|
||
|
||
Home >
Money > Business Headlines > Report June 6, 2002 | 1257 IST |
Feedback
|
|
IBM Labs working on gene projectShweta Rajpal Kohli IBM India Research Labs is working on a bioinformatics project to develop next-generation gene prediction algorithms. This is part of the various projects that the lab has taken up in bioinformatics and life sciences, the solutions of which can impact the quality of life. "To be able to use gene sequence information meaningfully, scientists need to identify the genes and the functions they serve," Manoj Kumar, director of IBM India Research Labs, said. The project, in collaboration with the department of biotechnology, is concentrating on the use of statistical and data mining tools (using the wealth of annotated genomic data available in public databases) to boost the accuracy of gene prediction by learning which algorithms perform the best in diverse scenarios. Another project taken up by the research centre, located at the Indian Institute of Technology campus at New Delhi, is to develop a comprehensive biochip information system to meet the challenges and opportunities being presented by microarrays today. According to IBM, the biochip information system will help users narrow down potential leads and accelerate development of new therapeutics by enabling development of the full spectrum of specialized analytics and bioinformatics tools required for making biological knowledge discoveries from biochip data. "Solutions such as these can help the biologists focus on their research by expediting access to and analysis of prior relevant research data which is stored in many different locations, each employing a different format," said Kumar. "Today, users are restricted in leveraging the information from these hundreds of data sources because they cannot access the data in terms of concepts from the biological domain that they are familiar with. The solution that we are building will help in providing single-query access to all databases, and will optimise search across heterogeneous data to speed information gathering," said Kumar. The research centre is also working towards automating the process of creating, updating and verifying annotations of medical documents. The goal of this research is to apply natural language processing techniques to identify relevant biological entities, discover relations between them and extract important biological information from the vast biomedical literature. The extracted information will be used in creating functional annotations for medical and biological terms to help in medical research, genomics and drug discovery. "Our goal is to apply natural language processing techniques to identify relevant biological entities, discover relations between them and extract important biological information. The extracted information can then be used in annotation tasks," said Kumar. ALSO READ:
|
ADVERTISEMENT |