.::. Edkal Business Solutions .::.

GLOBAL SITE |

home > solutions > life science >data mining

Overview

Industry Solutions

Life Science Solutions

	Data Mining
	Data Integration
	Knowledge Management
	Bioinformatics Services
	Edkal – Oracle Practice
	Our People

Health Care

Point Solutions

Emerging Technologies

Data Mining
The promise for bioinformatics is that public genome data, mixed with proprietary sequence data, clinical data from previous drug efforts and other stores of information, could unearth clues about possible candidates for future drugs. The challenge of extracting knowledge from the above data draws upon research in statistics, databases, pattern recognition, machine learning, data visualization, optimization, and high-performance computing, to deliver advanced business intelligence and web discovery solutions.

Edkal’s informatics team provides an array of services to analyze complex biological and chemical data and transform them to knowledge. The informatics team enables life science companies to improve productivity across the drug discovery R&D value chain by addressing certain key bottlenecks in the R&D. The problems and Edkal solution statements are addressed below:

Problem	Edkal Bio IT Solution
Biological Data is usually available in disparate and incompatible databases	Integration of all disparate data into a single database
Utilization of diverse public and privately available textual information and digital information	Extraction of Knowledge from text using powerful visualization tools
Require updates from disparate sources constantly	Automatic updates on relevant topics from various data sources
Generation of Knowledge requires comparison analysis and Correlation of all relevant information	Automatic correlation of in-house experimental data with World-Wide Knowledge

Image Mining
An increasing amount of experimental data comes in the form of digital images and mechanisms are needed to extract meaningful information from images that can be fed into data analysis tools. Edkal has developed medical and pathological image mining tools, to search confocal microscope images for biological objects, electron microscope images for nanoparticles and microscope images for cell differentiation.

Case Study
Manual examination of cancer images can get very tiring and time-consuming. Errors can result due to fatigue and subjectivity. The cancer detection algorithm developed by Edkal uses highly advanced technology to precisely locate cancerous cells based on a large number of cell characteristics

Text Mining - Chemical Structure Extractor
Medline has 14 million biographic units of information which is increasing at the rate of 40,000 every month*. This amount of knowledge must be able to contribute to future drug discovery and that is where text mining is now such an important technology. It would take five years to read the material that is now produced every 24 hours*. Not surprisingly the drug research industry is currently wasting $2.5 million* in retrieving existing data.

* As reported at UK –DTI Bioinformatics Conference (December 2005)

Edkal’s intelligent text mining tool helps Drug discovery chemists mine required information thereby significantly reducing time being spent downloading and reading published articles form journals

Discovery R&D chemists depend on published journal information (mostly in PDF format) to keep track of latest developments in new molecule/new chemical entity discoveries. Though all articles on new molecules are of interest to a chemist, it is important for chemists to look out for articles that contain information on those molecules that could potentially be lead candidates for the disease area that they are currently working on. Currently, chemists will have to look through all articles to decipher this information, or in another instance it could only be an article containing a brief about the chemical molecule of interest but has no structural information of the new molecule

Chemical structure extractor is a text mining tool used to extract structure information from downloaded PDF articles. The key features of this tool are:

Extracts structures form PDF articles
Structure formats compatible with ISIS and oracle databases
Query tool to search for structures of interest across PDF documents
Advanced query tool to do a substructure query and pull out structures that have the same substructure (for e.g. benzene would be a substructure and naphthalene could be a hit). This tool can also be used to query for reactions also seen in these PDF documents

Benefits

Targeted article search
Improved Productivity of chemists
Huge cost savings by subscribing only to articles of interest

Case Study
The chemical structure extractor is currently being used by Discovery chemists in a multinational pharmaceutical company’s research centre to extract structures from more than 150 articles downloaded from 18 different journals in a day.

^ Top

	Print this page
	Email this page