IBM Gives NIH Chemical Database

IBM announced plans to give the National Institutes of Health a database of more than 2.4 million chemical compounds. NIH will add this information to PubChem, a freely available database of chemical structures of small organic molecules and information on their biological activities. IBM pulled the chemical data from about 4.7 million patents and 11 million biomedical journal abstracts over the period of 1976 to 2000. According to the firm, the newly compiled data will help researchers more easily elucidate important relationships among chemical compounds and will aid drug discovery. IBM—which created the database in collaboration with AstraZeneca, Bristol-Myers Squibb, DuPont, and Pfizer—extracted the data using techniques such as automatic image analysis and enhanced optical recognition of chemical images and symbols.