Sunday 19th May '13 You are here: Home > Research > Sansone Group > Projects > Student Projects 2012
Personal tools

Student Projects 2012

The Sansone Group at the Oxford e-Research Centre offers several student projects throughout the year, focused on developing and applying new methods and tools for improving management, curation and sharing of data in life science.

Supervisors: Susanna-Assunta Sansone; Philippe Rocca-Serra and Eamonn Maguire.

 

PROJECT 1. Graph Based Bio-experiments Store

Abstract
A NoSQL version of the database component, part of the ISA software suite, used by a growing number of researchers and service providers community in the life science domains (Sansone et al., Nature Genetics, 2012). A natural solution would be to look at Neo4J, a graph based (learn the basics of graph based DBs here) NoSQL data store which is highly optimised for querying graph structures. It would also allow for easy translation to  Resource Description Framework (RDF) directly from database. 

Workplan outline:
1. Literature review
1. Review of the current BII database which uses hibernate and conventional relational databases and the BIIObjectModel. 
2. Assessment of Neo4J.
3. Implementation of persistence of the BIIObjectModel in to a Neo4J instance.
4. Investigatation of creation of a scalable web application/modification of one of the current BII web application (for an example, the one based at the European Bioinformatics Institute) which will sit on top of this Neo4J resource and query/present its content.
5. A further step would be to make the graph semantically aware, so queries on the graph, where nodes are URIs, would make available the information in those URIs, provided the content type is RDF for instance.
Additional links:
Learn the basics of graph based DBs here
Neo4J Tutorial is available here.
 

Suitable for:

 

PROJECT 2. Ontology Based Validation of Bio-experimental Records

Abstract

Add semantics validation to the relevant component of the ISA software suite, used by a growing number of researchers and service providers community in the life science domains (Sansone et al., Nature Genetics, 2012). The ISAvalidator component  is a syntax checking software developed to support ISA grammar. However, semantic validation is weak and error detection capability basic. We want to explore ontology-based reasoning approach for validation and develop a consistency checker against an existing ontology. The use case will be limited to experiments with chemical exposure and compound-based intervention studies, in order to develop a proof of concept.

Workplan outline:
1. Literature review and state of the art
2. Conversion of experimental test cases to Resource Description Framework (RDF)/Web Ontology Language (OWL) formats
3. Evaluation of reasoner performances
4. Definition of validation rules on top of existing logical constraints as defined in the ontology
5. Integration in user interface
 
Additional links:
 

Suitable for:


PROJECT 3. Design Driven Acquisition of Bio-experiments

Abstract

Enhance the data acquisition components of the ISA software suite, used by a growing number of researchers and service providers community in the life science domains (Sansone et al., Nature Genetics, 2012).  Develop a user interface exploiting the principles of design of biological/medical experiments (factorial design, parallel group design, cross over design), sample collection plan and sample testing applied to field trial and toxic exposure experiments. The goal is to increase consistency in experimental data reporting, awareness of experimental design and assist planning steps of experimental protocols.

 
Workplan outline:
1. Literature review and state of the art
2. Decomposition of classic experimental design in terms of intervals
3. Evaluation of performances
4. Integration in user interface
 
Additional links:
Allen’s algebra to deal with treatment timelines and sample collection timelines.

 

Suitable for: