Presentacin de PowerPoint

1
Case Study 1 The modeling of genetic regulatory
networks in cancer
Barbara, Laura, Marissa, Michael, Sander
1st Infobiomed Training Challenge, 16 September,
Barcelona, Spain
2
Participant expertise
Barbara Pharmacy PhD student (medicinal
chemistry), employement in a pharmacy Laura Biol
ogy, bio-informatics Integrative Biomedical
Informatics, Post-doc Marissa Medicine Resident
internal medicine, PhD student on epidemiology
of CRPS Michael Physics Molecular dynamics and
computer progamming, Phd student on protein
folding Sander Medical Biology PhD student
Immunogenetics of mucosal infectiions
3
From Way to PathorWhat group dynamic is all
about
4
Case description

• -Steroid hormones are involved in a wide variety
  of physiological processes
• Also Involved in the pathogenesis of several
  cancers including breast, prostate, uterine and
  ovarian tumours.
• Goal increasing the understanding of molecular
  mechanisms that underlie the cancer development
  regulated by steroid hormones

5
Breast Cancer
20 incidence of breast cancer in USA In women,
most frequent kind of cancer around the
world Strong relation with estrogen exposure
Important role for anti-estrogens in therapy
(adjuvant)
6
Anti-estrogen therapy

• Two major types
• Selective Estrogen Receptor Modulator (SERM)
• Advantages only slight increase of osteoporosis
• Disadvantages increased risk for thrombosis and
  endometrial cancer
• Aromatase inhibitors (block estrogen production)
• Advantage no increased risk for thrombosis and
  endometrial cancer
• Disadvantages increased risk for osteoporosis

7
Anti-estrogen therapy
Sometimes development of anti-estrogen resistence
occurs in previously anti-estrogen responsive
patients
8
SERMs (selective ER modulators)

• ER of different target tissues vary in chemical
  structure
• SERMs selectively stimulate or inhibit ER of
  different target tissues
• E.g. a SERM might inhibit ER in breast cells
  (proliferation inhibited) but activate ER in
  uterine endometrial cells (proliferation
  stimulated)

9
SERM activity
10
Resistance to SERMs in breast cancer treatment

• SERMs may bind in antagonistic and agonistic
  manner to the same receptor!
• Limitation of the effective treatment of
  hormone-responsive breast cancer
• ER? in most cases due to changes in cell
  signaling pathways! (only in small proportion due
  to mutations in ER?)

11
What are the clinical problems?
How to predict responsiveness to
anti-estrogen treatment in ER patiens? When is
additional chemotherapy neccessary? How can we
avoid/by-pass anti estrogen therapy
resistance? How to target the effect of anti
estrogens specific to breast cancer cells? To
answer insight in the mechanisms of estrogen
stimulated cell proliferation is necessary
12

• Research proposal
• General goal to advance in the understanding of
  proliferation of breast cancer cells and its
  relationship with steroid hormones and tamoxifen
  treatment.
• Specifics goals
• Model the genetic regulatory networks that
  underlie the proliferation of breast cancer cells
  by integration of high-throughput data and
  literature
• Predict the behaviour of the cell in response to
  estrogen and tamoxifen

13
Modelling of gene regulatory networks
(Davidson et al, Science 2002)
The analysis of genetic networks can give
information about the nature of the regulation of
gene expression.
14

• Approaches utilised in modelling of gene
  expression
•  
•  
•  
• Description of the expression of a few genes at
  the level of transcription by detailed
  mathematical models, that include the binding of
  transcription factors to DNA, effect of
  activators, the regulation by internal feedback
  loops or external regulators.
• Analysis in parallel of the changes on expression
  of thousand of genes over time or different
  experimental conditions. These gene expression
  profiles are analysed to search for clusters of
  co-expressed genes and to deduce regulatory
  information.
• Use both approaches in a complementary way to
  fill the unknowns components and/or connections
  of a given pathway and perform a dynamical
  simulation

15
(No Transcript)
16
Methods
Text mining tools Databases
Reverse engineering methods
pathway comparison
Dynamical modelling methods
Identification of the pathway
17
literature
Text mining tools (PathwayAssist, iHop) and
databases
pathway
Dynamical modelling methods (CellDesigner)
simulation
model of breast cancer proliferation
18
Clinical data
Many proteins are described to have an altered
expression in estrogen responsive breast cancer,
including cyclin D1. The cyclin D1 protein is
overexpressed in 50 of the breast cancer
cases. This implicates that several pathways are
involved in estrogen mediated tumor
proliferation. In most articles overexpression
of cyclin D1 is correlated with better response
to tamoxifen therapy and less recurrence of the
tumor. In contrast, there are few articles
describing a negative correlation.
19
NAFP pathway
20
microarray data
regulatory network of breast cancer
Reverse engineering methods

• Identification of the NAFP pathway within the
  network
• Comparison of both pathways NAFP pathway vs
  pathway obtained from microarray data
• Fill the missing information in the NAFP pathway
  using microarray data thus refining the model for
  the dynamic simulation

21

• Methods
• Text mining tools PathwayAssist, iHOP
• Reverse engineering methods ARACNE (Basso et al,
  Nature Genetics 2005)
• Dynamical modelling CellDesigner, Byodyn,
  Mathematica
• Data sources
• Microarray data from breast cancer and normal
  cells http//genie.rockefeller.edu/genomica/
• Literature PubMed
• Databases Biocarta, iPath, BIND, OMIM,
  SwissProt, Entrez

22
Proposal for future research
Since the NAFP cannot be the only pathway
involved in estrogen mediated cell
proliferation, it is important to model other
involved pathways. The pathway assist tool can
be useful in defining wich pathways are in the
centre and should be modeled. More knowledge
about the contribution of the different pathways
in the process will help to predict the effect of
therapy and give ideas of where these pathways
could be targeted.
23
Group results
Laura
Barbara
Sander
Michael
Marissa
24
Cross-talk
Laura
Barbara
Sander
Michael
Marissa
25
What did we learn within these training challenge?

• U have 2 leave (partially) your WAY (of thinking,
  talking, feeling,) to find a conjoint PATH
• U have 2 have RESPECT (not only for the others
  but also for your own knowledge)
• Dont be shy there are no stupid questions!
• FUN leads to better team-work and vice versa

26
Suggestions for future training challenges

• FUN can not be generated while feeling
  dis-stressed
• Eu-stress will be the driving force anyway (due
  to the researchers personality)!
• Some groups may need some kind of intervention.

27
ER
Cyclin D
PELP 1
Es
c-Src
PELP 1/ER/Es
PELP 1/ER/Es/c-Src
ERK1/2
ER/Es
Active Homodimer degradation
Active homodimerc
Inactive homodimer
inactive Homodimer degradation
28
Thank you!
to the organizers and the tutors!