Wednesday, August 19, 2015
10:00 AM - 10:45 AM
|Level: ||Technical - Intermediate|
The rapidly increasing amount of data in pharma industry provides new opportunities and challenges for large scale data mining. Apart from quantity, the inherent diversity and heterogeneity (e.g., data formats, information models and terminologies) of the data are significant barriers. Semantic Web (SW) standards have been shown to be successful in such environments to create coherence and integrate information “meaningfully” in a way that the data can be contextually understood, accurately interpreted and made actionable. Applications of SW technologies in pharma have matured to the point that we are beginning to see tangible benefits.
This talk will explore several real-world use cases relevant to pharmaceutical industry in applying SW for understanding molecular basis of complex diseases, conceptual integration of pharmaceutical databases and competitive intelligence. We will demonstrate that potentially powerful applications can be built by combining Semantic knowledge graphs with network analytics, natural language processing and inference reasoning methods.
We will discuss:
- Practical use cases of applying Semantic Web standards and technologies in Pharma
- How RDF and OWL can be effectively used to describe and build non-trivial relationships between different data sets that makes it easier to correlate and semantically integrate
- How complex queries can be easily formulated in SPARQL query language that can help to find new actionable insights from semantically aggregated RDF graphs
- How simple inference and reasoning capabilities empower discovering new knowledge and intelligence within the RDF graph
- How network analytics and natural language processing (NLP) are critical in supporting RDF graph based knowledge discovery
Ranga Chandra Gudivada PhD is a manager for Enterprise Medical Informatics at University of Pittsburgh Medical Center. At UPMC, he manages the semantic interoperability platform by harmonizing disparate discreet/non-discreet data from various clinical domains and systems. He also oversees efforts around unstructured text mining of clinical notes to actionable knowledge. In previous roles, as a scientist in big pharma, he applied Semantic Web Technologies for various problems in drug discovery and translational genomics. Dr. Gudivada earned his PhD in Biomedical Informatics from University of Cincinnati and work experiences include working at Eli Lilly and Corning Incorporated before joining UPMC.