ARTIFICIAL INTELLIGENCE SPECIAL SESSION ON "SEMANTIC TECHNOLOGY APPLICATIONS IN LIFE SCIENCE AND BIOINFORMATICS"

The Artificial Intelligence Special Session on "Semantic Technology Applications in Life Science and Biotechnology" is organized by MIMOS BERHAD during the Knowledge Technology Week 2012 (KTW2012).

This special session is a focused session featuring talks of leading experts in the field and demo applications on real-world problems. A panel discussion is designed to provide a great opportunity for senior members of government departments, ministries and industrial players who are directly related to life science and biotechnology sector, to address the topic with the right depth, giving it proper visibility, and allowing time for discussion and networking.


Tab Box

Speaker Profile

Dr. Sheng-Chuan Wu

Vice President
Corporate Development, Franz Inc., USA

Dr. Sheng-Chuan Wu received a Ph.D. degree in Scientific Computing and Computer Graphics from Cornell University in the U.S. He is currently the Vice President of Corporate Development at Franz Inc. (Franz is a leading technology and tool provider for Artificial Intelligence and Semantic Web, located in Silicon Valley). He has managed and consulted on many Semantic Technology projects in the US, China, Korea, India and Malaysia, and has routinely lectured on AI, RAD software development and Semantic Technology at conferences, and has conducted more than 20 week-long Semantic Technology workshops in Asia. Previously, Dr. Sheng-Chuan was the Vice President of Marketing and Customer Support at ATP, an integrated CAD/CAM/CAE software company in the US.

Dr. Amandeep S. Sidhu

Dr. Sidhu is working as a Research Fellow at Curtin Sarawak Research Institute (CSRI) at Curtin University, Sarawak. His research has resulted in over 50 publications that have been cited over 250 times till date with an average of 23.22 cites per year. He is also working as a Senior Lecturer at Department of Electrical and Computer Engineering (ECE) at Curtin Sarawak. He is the editor for Data, Semantics and Cloud Computing series of Springer-Verlag. He is also the Executive Committee Member of Asia Pacific Bioinformatics Network since 2006.He areas of expertise include High-Throughput Computing, Real-Time Data Streams, Semantic Web, Data Management, Bioinformatics and E-Health.

Presentation and Demo (I)

Presentation

Presenter: Sheng-Chuan Wu

Title: Semantic Technology - The Only Sensible Knowledge Integration Tool for Life Science Research

Abstract: Life Science data is typically stored in disconnected databases with different taxonomy and data schema, making it difficult to characterize and reference them. This lack of data uniformity severely affects drug discovery, system biology, and personalized medicine, all of which rely heavily on integrating and interpreting data sets produced by different experimental methods at different levels of granularity. Furthermore, life-science researchers need to constantly reference extant biological knowledge accumulated at many institutions to draw critical insight into the target biological processes. In this talk, Dr. Wu will explain how Semantic Technology provides the critical interoperability for the data and harmonization of knowledge to enable researchers to convert "reference" knowledge into "executable" knowledge, greatly strengthening life-science Translational Research. That's why many research institutes (such as NCBI, NCI, etc.) under NIH in the US have encoded their data with semantic technology. Dr. Wu will also demonstrate how easily semantic technology combines 11 public datasets to enable integrated search and query. He will also show how easy it is to use a semantic technology based tool to identify bio-markers for hepatotoxicants, and to discover and qualify the relationships among chemical compounds (e.g., drugs), genes and diseases.

1st Demo

Title: Integration of 11 Medical Science Knowledge Bases For Semantic Search

Abstract: This demo integrates 11 medical science knowledge, which entail 5 public life-science data sets: Drugbank, Dailymeds, Sider, Diseasome, and Clinical trials. The first 4 publicly available knowledge bases have already been semantically encoded in RDF. The clinical trial database contains more than 81,000 funded clinical trials result summaries. We apply entity extraction technique to the clinical trial data to create 5 Semantic databases: CT-discusses-drug, CT-discusses-side-effect, CT-discusses-target, CT-discusses-disease and CT-mentions-genes. Finally, we add a bridging ontology to mediate different taxonomies. With all the "data" in Semantic form, integration is simply loading all into a semantic database. We can now ask complex questions like "Find all clinical trials that resemble clinical trial NCT00130091 given diseases, drugs, targets, and side-effects", and "Find the title of all clinical trials that discuss the drug Lipitor and the side-effect Type-2 Diabetes" easily.

Screenshots:

2nd Demo

Title: Predictive Toxicology, Identifying Bio-Marker for Hepatotoxicants

Abstract: Panels of several hepatotoxicants administered to rats in groups of 4, using placebo, low, mid, high in single dosage. Then study effect at 6, 24 and 48 hours with metabolomic analysis of liver, serum and urine for 1,603 metabolites, plus microarray analysis of liver and whole blood for 31,096 transcript gene probes. The objective is to identify biomarkers for such liver toxins. Semantic technology makes such identification easy and straight forward, by mapping metabolomic analysis and microarray analysis results into semantic form in semantic database. Other public life science knowledge, such as pathway ontology, can then be loaded into the semantic database to enable integrated search for the bio-markers.

Screenshots:

3rd Demo

Title: Discover Relationship among Chemical Compounds, Genes and Diseases

Abstract: This demo shows how an NLP tool can extract key entities and their relationships (e.g., <:Sirolimus :inhibits :MTORGene> and <:RenalFailureChronicle :associateGene :MTORGene>) from the voluminous life science publications, and load them into Semantic database to enable complex search and queries.

Screenshots:

Presentation and Demo (II)

Presentation

Presenter: Amandeep Singh Sidhu

Title: Designing Structured Vocabularies for Biomedicine

Abstract: With the data explosion caused by various technologies today in biomedicine there is a great need for effective automated information-gathering and information-inference tools. Current biomedical informatics techniques addressed issues of data management and prediction, thereby reducing the economic burden of earlier techniques. These days numerous biomedical databases and analysis tools are independently administered in geographically distinct locations, lending them almost ideally to adoption of an integrated intelligent data management approach. This is especially needed as in the current ICT environments data sources enter and leave the systems. As result of emerging demands of huge amounts of biomedical data, new and improved data integration capabilities are required for supporting a wide range of scenarios including: data mining, data delivery to integrated data stores, data consolidation and inter-enterprise data sharing.
This session discusses need for a biomedical information infrastructure framework with data integration capabilities at the core to ensure increasing agility in various stakeholders and organizations involved to exploit biomedical data. In this brief session methodologies and tools to design structured vocabularies and ontologies for biomedicine are discussed:
1. Introduction to Structured Vocabulary, Ontologies and Semantic Web Tools
2. Strategies and tools to design biomedical vocabularies and ontologies from heterogeneous structured, semi-structured and unstructured data sources
3. Submission and visualization of the ontologies to NCBO BioPortal
4. Do's and Don'ts of building Biomedical Vocabularies



1st Demo

Title: Top Braid Composer

Abstract: Details to be announced.

Screenshots: Details to be announced.


2nd Demo

Title: Altova SemanticWorks

Abstract: Details to be announced.

Screenshots: Details to be announced.


3rd Demo

Title: NCBO BioPortal

Abstract: Details to be announced.

Screenshots: Details to be announced.


Organization Profile


Franz Inc.


Franz Inc was founded in 1984 by alumni from UC Berkeley. It is a leading vendor of Semantic Technology tools featuring AllegroGraph RDF Semantic Database and RDF Prolog inference engine, and Artificial Intelligence (AI) development tools including Allegro CL/CLOS Object System and AllegroCache Object Database. Based in Oakland, California near Silicon Valley, Franz Inc delivers leading-edge development products that enable software developers to build flexible, scalable, web-enabled production semantic applications quickly and cost-effectively



Curtin Sarawak Research Institute (CSRI)

Curtin Sarawak Research Institute (CSRI) is a new institute that is located at Curtin University Sarawak Campus in Miri. The initial funds for the establishment of CSRI have been assisted by Curtin University, Western Australia, and the Government of Malaysia.

CSRI has been established to promote research excellence in the energy and biotechnology areas. A strong emphasis of CSRI is on multidisciplinary and collaborative research to develop holistic solutions encompassing scientific, technological, social and economic aspects. CSRI promotes partnership between campuses of Curtin University and external partners including academics and industry.

For further details please contact:
Farouq H Hamed (MIMOS) / Norbaitiah Ambiah (MIMOS)
Tel: +60 3 8995 5000 Ext 57085
Fax: +60 3 8991 4212
Email: aiss@mimos.my