Conference on Semantics in Healthcare and Life Sciences (CSHALS)

Technical Talks

Updated February 24, 2011

Tech Talks showcase products and services of relevance to the CSHALS audience. Each Tech Talk is 10 minutes in length and designed to allow organizations to create awareness of new technologies, services, etc., in an informational presentation format.

For organizations interested in presenting a Tech Talk, please go to our Sponsor Opportunities page (click here) for further information.


Thursday, February 24, 2011 - Technical Talk 1
Time: 12:15 pm - 12:25 pm

Jans Aasman, Franz Inc.
Oakland, US

RDF Browser for Pharma Discovery and Visual Query Building


The free-form nature of RDF triplestores offers a lot of flexibility for constructing databases, but that freedom can also make it less obvious how to find arbitrary data for retrieval, error-checking, or general browsing. We will present a graphical triplestore browser that makes data retrieval more pleasant and powerful with a variety of tools for laying out cyclical graphs, displaying tables of properties, managing queries, and building SPARQL queries as visual diagrams.


Thursday, February 24, 2011 - Technical Talk 2
Time: 12:25 pm - 12:35 pm

Vishal Gupta, Elsevier
Ari Tuchman, Quantified

Presentation slides (.pdf)

SciVerse Platform Embraces Semantic Applications

With an exponential growth of information and wider distribution of services and data sources; integrated search and intelligent semantic discovery become crucial to the success of researchers.

SciVerse Applications is an innovative marketplace for applications that enhance the search capacity of researchers by adding semantic search functionality, text and data mining annotation to visualize data and relationships, or integrate data with ontologies and repositories. Developers and researchers can build customized applications that target specific researcher interests and workflows, and drive innovation on the SciVerse platform.

Using the SciVerse Framework, third party developers can build applications that appear alongside full text articles, abstracts and search results within the SciVerse product suite. Developers can also integrate external web services and APIs in their applications to mashup with the SciVerse APIs.

In this session we will showcase the SciVerse platform and demonstrate two applications driven by semantic technology and developed in collaboration with our partners at Stanford University and Quantifind. ODiSSea semantically expands your query with standard ontologies and public data resources. Quantifind extracts and aggregates data from the corpus associated with a user query and visualizes the associated data and trends.

For more information:

General information on Sciverse:  www.info.sciverse.com

SciVerse Applications:  www.applications.sciverse.com

SciVerse HUB:   www.hub.sciverse.com

SciVerse Developer Portal:  http://developer.sciverse.com


Friday, February 25, 2011 - Technical Talk 3
Time: 12:15 pm - 12:25 pm

Dexter Pratt
Selventa
Cambridge, US


Analysis of Omics Data Using Reverse Causal Reasoning (RCR) in an Integrated Analysis Environment

A critical challenge in modern biology is the development of informative mechanistic hypotheses from a large amount of information produced by today’s molecular profiling methods, such as microarray and next-generation sequencing methods. To address this challenge, SelventaTM utilizes the Genstruct® Technology Platform (GTP). In this talk, we present Reverse Causal Reasoning (RCR), one of the key analytic capabilities of the GTP, to translate large-scale data into meaningful, testable molecular mechanisms.

RCR is an automated reasoning technique that processes networks of causal relationships to formulate hypotheses, and then evaluates those hypotheses against data sets of differential measurements. RCR analysis attempts to answer the question “What signaling differences could lead to the observed differences in measured quantities?” The method is called “reverse” because the evaluation of each hypothesis effectively reasons from observed effects to identify potential causes. RCR is applicable to any molecular profiling data type, including microarray, RNA-Seq, proteomic/phosphoproteomic, and metabolomic data.

The output of RCR is a set of mechanistic hypotheses that represent upstream controllers of significant differential measurements, ranked by the calculation of statistical figures of merit for relevance and accuracy. These hypotheses can be assembled into causal chains and larger networks to interpret the data set at the level of interconnected mechanisms and processes.

Implementation of RCR uses a large knowledgebase encoded using BEL (Biological Expression Language) based on manual curation of the scientific literature. RCR methodology is available through a web-accessible investigation suite that includes facilities for managing experimental data, performing RCR and analyzing hypotheses generated by RCR.


Friday, February 25, 2011 - Technical Talk 4
Time: 12:25 pm - 12:35 pm

Chuck Rockey
IO Informatics
Berkeley, US


A 'Killer App' for Semantic Technologies:  Point-and-Click Data Integration Tools Make it Easy to Deliver Targeted Semantic Knowledge Bases

HCLS present unusually dynamic needs for data integration, application and search workflows. Semantic technologies are uniquely suited to provide game-changing integration tools for HCLS through their far more flexible methodologies for data integration and knowledge building.

Traditional data warehousing and federation approaches often choke on the heterogeneity of data sets involved in HCLS (e.g., multiple experimental data silos, public data sets, and clinical data as well as NLP results from scientific, competitive and patent literature.)  Even if projects enjoy initial successes, traditional solutions quickly start to break down as novel application requirements and data sets are added.

This talk will describe and demonstrate semantic methods that make it possible for domain experts as well as ontologists and informatics experts to quickly build, modify and extend integrated knowledge bases. Formal ontologies or application ontologies are assembled on the fly as part of the integration process - all without programming or hand-coding of RDF.   We’ll show you how, in 10 minutes.


[Top]