Creating Smart Clients with the
Collaboration Notebook
Greg Quinn
Principal Investigator
Desktop and Mobile Data Management
San Diego Supercomputer Center
Topics
Data overload
The Collaboration Notebook: a data
management solution
Smart client development framework
Future work
Science research: it’s about the DATA!
Locally generated, e.g. observations of a
researcher at the lab bench
Theoretical data
Archived local data, collected from many
researchers
Data in legacy databases
Data from collaborators, both near and far
Data from Internet sources such as NCBI
(NIH)
The Encyclopedia of Life Project
More than 800 genomes
that have been completely
or partially sequenced
For each publicly available
protein sequence derived
from genomic data, EOL
attempts to locate structural
domains and correlate this
data with other publicly
available sequence
annotation
A large amount of
information for end-users to
collate
Data management app
Many Excellent Online Data Sites
BLAST
NCBI
BLAST
Protein NR updated
nightly
(~2 million
sequences)
XML
DATA OVERLOAD!
Online data sources
Legacy and archived data
Collaborative research
Research notes and
observations
DATA OVERLOAD!
No meaningful way to store data
No simple way to re-purpose online data
No mechanism to share data
Difficult to keep track of data acquired during an
online session
Resource-intensive searches frequently repeated
When data is retrieved, it can be difficult to
manipulate and/or search
No simple way to annotate downloaded data
The Collaboration Notebook
A desktop application to better enable the scientific
researcher and knowledge worker utilize network
information resources and manage data
Leverages features of Windows and the .Net development
Powerful local db with search functionality
“Knowledge” of data types through the use of Ontologies
Ability to annotate stored data
Peer-to-peer querying of stored data and annotations
Data export capability to popular formats
Unattended/automatic data updates via web services & HTTP
User notification of new data
Plugin API for data visualization components – c/w basic data
viewers for popular Bio-data types, e.g. text, DNA seqs, etc.
Smart client framework for SOAP-based web services
Point-and-click interface to support Tablet PC’s and ink data types
Data Sources
XML doc
Local datastore
personal
database
Data presentation and Smart
client for network data services
personal
database
personal
database
The Connected Research Environment
Web
Interfaces
SOAP
Services
Other Data
Sources
Report and paper preparation
P2P
Research data access, input & annotation
Group collaboration
GUI Design schematic
Data browser
Smart Client
display area
N-App
availability
P2P
Collaboration
group
Fast search
options
Winforms
Avalon
Embedded apps
Collaboration Notebook N-Apps
Networked
Data services
N-App
Server
EOL
PDB
NCBI
Workbench
Work
bench
N-App Collection
…
Smart Client..?
Local resources
Connected
Offline capable
Intelligent deployment and update
Collaboration Notebook
N-App
Interface class
Reference
Data service
wrappers
Available Services
Run request
Data services
Service
wrapper
SOAP
services
Service
wrapper
Web
Pages
Service
wrapper
DB’s
data
Data changed
event
manager
Service
wrapper
Smart Client
Local DB
Collaboration
subsystem
Indigo
Local
Instrumentation
Stages in Smart Client Dev
Write your application with Windows forms
or Avalon (control library project)
Reference Notebook interface class and
Notebook data objects class to gain
access to Notebook services (e.g. data
access, persistence and retrieval)
Write a wrapper for the online service (if
not using a pre-existing one)
Write a data translator to convert incoming
data into internal canonical format
Debugging Smart Clients
Service
wrapper
Test Harness
Data
Service
Embedding a Pre-existing
Application
Create a new Class Library project in
Visual Studio
Add reference to executable
Use Visual Studio object browser to
determine available methods
Add references to Notebook interface
class and Notebook data objects class
Wrapper communicates with
Embedded app
Wrapper
Worldwind.menuItemEditPaste.PerformClick()
N-App
System Clipboard
WorldWind://lat=x&long=y
Lat/Long data
Notebook
Service
wrapper
Data
Service
Example N-Apps
BLAST interface
Winforms
BioConductor Interface
Avalon (XAML)
World Wind
Embedded
BLAST interface
BLAST –
Basic Local Alignment Search Tool
Several BLAST SOAP services available:
NCBI
PDB
EBI
DDBJ
Unified interface to these services
BLAST Interface Smart Client
Microarray Data Analysis
Developed by Dr. Robert Byrnes
Statistical analysis of microarray data can be used to determine
up- and down-regulation of genes, expression signatures, etc.
The Bioconductor package is a widely used set of biostats
routines that can be used to analyze microarray data
Written in “R” language
Current GUI’s to use Bioconductor are relatively crude
XAML-based Applications
C#
XAML
<FlowPanel
xmlns="http://schemas.microsoft.
com/2003/xaml">
<Text>Hello World</Text>
<Button>Click me!</Button>
</FlowPanel>
+
public partial class MyPanel
{
public string Hello()
{
return "Hello!";
}
}
Microarray Data Analysis
Developed by Dr. Robert Byrnes
R Interpreter
Notebook App
Bioconductor
packages
Avalon Control
XML
Metadata
SQL
SERVER
Microarray App
Microarray Data Analysis
Developed by Dr. Robert Byrnes
Embedded Applications
.Net-based app’s can be embedded with no
modification required
Need to creatively use existing methods within the
application for data analysis, persistence and retrieval
Embedded Applications
Smart Clients Under Development
Protein Data Bank (PDB)
Next Generation Biology WorkBench
National Ecological Observatory
Network (NEON)
Homeland Security projects
CORE APP DEV
ADVANCED FEATURES
CORE FEATURES
DATA PLUGIN API
SERVICE PLUGIN API
XAML DEV FRAMEWORK
SQL SERVER
FIXED INTERNAL DATA MAPPING
P2P COMMUNICATIONS
INTEGRATION OF DATA MAPPING SOLUTION
NOTEBOOK RUNTIME SERVICE
BACKGROUND DATA UPDATES
INTEGRATION INTO “LONGHORN” PLATFORM
DATA MAPPING &
DISTRIBUTED QUERY ADVANCED QUERY PROCESSING LAB
ONTOLOGY-BASE DATA MAPPING
DISTRIBUTED QUERY OPTIMIZATION
CROSS-DOMAIN QUERY
N-APP DEV
CHARTER N-APPS
MICROARRAY DATA ANALYSIS APP
GEOSCIENCES-BASED APP
OTHERS…
FIELD TRIALS & TESTING
INTERNAL
UCSD RESEARCH LABS
SOCIAL MONITORING
2004
2005
PUBLIC
BETA TESTING
BUG TRACKING
2006
Hands-on One Day Workshop:
Creating applications with the
Collaboration Notebook
Date: October 2005
Location: San Diego Supercomputer Center
Further info: notebook@sdsc.edu
Acknowledgements
Programmers
Blair Jennings (lead)
Robert Byrnes
Martin Dubcovsky
Kevin Fowler
Acknowledgements
Project Support
Dan Fay & Microsoft Research
Mark Miller, SDSC
The SDSC Synthesis Center
Questions?
Greg Quinn
quinn@sdsc.edu
http://www.notebookproject.org
© 2005 Microsoft Corporation. All rights reserved.
This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.