LBNL Speaker Series in Washington to Feature CRD Expertise

Beginning in February, scientists from CRD will launch a series of presentations at Berkeley Lab’s project office in Washington, D.C. The goal of the series is to better inform the Washington research community about the achievements and expertise of LNBL staff. The LBNL office is located at 901 D Street, SW, Suite 950. The office is in the Aerospace Center, across D Street from L’Enfant Plaza.

Wes Bethel, leader of the Visualization Group in CRD, will give the first talk at 9 a.m. Thursday, Feb. 17. His presentation is entitled, “Finding the Unknown in a Sea of Data: Leveraging Human Intuition with Scientific Visual Data Analysis.”

In his talk, Bethel will describe several approaches to help researchers tackle the problem of detecting patterns in their increasingly complex data and possibly even finding the unexpected. One example is ProteinShop, a software application used for predicting the 3D shape of proteins from amino acid sequences. ProteinShop takes advantage of human intuition to accelerate the protein structure prediction process, thereby reducing the time required from weeks to hours. Another example is a combination of visualization technologies and data management to produce what is known as "query-driven" visualization. The main idea is to perform intelligent, user-guided selection to find and analyze only the "interesting data," rather than a brute force method to perform analysis on the entire dataset.

The March presentation will feature Victor Markowitz talking about the Biological Data Management and Technology Center (BDMTC). Markowitz’ talk will be at 9 a.m. Friday, March 11. Before rejoining the Lab in January 2004, Markowitz was CIO and Senior VP, Data Management Systems at Gene Logic, where he was responsible for the development and deployment of the data management and analysis platform for the company's gene expression data. Prior to joining Gene Logic in 1997, he was a staff scientist at LBNL, where he led the development of data management tools applied to biological databases.

The BDMTC is based on the premise that effectively addressing biological data management challenges requires consolidating data management and system development expertise in a central core. Biological data management involves data generation and acquisition, data modeling, data integration and data analysis. Data management challenges are posed by increasing amounts of experimental data generated by life science applications, difficulty of qualifying data generated using inherently imprecise tools and techniques, and complexity of integrating data residing in diverse and poorly correlated repositories. Biological data management systems in academic settings such as LBNL are developed with minimal or no user and system analysis, without following system development practices, and without considering system evolution, maintenance and scalability.

This presentation will discuss the main challenges of biological data management, the problems encountered by academic groups in addressing these challenges, the rationale for BDMTC, its activities to date and future plans.