A Systematic approach to the LargeScale Analysis of GenotypePhenotype correlations

1
A Systematic approach to the Large-Scale Analysis
of Genotype-Phenotype correlations
Supervisee Paul Fisher Supervisors Dr. Robert
Stevens Prof. Andy Brass Advisor Dr. Steve
Pettifer
2
PhD Hypothesis

• Utilising the capabilities of workflows and a
  pathway-driven approach, we are able to provide a
  more
• - systematic
• - efficient
• - scalable
• - un-biased
• - unambiguous methodology
• the benefit will be that new biology results
  will be derived, increasing community knowledge
  of genotype and phenotype interactions.

3
Summary of 1st / (early) 2nd Year

• Identified issues with current methods
• Developed own methodologies
• Constructed methods using workflows
• Tested methods through use cases
• Obtained Biological results
• Published results in Biology and Bioinformatics
  journals

4
QTL mapping study
Microarray gene expression study
Statistical analysis
Identify genes in QTL regions
Identify differentially expressed genes
Genomic Resource
Annotate genes with biological pathways
Annotate genes with biological pathways
Pathway Resource
Select common biological pathways
Hypothesis generation and verification
Wet Lab
Literature
5
Scientific Literature A Knowledge resource
Pathway information
Literature database
Implicit interactions
Phenotype information
6
What Does the Text Hold?
Protein Info
Related Proteins
Protein-Protein Interactions
Pathways
Biological processes
7
What Next ?
Biological processes
Generate a Profile for Pathway / Phenotype
Apoptosis Cell Death Stress response ..
MeSH vocabulary
8
Two Profiles Phenotype and Pathway
Find common terms

• Phenotype Terms
• Apoptosis
• Cholesterol
• Diabetes
• Jak-Stat
• Ribosome
• Cell Adhesion Molecules

• Pathway terms
• Apoptosis
• Cholesterol
• Diabetes
• Cell Death
• JNK pathway

High chance pathway is linked to phenotype
9
And So
Which pathways do I investigate in the Wet Lab?
High Priority
Prioritise those that have links in the literature
Phenotype Terms
Pathway Terms
Lower Priority Pathways
10
The Prototype Workflows
Find common terms
Get terms from abstracts
Get abstracts for pathways / phenotype
11
Future work (3rd Year)

• Provide a corpus of specific workflow fragments
• Text mining
• Quality Assurance (adding a level of
  significance to returned results)
• Evaluate text mining workflows using test cases
  including
• Analysis of data obtained from Trypanosomiasis
  project
• Analysis of data for studying the infection of
  Mice by Trichuris muris
• Publish findings
• Biology journals for biological results
• Bioinformatics journals for text mining workflow
  methods
• Publish workflows within the myExperiment
  project for sharing and re-use
• Compile thesis on findings over 3 years

12
(No Transcript)
13
Questions
14
Genotype to Phenotype
15
Current Methods
Genotype
Phenotype
200
?
What processes to investigate?
16
Phenotype
Pathway A
CHR
literature
Pathway linked to phenotype high priority
QTL
Gene A
Pathway B
Gene B
literature
Pathway not linked to phenotype medium priority
Gene C
Pathway C
literature
Genotype
Pathway not linked to QTL low priority
17
Issues with current approaches

• Scale of analysis task is huge
• User bias and premature filtering
• Hypothesis-Driven approach to data analysis
• Constant flux of data - problems with
  re-analysis of data
• Implicit methodologies (hyper-linking through web
  pages)
• Error proliferation from any of the listed issues
• Solution Automate (through workflows)

18
Phenotype
Genotype
200
?
Metabolic pathways
Phenotypic response investigated using microarray
in form of expressed genes or evidence provided
through QTL mapping
Genes captured in microarray experiment and
present in QTL (Quantitative Trait Loci ) region
Microarray QTL
19
QTL mapping study
Microarray gene expression study
Statistical analysis
Identify genes in QTL regions
Identify differentially expressed genes
Genomic Resource
Annotate genes with biological pathways
Annotate genes with biological pathways
Pathway Resource
Workflow methods
Select common biological pathways
Hypothesis generation and verification
Wet Lab
Literature
Manual methods
20
Evaluation - Results

• Identified many issues with current investigative
  techniques
• A strong candidate gene was found for
  Trypanosomiasis resistance
• Daxx gene not found using manual investigation
  methods
• The gene was identified from analysis of
  biological pathway information
• Sequencing of the Daxx gene in Wet Lab showed
  mutations that believed to change the structure
  of the protein
• Mutation was published in scientific literature,
  noting its effect on the binding of Daxx protein
  to another protein other protein controls one
  of the phenotypes of Trypanosomiasis resistance
• Sex dependant biological pathways identified in
  Trichuris muris infection
• 2 year study of candidate genes did not identify
  these pathways directly
• Sex dependence identified as a possible factor in
  expulsion of parasite
• Current studies still ongoing in Life Science
  department
• FOUND NEW BIOLOGY !!!!

21
Taverna Workflow Workbench
http//taverna.sf.net
22
Ideally What Next?
Protein Info
Protein-Protein Interactions
Fill in the blanks and verify hypotheses
Biological processes
23
Link Phenotype to Pathways
Find common terms

• Phenotype Terms
• apoptosis
• Cholesterol
• Diabetes
• Jak-Stat
• Ribosome
• Cell Adhesion Molecules

apoptosis Cell Death JNK pathway
Cholesterol Cell Death JNK pathway
Simple means of linking pathways

• apoptosis
• Cholesterol
• Diabetes
• Cell Death
• JNK pathway
• Another pathway

24
Publication of Results

• Published methods and biological findings
• Nucleic Acids Research
• Presented Poster at ISMB Vienna (2007)
• Abstract for Poster to be published in BMC
  Bioinformatics Journal
• Assisted in Writing Book chapter on Building
  workflows that traverse the bioinformatics data
  landscape, To appear in Data Mining Techniques
  in Grid Computing Environments