Protein-protein interaction networks

Protein-protein interaction networks

1. Protein‐Protein
Interac-on
Networks

Predic-on,
Visualiza-on
and
Analysis

Vijayaraj
Nagarajan
PhD

Computa4onal
Molecular
Biology
Specialist

Bioinforma4cs
and
Computa4onal
Biosciences
Branch

Na4onal
Ins4tute
of
Allergy
and
Infec4ous
Diseases


Office
of
Cyber
Infrastructure
and
Computa4onal
Biology


2. Outline

• Introduc4on
to
Interac4on
Networks

• Basic
components
of
an
interac4on
network

• Types
of
interac4on
networks

• Predic4ng
Protein‐Protein
Interac4on
Networks

• Methods

» Logic
and
concept

• Available
Interac4on
data

• Integrated
protein‐protein
interac4on
databases

• Searching
for
interac4on
data

• Network
Visualiza4on
and
Analysis
tools

• Popular
visualiza4on
tools

• Network
analysis

• Demonstra4on

• Predic4on
of
interac4on
(PPI…?)
network
from
microarray
(ARACNE)

• Visualiza4on
and
analysis
of
the
predicted
network
(Cytoscape)


3. • Nodes

– DNA/RNA/Protein/Metabolite

• Edges

Directed

– Dis4nc4on
between
source
and
target

» Ac4va4on
(direct/indirect)

» Repression
(direct/indirect)

Undirected

– No
dis4nc4on
between
source
and
target

» Co‐expression
(indirect)

» Binding
(direct)

Interac-on
Networks
–
Basic
Components


4. Interac-on
Networks
‐
Basic
Features

• Degree

• Number
of
connec4ons
that
a
node
has

• Distance

• Number
of
connec4ons
between
two

nodes,
in
a
shortest
path

• Path

• A
sequence
of
connec4ons

• Is
there
a
path
(reachability)

• Mean
Shortest
Path
distance
(closeness)

• In
how
many
shortest
paths

(betweenness)


5. • Size
of
a
network
(Number
of
nodes)

• Density
of
a
network
(Propor4on
of
the
connec4ons)

• Mo4fs/Cliques/Clusters/Sub‐networks

Loops
Chains
Parallels
Multi-input Single input
Interac-on
Networks
‐
Basic
Features


6. Types
of
Interac-on
Networks

• DNA‐Protein

» Transcrip4onal
regulatory
networks

» Methyla4on
networks

• RNA‐RNA

» miRNA
regulatory
networks

• RNA‐Protein

» Splicing
regulatory
networks

• Protein‐Protein

» Co‐expression
networks

» Co‐localiza4on
networks

» Co‐evolu4on
networks

» Structure
networks

» Pathway
networks

» Protease
regulatory
networks

» Signal
transduc4on
networks

» Gene
Ontology
networks


7. Single
gene




– Regulators/Co‐regulators

– Upstream/Downstream
elements
in
the
network

– Global
connec4vity/interconnec4vity

– Func4onal
features

– Differen4ally
expressed
subnetworks

– One
gene
–
one
disease
:
bunch
of
genes
–
pathways

– Nextgen
sequencing
data

List of genes
Why
Build/Analyze
Interac-on
Networks
?


8. Outline

• Introduc4on
to
Interac4on
Networks

• Basic
components
of
an
interac4on
network

• Types
of
interac4on
networks

• Predic4ng
Protein‐Protein
Interac4on
Networks

• Methods

» Logic
and
concept

• Available
Interac4on
data

• Integrated
protein‐protein
interac4on
databases

• Searching
for
interac4on
data

• Network
Visualiza4on
and
Analysis
tools

• Popular
visualiza4on
tools

• Network
analysis

• Demonstra4on

• Predic4on
of
interac4on
(PPI…?)
network
from
microarray
(ARACNE)

• Visualiza4on
and
analysis
of
the
predicted
network
(Cytoscape)


9. How
to
Build
Interac-on
Networks
?


• Search/Retrieve
from
knowledge
bases

• Predict
from
genome
sequences

• Predict
from
“omics”
data

• Predict
from
literature

• Integrate
and
analyze


10. Protein Engineering, Vol. 14, No. 9, 609-614, September 2001
Predic-on
from
genome
sequences

• Gene
neighbor
(gene
cluster,
gene
order)

• Gene
fusion
(Rose\a
stone)

• Phylogene4c
profiling

• Co‐evolu4on

• Mirror
tree


11. Predic-on
from
“omics”
data

• Co‐expression
(Correla4on,
Mutual
Informa4on)

• Integrated
(Classifica4on)

• Literature
mining
(Natural
Language
Processing)


12. Outline

• Introduc4on
to
Interac4on
Networks

• Basic
components
of
an
interac4on
network

• Types
of
interac4on
networks

• Predic4ng
Protein‐Protein
Interac4on
Networks

• Methods

» Logic
and
concept

• Available
Interac4on
data

• Integrated
protein‐protein
interac4on
databases

• Searching
for
interac4on
data

• Network
Visualiza4on
and
Analysis
tools

• Popular
visualiza4on
tools

• Network
analysis

• Demonstra4on

• Predic4on
of
interac4on
(PPI…?)
network
from
microarray
(ARACNE)

• Visualiza4on
and
analysis
of
the
predicted
network
(Cytoscape)


13. Available
Interac-on
Data

• MINT

• Molecular
Interac4on
Database

• BIND

• Biomolecular
Interac4on
Network
Database

• DIP

• Database
of
Interac4ng
Proteins

• IntAct

• InterAc4on
Database
at
EBI


14. Integrated
Data
Sources

• Pathway
Commons

• BioGRID

• MiMI
(Michigan
Molecular
Interac4ons)

• STRING
(Search
Tool
for
Retrieval
of
Interac4ng
Genes/Proteins)

• Genes2Network

• VisANT
(Integra4ve
Visual
Analysis
Tool)


• BIOBASE

• IPA
(Ingenuity
Pathway
Analysis)

• MetaCore

Open Source
Proprietary

15. IP3Rs

• Inositol
triphosphate
receptor

• ip3r1

• ip3r2

• ip3r3

» Calcium
ion
channels

» ER

• s4m1
(stromal
interac4on
molecule
1)

» PM

» Calcium
sensor
protein

• orai1
(calcium
release
ac4vated
calcium
modulator)

» PM

» Accessory
protein
downstream
of
s4m1


16. Searching
for
Interac-on
Data

STRING

IPA


25. IPA


32. Outline

• Introduc4on
to
Interac4on
Networks

• Basic
components
of
an
interac4on
network

• Types
of
interac4on
networks

• Predic4ng
Protein‐Protein
Interac4on
Networks

• Methods

» Logic
and
concept

• Available
Interac4on
data

• Integrated
protein‐protein
interac4on
databases

• Searching
for
interac4on
data

• Network
Visualiza4on
and
Analysis
tools

• Popular
visualiza4on
tools

• Network
analysis

• Demonstra4on

• Predic4on
of
interac4on
(PPI…?)
network
from
microarray
(ARACNE)

• Visualiza4on
and
analysis
of
the
predicted
network
(Cytoscape)


33. • Cytoscape

• VisANT
(Integra4ve
Visual
Analysis
Tool)

• Osprey

• BioLayout

Popular
Visualiza-on
Tools


34. VisANT


43. Analysis
Tools

• Phunkee
(Pairing
subgraphs
Using
Network
Environment
Equivalence
–
finds

similar
subnets)

• mfinder
(mo4f
detec4on)

• MAVisto
(Mo4f
Analysis
and
Visualiza4on
tool
–
explora4on
of
mo4fs)

• GraphMatch
(Search
for
similar
sub‐nets)

• NeAT
(Network
Analysis
Tools
–
mo4f
finding,
node
sta4s4cs)

• Cfinder
(Finds
clusters
–
dense
group
of
nodes)

• NetworkBLAST
(Compares
mul4ple
networks
to
infer
complexes
&
paths)


44. Outline

• Introduc4on
to
Interac4on
Networks

• Basic
components
of
an
interac4on
network

• Types
of
interac4on
networks

• Predic4ng
Protein‐Protein
Interac4on
Networks

• Methods

» Logic
and
concept

• Available
Interac4on
data

• Integrated
protein‐protein
interac4on
databases

• Searching
for
interac4on
data

• Network
Visualiza4on
and
Analysis
tools

• Popular
visualiza4on
tools

• Network
analysis

• Demonstra4on

• Predic4on
of
interac4on
(PPI…?)
network
from
microarray
(ARACNE)

• Visualiza4on
and
analysis
of
the
predicted
network
(Cytoscape)


45. PPI
Predic-on
Using
Microarray
Data

• Co‐expression
concept

• Correla4on
Coefficient

» SIMoNE
(Sta4s4cal
Inference
for
Modular
Networks)
‐
R

• Mutual
Informa4on

» Reference
Networks

» ARACNE
(Algorithm
for
Reconstruc4on
of
Accurate
Cellular

Networks)
–
R,
geWorkbench

» CLR
(Context
Likelihood
of
Relatedness)
–
R

» MRNET
(Maximum
Relevance/Minimum
Redundancy)
–
R

» MONET
(Modularized
NETwork
Learning)
‐
Cytoscape

• Bayesian
Network


46. PPI
Predic-on
Using
Microarray
Data

• Melanoma
metastasis

• 83
samples

• Affymetrix

• GEO

• ARACNE

• geWorkbench

• IP3R1,
IP3R2,
IP3R3,
STIM1,
ORAI1


58. GML File

59. Predicted
PPI
Network

• Could
form
a
complex

• Could
be
func4onally
associated

• Could
be
involved
in
a
same
metabolic
pathway

• Could
be
involved
in
a
specific
signal
transduc4on
path

• False
posi4ve


60. Visualiza-on
and
Analysis
Using
Cytoscape

• Cytoscape

• Opensource

• Works
in
Windows,
Linux,
Mac

• One
of
the
first
developed
tools

• Great
number
of
PLUGINS

• Excellent
help
from
community


61. Literature
Based
PPI
Predic-on


Using
Cytoscape


65. Retrieving
PPI
Using
Cytoscape


69. Data
Visualiza-on,
Integra-on,
Analysis

• Import
ARACNE
predicted
network

• Import
VisANT
retrieved
network

• Merge
all
networks

• Mo4f
finding


79. NetCirChro
(Networks
on
Circular
Chromosomes
)
‐
Citrate
cycle

B.
sub'lis
E.
coli

 S.
typhi
 S.
aureus

Gram-negative Bacteria Gram-positive Bacteria

80. [email protected]

Thank
You