INCREASING THE EFFICIENCY OF PHARMACEUTICAL RESEARCH THROUGH DATA INTEGRATION Dr. Roland Bauer 12-15.Oct. 2014 ICIC 2014 Heidelberg Project Manager Content Integration & Development Elsevier Information Systems GmbH, Frankfurt [email protected] Matthew Clark Ph.D. Consultant, Life Science Services Elsevier Inc. Philadelphia, PA [email protected]
ICIC 2014 Increasing the efficiency of pharmaceutical research through data integration
Jun 11, 2015
The pressures of pharmaceutical research and development demand increasing efficiency from scientists. High-quality decisions must be made faster and encompass all available information. At the same time there is a growing desire to better utilize the multi-billion dollar research investment recorded in laboratory notebooks and bioassay databases. Key values for data integration in a data exploration environment include gathering data from disparate E-notebooks and bioassay databases into a single searchable “virtual” system and increased discoverability by accessing data through a system designed for exploration. Key benefits are better chemistry decisions through easier access to broader data and reduced time for preparing patent filings. The ability to interlink in-house and reported assay data with in-house and published chemistry provides a data-rich environment for developing insights and predictive models. We will discuss our experience with integrating information from journals, patents, bio-assay databases, and E-lab notebooks to address these needs.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INCREASING THE EFFICIENCY OF PHARMACEUTICAL RESEARCH THROUGH DATA INTEGRATION
Dr. Roland Bauer
12-15.Oct. 2014 ICIC 2014 Heidelberg
Project Manager Content Integration & Development Elsevier Information Systems GmbH, Frankfurt [email protected]
Matthew Clark Ph.D. Consultant, Life Science Services Elsevier Inc. Philadelphia, PA [email protected]
2
ABOUT ME:
- “Babes –Bolyai” University, Cluj-Napoca,
Romania
- Max-Planck-Institute for Polymer Research,
Mainz, Germany
- Elsevier
3
Introduction & Setting the Stage
Why?
Content Integration : The Reaxys Case
Integration Process Project Overview
AGENDA
4
INTRODUCTION & SETTING THE STAGE: THE DRUG DISCOVERY INFORMATION LANDSCAPE
5
INTRODUCTION & SETTING THE STAGE: TENDENCIES IN THE DRUG DISCOVERY INFORMATION LANDSCAPE
6
INTRODUCTION & SETTING THE STAGE: TENDENCIES IN THE DRUG DISCOVERY INFORMATION LANDSCAPE
WHY?
7
ISSUE: CHEMICAL INFORMATION ACCESS IS FRAGMENTED
• End users must learn many
interfaces
• Different data sources
have different
capabilities for searching
• Scientists may not search all
appropriate data sources
Licensed
Database
Licensed
Database
Catalog
Catalog
E-Notebook
E-Notebook References/
Full Text
INTEGRATION OF DATA PROVIDES BETTER ANSWERS
Searching multiple sources with one search via a single
interface increases efficiency
Harmonized indexing allows asking similar question
among all sources
9
Easier
Access Enhanced
usage
Better
value for
investment
Better
decisions
Faster
progress
CONTENT INTEGRATION : THE REAXYS CASE
10
11
THE REAXYS DATABASE: CONTAINS INTEGRATED PUBLISHED CHEMISTRY DATA
12
THE REAXYS DATABASE: …ALONG WITH EXPANDED BIBLIOGRAPHICAL INFORMATION
13
THE REAXYS TREE : BROWSE CONTENT BY ONTOLOGY
TWO APPROACHES TOWARDS INTEGRATED CONTENT
14
Analysis system
End-User
Central
Storage
FEDERATED MODEL WAREHOUSE MODEL
TWO APPROACHES TOWARDS INTEGRATED CONTENT
15
FEDERATED MODEL WAREHOUSE MODEL
Pros: - Easy scalability in case of new
data sources - Delivery of short term „wins“
- Maintenance costs
Cons: - Lack of normalization and
harmonized indexing
- Performance and availability dependent on the source systems
Pros: - High data quality trough
normalization
- Unified Queries and Filters applicable
Cons: - Long implementation times &
higher starting costs
- Expensive and difficult to accommodate changes in data types
16
REAXYS EXTERNAL CONTENT INTEGRATION
Database
End-User
ELN 1
ELN 2
CUSTOM IN HOUSE
REACTIONS SOURCE
Indexed Storage
RX CONTENT EXTERNAL CONTENT
Customer Hosted
17
REAXYS EXTERNAL CONTENT INTEGRATION: IN HOUSE SCENARIO
Database
End-User
ELN 1
ELN 2
CUSTOM IN HOUSE
REACTIONS SOURCE
Indexed Storage
RX CONTENT EXTERNAL CONTENT
Customer Hosted
Elsevier Hosted
18
REAXYS EXTERNAL CONTENT INTEGRATION: ELSEVIER HOSTED SCENARIO
Database
End-User
ELN 1
ELN 2
CUSTOM IN HOUSE
REACTIONS SOURCE
Indexed Storage
RX CONTENT EXTERNAL CONTENT
Customer Hosted Elsevier Hosted
19
REAXYS EXTERNAL CONTENT INTEGRATION: HYBRID HOSTING SCENARIO
Database
End-User
ELN 1
ELN 2
CUSTOM IN HOUSE
REACTIONS SOURCE
Indexed Storage
RX CONTENT EXTERNAL CONTENT
REAXYS PROVIDES A UNIFIED INFORMATION PORTAL
• Provides a single powerful
interface
• Can integrate several
notebook systems
• Links chemistry, structures,
sourcing, citations, and
full-text of articles
Structures,
reactions, and
Full-Text
Licensed
Reaction and
Structure
Databases
E-Notebook Binding,
Properties
E-Notebook
Patents
INTEGRATED SOLUTION SEARCH
21
List of integrated
sources
Sources list can include licensed
databases, and multiple e-
notebooks from organizational
units
All e-notebooks can be integrated
and searched together
REACTION SEARCH RESULTS SEPARATED BY SOURCE
22
Results from
each source on
separate tab
Show corresponding
substances in …
PubChem
eMolecules
Licensed
PharmaCo e-notebook
PharmaCo2 e-notebook
Cross link to
substance in all
other sources
where it is found
E-notebooks
SUBSTANCE RESULTS
23
Results from each source on separate tab
Including PubChem and eMolecules
All filters fully
active
INTEGRATION CASE STUDY: ROCHE IN HOUSE HOSTED
Integrated Reaxys with several data sources:
• Medicinal Chemistry E-notebooks
• Development Chemistry E-notebooks
• Several E-notebook systems of acquired organizations
• Licensed Databases
• Current Chemical Reactions
• Several other databases
Links to many more sources
• Roche stockroom availability
• Patent/Literature full text
• Link to original e-notebook pages
24
Reaxys integrates these e-
notebooks with each other,
while they are still maintained
as separate systems
CASE STUDY: ROCHE KEY DRIVERS
From ACS Presentation by
Michael Kapler, Roche Pharma Research and Early Development
http://abstracts.acs.org/chem/245nm/program/view.php?obj_id=188977
INTEGRATION PROCESS
PROJECT OVERVIEW
26
PROCESS OVERVIEW FOR AN INTEGRATION PROJECT
Initialisation:
- Evaluation of Datasources and needed
resources
- Determine hosting scenario
- Commercial and legal framework
Kick-of:
-requirements harvesting
-establish milestones and top down workstreams
-Refine & finalize plans
Execution:
-implement automatised ETL process
-implement application customisation
-install IT infrastructuee and interfaces
Delivery
- BETA release
- Refinement
- GoLive
Sprint 1
Sprint 2
Sprint 3
Sprint 4
Hand over to BAU /Maintenance
- Sprint Iterations a 3 weeks -Sprint number dependent on complexity (5-…)
DATA MODEL AND USER INTERFACE PROCESS
Determine
data
sources
Map to Reaxys
Integration Data
Model (XML)
User interface
configuration for new
fields
Unit
conversions,
data cleaning
E-notebooks
Licensed databases
°C, K
moles, grams
Identify URL
links to E-
notebooks
an other
resources
This is a key step for
the integration project Design location, nature of displaying
the fields, urls etc.
AUTOMATISED FABRICATION PROCESS (ETL)
E-notebook
1
E-notebook
2
Bioassay db
Transmit to
fabrication
server (sftp,
scp)
Fabrication
combines data
with Reaxys data
for production
Daily
extraction to
XML using
defined data
model …
30
THANK YOU – QUESTIONS?
Project Manager Content Integration & Development Elsevier Information Systems GmbH, Frankfurt [email protected]
Matthew Clark Ph.D. Consultant, Life Science Services Elsevier Inc. Philadelphia, PA [email protected]
Dr. Roland Bauer
Related Documents
