Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Introduction to Pathway and Network Analysis

This is a talk I will give at the 2010 Canadian Bioinformatics Workshop on Interpreting Gene Lists

Lincoln Stein

on 5 July 2011

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Introduction to Pathway and Network Analysis

Introduction Pathway Databases Pathway and Network Analysis of Omics Data Lincoln Stein
Ontario Institute for Cancer Research Relationship Maps Gene Function Prediction Hypothesis Generation from Gene Lists Patient/Sample Classification Conclusions Overview Why pathway analyses?
What are pathways & networks?
Where can you get pathway & network information?
Using pathways & networks to predict gene function,
...to generate hypotheses from gene lists
...to classify patients & samples. Relationship Maps Networks Pathways Network Databases Functions A,B,C Functions B,C,Z Functions Z,F Function F Functions D,E,F Function H Functions A,I Functions M 1 2 3 4 5 6 7 8 5 8 4 7 2 6 1 3 Pathway A Pathway B Outlier? Pathway/Network Clustering Getting Pathway/Network Data Using Pathway/Network Data Pathway Databases Process diagrams
Biochemical model
Typically hand-curated from literature
Really useful for extracting biological knowledge.
But concept of "pathway" is very fuzzy KEGG
NCI/Nature PID
Pathway Commons Boehringer Mannheim Metabolic Pathways Chart 5.2 million genes; 1024 species; 100,000 pathways.
Manually curated pathway diagrams; but most are projecte across species.
Heavy reliance on EC numbers for ID mapping.
Free for academic use
Pathway/gene lookup; Colorizing service. Beautiful hand-drawn pathway diagrams
"120,000 genes from multiple species"; 154 pathways.
Community-annotation service, mostly used by drug companies to overlay drug ads.
No underlying database; can't automate to use to interpret gene lists Desktop application
Provides pathway search & colorization

Works with WIKIPathways, an online version
19 species: 178 human pathways and 4279 genes www.reactome.org www.genmapp.org www.biocarta.com www.genome.jp/kegg pid.nci.nih.gov/index.shtml www.ingenuity.com Popular $$$$ commercial application.
Very polished user interface
Combination of curation, integration and macine learning, but algorithms unpublished.
Content statistics unavailable.
Features: Identify pathways containing lists of genes; extract and build custom pathways/networks.
Integration with pharmaceutical information. Popular Pathway Databases www.reactome.org www.pathwaycommons.org/pc/ BioPax Format Pathways from multiple sources in uniform format
Basic summary & search tools.
Web service export to Cytoscape.
Planned integration & unification of pathways, but not yet. Networks Built on top of bimolecular interactions
Physical interaction
Sharing of GO terms
Literature co-mention
Adjacency in pathways Networks can be built automatically or via curation. Pathway Colorizing Services Popular Sources of Curated Interaction Networks Some More Network Concepts www.pathguide.org
Sloan Kettering BioGRID -- www.thebiogrid.org; Curated interactions from literature; 529,000 genes, 167,000 interactions.
InTact -- www.ebi.ac.uk/intact; Curated interactions from literature; 60,000 genes, 203,000 interactions.
MINT -- mint.bio.uniroma2.it; Curated interactions from literature; 31,000 genes; 83,000 interactions. Automated Interactions via Text Mining Computationally extract gene relationships from text, usually PubMed abstracts
Useful if network is not in a database
Literature search tool
BUT not perfect
Problems recognizing gene names
Natural language processing is difficult
Agilent Literature Search Cytoscape plugin
iHOP (www.ihop-net.org/UniPub/iHOP/) Automated Interactions from High-Throughput Experiments Yeast 2 hybrid protein interactions
Protein complex pulldowns + mass spec
Genetic screens, e.g. synthetic lethals
BUT not perfect
Y2H interactions have taken proteins out of natural context; high number of false positives.
Protein complex pulldowns plagued by "sticky" proteins such as actin.
Genetic screens highly sensitive to genetic background (due to "network effects"). Integrative Approaches Combine multiple sources of evidence to increase accuracy.
Simple example: "Party hubs": Y2H + gene coexpression & subcellular compartment data.
Complex example: Combine Y2H, proteomics, curated data, literature mined data, coexpression and GO data. Relationship Maps Different pathway databases have different names for similar processes Are they the same, overlapping, or distinct?
"Heart disease", "cardiovascular disease", "vascular development"
Relationship maps integrate gene information with phenotype information to discover these relationships. KEGG Reactome Reactome Beta Active Subnetwork Extraction and Annotation Example Integrated Network: Reactome Functional Interaction Network The Paradigm What makes a Cancer Stem Cell? Cluster Overview Clustering Stats Reactome Beta Beta features:
Google-map style reaction diagrams
Overlays with protein & chemical interactions Before you start... Normalization
Background adjustment
Quality control

Use statistics that will reduce the noise in your experiment.
Make sure to use IDs that are compatible with the softare. Gene Set Enrichment Maps Birds of a Feather Principle What to do if many of the genes in your list have no known function?
Look at genes in their neighbourhood.
Guilt by association. What have we Learned? Pathway databases vary considerably in content, curation policies, and data model.
Pathway databases provide excellent qualitative information, but are limited in coverage and are not good for quantitative hypothesis testing. Pathways Interaction Networks Networks constructed from integration of many data sources provide better coverage, and do not necessarily sacrifice accuracy.
When combined with pathway information, networks can provide clues to mechanism.
Network analysis can be used to aid biological hypothesis generation, to predict gene function, and to classify patients & samples. Hand-curated pathways in human.
Rigorous curation standards -- every reaction trackable back to primary literature
Automatically-projected pathways to other species.
22 species; 1045 pathways; 4605 unique proteins.
Exportable in multiple formats.
Features: Find pathways containing genes of interest; Calculate gene overrepresentation in pathways; Find corresponding pathways in other species. Curated signaling pathways; human only
110 pathways curated by Nature
329 pathways imported from Reactome & BioCarta
Very easy to use gene list interpretation function. NCI/Nature PID 325 pathway/network dbs and counting!
Full transcript