Integrating Clinical and Genomic Information through the PrognoChip Mediator

1
Integrating Clinical and Genomic Information
through the PrognoChip Mediator

• Implemented by
• Manos Kalaitzakis, Dimitris Manakanatas,
  Haridimos Kondylakis, Manolis Nikoloudakis, and
  Maria Michou
• Under the guidance of
• Anastasia Analyti Dimitris Plexousakis
• Institute of Computer Science, FORTH-ICS, Greece
• Department of Computer Science, University of
  Crete, Greece
• Medical Doctors
• Elias Sanidas, Efstathios Stathopoulos, Maria
  Kafousi
• Medical School, University of Crete, Greece
• Stamatis Vassilaros
• Prolipsis Diagnostic Breast Center, Athens,
  Greece

2

• The aim of PrognoChip is
• the identification of molecular markers for the
  classification and prognosis of Breast Cancer,
  based on
• the correlation of patients clinicohistopatholog
  ical parameters and therapy response with their
  tumor gene expression profiles

3
Tumor Tissue Route
liquid nitrozen or ice
Surgery
Tumor sample
Histopathologist
PAGNH University Hospital PROLIPSIS Breast
Cancer Center
4
The Integrated Clinico-Genomics Environment of
PrognoChip

• Consists of
• a Clinical Information System to keep patient
  clinical information (i.e., clinical, laboratory,
  and histopathological information),
• a Genomic Information System to manage DNA
  microarray experiments, analyze raw hybridization
  data, and store gene expression profiles,
• a middleware layer, called PrognoChip Mediator,
  for the integration of the Clinical and Genomic
  Information Systems, and
• a Data Mining layer, for the intelligent
  processing of the retrieved clinico-genomic data,
  knowledge extraction, and visualization.

5
Three Databases to Integrate

• Genomic Information System (extension of BASE
  v1.2.16, developed by Lund Univ.)
• microarray design, samples, hybridizations, etc.
• Clinical Information System (extension of
  Integrated Care Solutions, developed by
  FORTH-ICS.)
• patient medical history, current findings,
  diagnostic test results, tumor histopathologic
  diagnosis, etc.
• Gene Ontology (GO) Database
• ontology terms for the annotation of gene products

6
Visualization Tools
findings
PrognoChip The Integrated Clinico-Genomics
Environment
metadata
Findings Repository
Data Mining Tools
store/query findings
Specify parameters
output XML file
biomedical investigator
(local access)
Mediator DB
Mediator
store administration temporary data
virtual query
Query Engine
(web-based)
biologist
Clinical IS
Genomic IS
Submit/search/ update
GO database
(web-based)
(client-server based)
local installation
Normalization Plug-ins
7
The PrognoChip Mediator

• The authorized biomedical investigator can form
  clinico-genomic queries through a web-based
  graphical user interface.
• Specifically, the user can specify criteria for
  selecting tumors
• excised from patients with a clinical profile of
  interest,
• having histopathologic characteristics of
  interest, and
• participating in a PrognoChip microarray
  experiment of specified quality and
  characteristics.
• Returned tumors are accompanied with desirable
  clinico-genomic information (e.g., gene
  expression profiles).

8
Clinical Criteria

• Patient past breast diseases
• Gynecological history (e.g. age of menarche,
  menopausal status, etc.)
• Total duration of hormone intake
• Other risk factors (e.g. smoking habits, etc. )
• Family history of malignancy
• Current findings and blood tumor markers
• Therapies before and after surgery
• Effectiveness of therapies

9
Clinical Query Forms
10
(No Transcript)
11
10
11
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
Tumor Histopathologic Criteria

• Tumor histological type
• Diameter
• Grade
• Lymphatic, Veinous, Perineural, Skin, Nipple, and
  Lymph node invasion
• Histological TMN
• Immunohistochemical tumor markers
• (ER, HER-2, etc.)

21
Histopathologic Query Forms
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Genomic Criteria

• The channel of the cancerous tissue
• Hybridization quality indicators
• Dates of hybridizations
• The name of the desired normalization procedure
• Reporter/Gene filtering criteria
• GO annotations and evidence code
• Molecular Function, Biological Process, and
  Cellular Location ontology terms

27
Genomic Query Forms
28
(No Transcript)
29
(No Transcript)
30
Select Page After patient/tumor criteria
specification, the user selects the clinical,
histopathologic, and genomic fields of interest
that will accompany the cancerous sample in the
output.
31
Select Page Part 1
32
(No Transcript)
33
Select Page Part 2
34
(No Transcript)
35
Query Processing

• When a Query Form is saved, part of the related
  subquery is formed.
• When the Select Page is filled out, final
  clinical subqueries are formed.
• Clinical subqueries are submitted to Clinical IS
  and results are composed.
• Based on the tumor names returned in 3), final
  genomic queries are formed and submitted to
  Genomic IS.
• Results from Clinical IS and Genomic IS are
  joined based on tumor name.
• Output XML file is formed, based on a
  predetermined schema.

36
Output XML File Schema

• For the full Output XML File schema, see
• http//www.ics.forth.gr/analyti/PrognoChip/
  OutputXMLFileSchema.jpg
• Description is provided in
• http//www.ics.forth.gr/analyti/PrognoChip/
  OutputXMLFileDescription.pdf

37

• Result Page
• In the Result Page, the user can
• download the complete set of output files to
  his/her machine,
• view the output XML file, and
• download previous results.

38
(No Transcript)
39
Output XML File
40
Data Mining Layer

• It supports decision making operations, on output
  XML files.
• Examples
• Retrieve the tumor gene expression profiles of
  patients that meet clinical profile A and
  clinical profile B. Then, find the genes that
  best discriminate between these two groups.
• Cluster patients based on their tumor gene
  expression profiles. Then, identify indicative
  patient clinical profiles, for these clusters.

41

• Details on the PrognoChip Mediator can be found
  in
• Anastasia Analyti, Haridimos Kondylakis, Dimitris
  Manakanatas, Manos Kalaitzakis, Dimitris
  Plexousakis, George Potamias, Integrating
  Clinical and Genomic Information through the
  PrognoChip Mediator, Procs. of the 7th
  International Symposium on Biological and Medical
  Data Analysis (ISBMDA-2006), 250-261, 2006.
  Springer-Verlag

42
PrognoChip Mediator Web Site

• http//www.ics.forth.gr/isl/projects/PrognoChip/

43
Acknowledgements

• The development of PrognoChip Mediator was
  supported by National funds within the context
  of
• the PrognoChip project (GSRT-EPAN).

44
Thank you !!!