A Computer-Based Chemical Information System

Abstract

A considerable improvement in the power of computers in the 1960's prompted work at the National Institutes of Health (NIH) and at the headquarters of the Environmental Protection Agency (EPA) to explore the feasibility of an online chemical information system (1). Several capabilities were identified as necessary in such a system, including storage capacity, search and retrieval from libraries of chemical data, comnputer-assisted analysis of chemical data, and retrieval of information from the chemical literature. The types of numerical data under consideration included mass spectra, carbon-13 nuclear magnetic resonance (NMR) spectra, and x-ray diffraction data. Bibliographic data dealing with mass spectrometry and x-ray crystallography have also been incorporated into the system.In view of the well-known difficulties implicit in the use of batch-oriented computers in information retrieval, it was considered essential that an on-line interactive computer be used for these tasks and so the NIH PDP1O was chosen as the computer to be used in these experiments. A chemist who searches through a data base often possesses a good deal of information, perhaps poorly defined, that is related to the problem at hand. A Dr. Heller is a computer specialist in the Management and Information

Chemical toxicology searching: a collaborative evaluation, comparing information resources and searching techniques

“…Direct substructure searching facilities are also advantageous for rapid and precise searches of this sort, and were used by some participants to search RTECS and PHS-149 on the NIH-EPA system [12,13]. (The recently introduced CAS ONLINE and DARC substructure searching systems for the Chemical Abstracts bibliographic files were not available to participants at the time the searches were carried out).…”

Protein and Nucleic Acid Sequence Database Systems

“…More recently, information systems such as the Chemical Information System (CIS) (3,20) and the PROPHET system (2,29), as well as the on-line nucleic acid and protein databases (4,22,28), have been developed that BIOLOGICAL SEQUENCE DATABASES 421 allow the user to interactively interrogate, manipulate, and analyze the scientific data in the database.…”

Computer identification of mass spectra

“…Most systems do not compare aU of the peaks (many use < 10 peaks), usuaUy selecting the most abundant, The PEAKoption of the Mass Spectral Search System (Ref. 3) operates in a "conversational" mode to allow us er choice of peaks, Compilations are available (Ref. 4), however, so that matehing of this kind can be done without the aid of the computer, Even with our present reference file representing over 30,000 different compounds, it is possible that a mass spectrum can be uniquely defined using only a few peaks because of the unusually high information content of a mass spectrum (Ref.…”