Computational research for medical discovery at Boston College Biology - PowerPoint PPT Presentation

Loading...

PPT – Computational research for medical discovery at Boston College Biology PowerPoint presentation | free to download - id: 7cdbc1-YjFmZ



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Computational research for medical discovery at Boston College Biology

Description:

Computational research for medical discovery at Boston College Biology Gabor T. Marth Boston College Department of Biology marth_at_bc.edu http://clavius.bc.edu/marthlab – PowerPoint PPT presentation

Number of Views:102
Avg rating:3.0/5.0
Slides: 23
Provided by: Gabo70
Learn more at: http://clavius.bc.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Computational research for medical discovery at Boston College Biology


1
Computational research for medical discovery at
Boston College Biology
Gabor T. Marth Boston College Department of
Biology marth_at_bc.edu http//clavius.bc.edu/marthla
b
2
We study genetic variations because
they underlie phenotypic differences
3
Our current projects investigate three essential
aspects of genetic variations
  • how to discover inherited genetic polymorphisms
    that lead to disease?

4
1. We build computer tools for variation
discovery
inherited (germ line) polymorphisms are important
as they can predispose to disease
1.
5
we are currently expanding our polymorphism
detection capabilities.
  • for automated detection of somatic single base
    pair mutations in diploid samples

6
2. We measure genome-wise distributions of DNA
polymorphism data
1. marker density (MD) distribution of number of
SNPs in pairs of sequences
7
we build models of these distributions under
competing scenarios of human demographic history
stationary
expansion
collapse
bottleneck
past
history
present
MD (simulation)
AFS (direct form)
8
and determine the best-fitting models.
European data
African data
Marth et al. PNAS 2003 Genetics 2004
9
3. The HapMap project aims to map out human
polymorphism structure to aid gene mapping
10
we generate quasi-samples with computational
means to study sample-to-sample variability
Instead of genotyping additional sets of
(clinical) samples with costly experimentation,
and comparing the variation structure of these
consecutive sets directly
11
and to optimize tag SNP (marker) selection for
clinical association studies.
12
We are developing projects to expand
  • from single-nucleotide DNA changes to developing
    computer tools for the detection of other types
    of genomic and epigenetic changes (e.g. in
    cancer)
  • to developing visualization and statistical
    tools for the integration of diverse genetic and
    epigenetic data

(Image from Nature Reviews Genetics)
13
Detecting SNPs in medical re-sequencing data,
short insertions / deletions
  • detection in new data types produced by the
    latest, super-high throughput sequencing
    technologies (i.e. 454 Life Sciences sequencing
    machines) that will be used for individual
    medical re-sequencing

14
Using SNP array data intelligently to detect
chromosomal aberrations
Speicher Carter, NRG 2005
15
Software development for other genetic and
epigenetic data (focus on data confidence)
copy number detection
16
Integrate genetic and epigenetic data from varied
sources to find common themes during cancer
development
chromosome rearrangements
17
Using new haplotype resources to connect genotype
and clinical outcome in pharmaco-genetic systems
  • the HapMap was designed as a tool to detect
    high-frequency (common) phenotypic (e.g.
    disease-causing) alleles
  • important drug metabolizing enzymes are
    relatively few in number, well studied, are at
    known genome locations, many associated
    phenotypes are well described
  • many functional alleles are known, and of high
    frequency (common)
  • multi-SNP alleles are highly predictive of
    metabolic phenotype
  • clinical phenotype (adverse drug reaction) less
    predictable
  • ideal candidate for applying haplotype resources

18
Multi-marker haplotypes as accurate markers for
ADRs?
molecular phenotype (drug concentration measured
in blood plasma)
functional allele (known metabolic polymorphism)
19
Resources
  • specifics of enzyme-drug interactions

20
Evolutionary / PopGen questions
  • mutation age?

21
Proposed steps of analysis
  • haplotypes vs. metabolic phenotype?
  • haplotypes vs. ADR phenotype?

22
Funding sources / plans
  • polymorphism discovery medical re-sequencing
    data analysis 5-year NIH R01 research grant
    awarded
  • pop-gen modeling haplotype analysis marker
    selection system NIH R01 application pending
  • informatics tools for genomic and epigenetic
    changes in cancer need a postdoc to establish
    project (startup or NIH R21 or private funding)
  • haplotypes in Pharmacogenomics need a postdoc
    to establish project (startup or NIH R21 or
    private funding)
About PowerShow.com