KnowledgeBased Integration of Neuroscience Data Sources

1
Knowledge-Based Integration of Neuroscience Data
Sources

• Amarnath Gupta
• Bertram Ludäscher
• Maryann Martone
• University of California San Diego

2
A Standard Information Mediation Framework
Client Query
Integrated XML View
Mediator
XML View
XML View
XML View
Wrapper
Wrapper
Data Source
XML Data Source
Data Source
3
A Neuroscience Question
Cerebellar distribution of rat proteins with more
than 70 homology with human NCS-1? Any structure
specificity? How about other rodents?
Integrated View
Mediator
Wrapper
Wrapper
Wrapper
Wrapper
WWW
CaBP, Expasy
protein localization
morphometry
neurotransmission
4
Integration Issues

• Structural Heterogeneity
• Resolved by converting to common semistructured
  data model
• Heterogeneity in Query Capabilities
• Resolved by writing wrappers with binding
  patterns and other capability-definition
  languages
• Semantic Heterogeneity
• Schema conflicts
• Partially resolved by mapping rules in the
  mediator
• Hidden Semantics?

5
Hidden SemanticsProtein Localization

• ltprotein_localizationgt
• ltneuron typepurkinje cell /gt
• ltprotein channelredgt
• ltnamegtRyRlt/gt
• .
• lt/proteingt
• ltregion h_grid_pos1 v_grid_posAgt
• ltdensitygt
• ltstructure fraction0.8gt
• ltnamegtspinelt/gt
• ltamount nameRyRgt0lt/gt
• lt/gt
• ltstructure fraction0.2gt
• ltnamegtbranchletlt/gt
• ltamount nameRyRgt30lt/gt
• lt/gt

6
Hidden Semantics Morphometry

• ltneuron namepurkinje cellgt
• ltbranch level10gt
• ltshaftgt
• lt/shaftgt
• ltspine number1gt
• ltattachment x5.3 y-3.2 z8.7 /gt
• ltlengthgt12.348lt/gt
• ltmin_sectiongt1.93lt/gt
• ltmax_sectiongt4.47lt/gt
• ltsurface_areagt9.884lt/gt
• ltvolumegt7.930lt/gt
• ltheadgt
• ltwidthgt4.47lt/gt
• ltlengthgt1.79lt/gt
• lt/headgt
• lt/spinegt

7
The Problem

• Multiple Worlds Integration
• compatible terms not directly joinable
• complex, indirect associations among schema
  elements
• unstated integrity constraints
• Why not use ontologies?
• typical ontologies associate terms along limited
  number of dimensions
• Whats needed
• a theory under which non-identical terms can be
  semantically joined

8
Our Approach

• Modify the standard Mediation Architecture
• Wrapper
• Extend to encode an object-version of the
  structure schema
• Mediator
• Redesign to incorporate auxiliary knowledge
  sources to
• Correlate object schema of sources
• Define additional objects not specified but
  derivable from sources
• At the Mediator
• Use a logic engine to
• Encode the mapping rules between sources
• Define integrated views using a combination of
  exported objects from source and the auxiliary
  knowledge sources
• Perform query decomposition
• We still use Global-as-View form of mediation

9
The KIND Architecture
Integrated User View
View Definition Rules
Auxiliary Knowledge Source 1
Logic Engine
Integration Logic
Auxiliary Knowledge Source 2
Schema of Registered Sources
Materialized Views
Src 2
Src 1
10
The Knowledge-Base

• Situate every data object in its anatomical
  context
• An illustration
• New data is registered with the knowledge-base
• Insertion of new data reconciles the current
  knowledge-base with the new information by
• Indexing the data with the source as part of
  registration
• Extending the knowledge-base
• Creating new views with complex rules to encode
  additional domain knowledge

11
F-Logic for the Mediation Engine

• Why F-Logic?
• Provides the power of Datalog (with negation) and
  object creation through Skolem IDs
• Correct amount of notational sugar and rules to
  provide object-oriented abstraction
• Schema-level reasoning
• Expressing variable arity
• F-Logic in KIND
• Source schema wrapped into F-Logic schema
• Knowledge-sources programmed in F-Logic
• Definition of Integrated Views

12
Wrapping into Logic Objects

• Automated Part

lt!ELEMENT Studies (Study)gt lt!ELEMENT Study
(study_id, animal,
experiments, experimentersgt lt!ELEMENT experiments
(experiment)gt lt!ELEMENT experiment (description,
instrument, parameters)gt
studyDBstudies ? ? study. studystudy_id ?
string animal ? animal
experiments ? ? experiment
experimenters ? ? string.

• Non-automated Part
• Subclasses
• Rules
• Integrity Constraints

mushroom_spinespine
Smushroom_spine IF Sspinehead?_neck ?_.
ic1(S)alerttype ? invalid spine object S IF
Sspineundef ? ? head, neck.
13
Computing with Auxiliary Sources

• Creating Mediated Classes
• Reasoning with Schema

animalM?R IF Ssource, S.animal M?R
. animaltaxon ? TAXON.taxon. Xtaxon?T IF X
PROLAB.animalname ?N,
words(N,W1,W2_), T
TAXON.taxongenus ?W1species ?W2.
14
Integrated View Definition

• Views are defined between sources and knowledge
  base
• Example protein_distribution
• given organism, protein, brain_region
• KB Anatom
• recursively traverse the has_a paths under
  brain_region collect all anatomical_entities
• Source PROLAB
• join with anatomical structures and collect the
  value of attribute image.segments.features.featur
  e.protein_amount where image.segments.features.f
  eature.protein_name protein and
  study_db.study.animal.name organism
• Mediator
• aggregate over all parents up to brain_region
• report distribution

15
Query Evaluation Example

• protein distribution of Human NCS-1 homologue
• from wrapped CaBP website
• get the amino acid sequence for human NCS-1
• from wrapped Expasy website
• submit amino acid sequence, get ranked homologues
• at Mediator
• select homologues H found in rat, and homology gt
  0.70
• at Mediator
• for each h in H
• from previous view
• protein_distribution(rat, h, cerebellum,
  distribution)
• Construct result

a second integrated view
16
Implementation

• System
• Flora as F-Logic Engine
• Communicate with ODBC databases through
  underlying XSB Prolog
• XML wrapping and Web querying through XMAS, our
  XML query language and custom-built wrappers
• Data
• Human Brain Project sites
• NPACI Neuroscience Thrust sites

17
Work in Progress

• Architecture
• plug-in architecture for
• domain knowledge sources
• conceptual models from data sources
• Functionality
• better handling of large data
• operations
• expressive query language
• operators for domain knowledge manipulation
• query evaluation
• query optimization using domain knowledge
• Demonstration
• at VLDB 2000