
YBL026WYNL091W YOR178C YIL113WYLR321C YML064C YMR117C YDL194WYNR007C
YOL058WYBR045CYER065CYNL167C YGL097WYHR071W YDL078C YDL081CYDR354W
YER145C YGR136WYDR311W YPR119WYER112W YLR214W
YER143W YBR043CYKL204W
YGR019WYEL041W YER133W
YBR118WYAL038W YDR167WYMR058WYER079W YMR291W
YKL012W YDL113CYDR299W YDL075W YDL236WYLR377C YNL145W
YNL236W
YOL156W
YGL013C
YHR171W
YMR021C YFL038C
YER090WYPR062W YAR007C YNL307CYML024WYDR335W YLR075W
YNL050CYGR046W YAL040CYLR191W YMR138WYIL045W YHR005C
YKL211CYLR452C YPL075WYML051W YOL123WYHR198C YMR300C
YJR060W YMR043WYPR124WYLR081W YLR319CYKL074C YKL001C
YDR100W
YDR395W
YDR009W
YDR309C
YPR102C
YAL030W
YHR084W
YLR345W
YBR170C
YJL089WYFL026W YBR018C YGL115W YDL215CYGR009C YOL120C
YFL017C YDR429C
YIL052C YGL073W
YGR108WYPR035W
YJL190CYOL086CYBL005WYKR026C
YBR155W
YOR264W YKL109W
YOR167C
YDR070CYEL015W
YIL133C
YGL166WYHR030CYGL008C
YMR146C YBR160W
YBR020W YBR190WYDR323CYLR197W YFR014CYKL161C
YML054C YKR099WYLR340WYGL106W YBR093CYCL040W
YLR044C
YCR086WYDL130W
YJL203W
YEL009CYBR135W
YOR361C YGR085C
YNL216W
YBR109C
YER124C
YJL157C YDR461WYNL154CYLR117C YKR097W YIL069CYMR186W YJR109CYIL015W
YER040W YGR074WYER052C YIL160CYOR290C YLR249W
YGL153WYOR215CYGR254W YLR432WYCR084CYOR089C YOR303W
YGL161C YLR293CYDL030WYNL036W YHR135CYER179W YDR277CYDR184C YML114C YFL039CYER054C YER110CYLR109W YLR116WYNL214W YBL069W
YHR141CYER116CYJL219W YDL023C YGL202WYER062C YMR183CYFR034CYGL122C
YIL105C YDL088CYPR010C YJR048W YIL070C YEL039CYDR412WYMR108W
YOR204W YMR255W YLR175W YHR115CYNL164C YJL013C YDL063C YNL117W
YIL143CYOR315W YDR146CYLR310CYGR014WYBR217W
YJL036W
YNL116W
YOR120W
YDR032C
YPR113W
YLR153C
YGR048W
YGR203W
YNL113WYOR202W
YNR050C
YCL030C
YJL159W
YHR053CYPR110C?YLR258W YBL079W
YNL069C YNL311CYDR142C YGL044CYMR044W
What is Great
Visualization…?

Design is complicated,
because humans are complicated.
Design is a process to avoid bad designs.
Mike Bostock (New York Times Visualization Team. Creator of D3.js)

It is hard to generalize the design process, but we
can avoid pitfalls by following some basic rules.

Every pixel should
carry information.
Edward Tufte

Avoid Data Overload
• Mapping too many attributes makes your visualization
awful!
• It is hard to see the overall trend of your data sets if too
many channels are used in a image

“Great Artists Steal…”

MUD
HAP4
GC
HA
GAL1
GAL7
GAL80
GAL3
GAL11
GAL4
GAL2
SIP4
FBP1
GAL10
SWI5
SUC2
MIG1
ADH1
PGK1
CDC19
GCR1
CBF1
ENO1
ENO2
MCK1
NCE103
SSL2
TFB1
YNL091W
TRP4
ARG1
GCN4
SKO1
HIS3
ADE4 ILV2
RPS17A
BAS1
HIS7
RPS24B
MSL1
HIS4
PDC5
PHO84
PHO4
YIL105C
MET16
RPL11B
RPS8B
RPL11A
RPL31A
PHO13
PDC1
SXM1
RPL34B
RPL16B
ATC1
CAR1
FCY1
ICL1SRP1
TPI1
RPL18B
RPL25
PHO5
RPS24A
RPL18A
DMC1
RAP1
RPL16A
HSP42

MUD
HAP4
GC
HA
GAL1
GAL7
GAL80
GAL3
GAL11
GAL4
GAL2
SIP4
FBP1
GAL10
SWI5
SUC2
MIG1
ADH1
PGK1
CDC19
GCR1
CBF1
ENO1
ENO2
MCK1
NCE103
SSL2
TFB1
YNL091W
TRP4
ARG1
GCN4
SKO1
HIS3
ADE4 ILV2
RPS17A
BAS1
HIS7
RPS24B
MSL1
HIS4
PDC5
PHO84
PHO4
YIL105C
MET16
RPL11B
RPS8B
RPL11A
RPL31A
PHO13
PDC1
SXM1
RPL34B
RPL16B
ATC1
CAR1
FCY1
ICL1SRP1
TPI1
RPL18B
RPL25
PHO5
RPS24A
RPL18A
DMC1
RAP1
RPL16A
HSP42
Map gene expression
values to color
Avoid using more colors in other
components (edge/label)
If necessary, map other data
into non-overlapping visual properties
(edge score to width)

Part 1: Session File and Basic Navigation

Cytoscape 3.4 Desktop
Toolbar
Network Panel
Bird’s Eve View
Table Browser
Network Views

Local Column
Table Tabs
List Data
(Values in [ ])
Shared Column

Session File
- Snapshot of your workspace
- Networks
- Tables
- Visual Styles
- System Properties

Open a Session
- Click folder icon
- Or, File → Open

Exercise 1: Loading a session

Navigation
- Pan: Drag
- Zoom
- IN: Mouse Wheel UP
- OUT: Mouse Wheel DOWN
- Selection: Shift + Drag
- Fit to Window
- Selected region
- Entire network

First Neighbor of Nodes
CTR+6

Create New Sub-Network From Selection
CTR+N

- Grid View

- Detached View

Part 2: Data Import

Network Data Formats
- SIF
- GML
- XGMML
- GraphML
- BioPAX
- PSI-MI
- SBML
- KGML (KEGG)
- Excel
- Text Table
- CSV
- Tab

NCBI Gene ID 672
On Chromosome 17
GO Terms
DNA Repair
Cell Cycle
DNA Binding
Ensemble ID
ENSG00000012048
BRCA1

Data Tables for Cytoscape
- Example:
- Numeric
- Gene expression profiles
- Network statistics calculated in other
applications, such as R
- Confidence scores for edges
- Text (or categorical)
- GO annotation for genes
- List of genes related to disease X
- Targets for FDA approved drugs
- Genes on KEGG Pathway Y
- Clusters / group / community calculated
in external programs
- …

Your Data Sets
- Anything saved as a table can be
loaded into Cytoscape
- Excel
- Tab Delimited Document
- CSV
- As long as proper mapping key is
available, Cytoscape can map them to
your networks.

Mapping Key in
the Network
Mapping Key
in the Table

Exercise 2: Loading network and tables

Part 3: Visualization

Layouts

Force-Directed + Edge Bundling

Stacked-Node Layout + Default Edge Bend

Circular + Edge Bend

Automatic Layout
- Choose proper algorithm
- Tree-like data - Hierarchical Layout
- Scale-Free Network - Force-directed
- Circular process - Circular Layout
- Tweak parameters if necessary

Manual Layout
- Tweak result from
automatic layout
- Scale
- Align
- Rotate

Exercise 3: Apply layouts

Visual Style
- Collection of mappings from
Attributes to Visual
Properties

Visual Styles
- Defaults + Mappings
- Expression values to node color
- Gene function to node shape
- Interaction detection method to edge line
type
- Confidence score to edge width

Core Idea: Data Controls The View

Data Controls The View
• Photoshop / Illustrator
• You control the pixels and objects on the display
• Data Visualization Tools (including Cytoscape)
• Data points are mapped to visual properties
• Color
• Size

Data Controls The View

Expression Values
To Node Colors

Discrete Mapping Editor
Continuous Mapping Editor

Exercise 4: Create New Visual Style

Preview of Lecture 2:
Reproducible Workflow with Jupyter Notebook and cyREST

Cline, Melissa S., et al. "Integration of
biological networks and gene expression
data using Cytoscape." Nature protocols
2.10 (2007): 2366-2382.

Cline, Melissa S., et al. "Integration of
biological networks and gene
expression data using Cytoscape."
Nature protocols 2.10 (2007):
2366-2382.

Results

Sharing Results
😐

Sharing Results and Process
😃

Point & Click
Operation is Easy, but
not Reproducible…

Goal:
Reproducible Science

Goal: Reproducible Science
REST

Tools You Need

REST
- GitHub
- For source code sharing
- IPython (Jupyter) Notebook
- Your electronic lab notebook
- cyREST
- RESTful API module for Cytoscape

-
- Two Google Groups
- cytoscape-
discuss@googlegroups.com
- cytoscape-
helpdesk@googlegroups.com
- ANY question is OK!
Getting Help

Further Readings

Further Readings
• My presentation slides
• http://www.slideshare.net/keiono
• This deck of will be uploaded today

Further Readings 1
- Introduction to Network Biology
- Deciphering Protein–Protein Interactions. Part I. Experimental
Techniques and Databases
Shoemaker BA, Panchenko AR (2007) Deciphering Protein–Protein
Interactions. Part I. Experimental Techniques and Databases. PLoS
Comput Biol 3(3): e42.doi:10.1371/journal.pcbi.0030042
- Deciphering Protein–Protein Interactions. Part II. Computational
Methods to Predict Protein and Domain Interaction Partners
Shoemaker BA, Panchenko AR (2007) Deciphering Protein–Protein
Interactions. Part II. Computational Methods to Predict Protein and
Domain Interaction Partners. PLoS Comput Biol 3(4): e43. doi:10.1371/
journal.pcbi.0030043

Further Readings 2
- Overview of Cytoscape Apps (Plugins)
- A travel guide to Cytoscape plugins
Rintaro Saito, Michael E Smoot, Keiichiro Ono, Johannes Ruscheinski,
Peng-Liang Wang, Samad Lotia, Alexander R Pico, Gary D Bader, Trey
Ideker (2012) Nature Methods 9 (11) p. 1069-1076
- Sample Protocol (based on 2.x)
− Integration of biological networks and gene expression data using
Cytoscape
Cline, et al. Nature Protocols, 2, 2366-2382 (2007).

Further Readings 3
- Cytoscape Tutorial Booklet:
Analysis and Visualization of Biological Networks with Cytoscape
- http://www.rbvi.ucsf.edu/Outreach/Workshops/ISMBTutorial.pdf

2016 Keiichiro Ono
kono@ucsd.edu