Teaching the Use of CAS ONLINE at Indiana University

by Gary Wiggins

11 June 1985

Online bibliographic searching of the Chemical Abstracts database has been available in the IU Chemistry Library since 1973. However, until recently the person who wanted to search the database had to do so through a "search intermediary"--a librarian. The costs of an online search are generally variable in most libraries and are usually assessed to the user. Since Chemical Abstracts is one of the more expensive databases to search, it was a rare user in our library who paid less than $25 for a search. In fact, it was not unusual for the cost to be triple that amount. That situation changed in July of 1984 when the Indiana University Chemistry Library became a participant in Chemical Abstracts Service's Academic Program.

Under the original Academic Program, all variable search fees were abandoned. In their place was a $500 monthly fee. The only restriction was that the database could not be searched during CAS's peak operating hours (originally, 8:00 a.m. to 2:00 p.m. eastern time). It did not take long for academic libraries around the world to realize what a bargain this was. Soon Chemical Abstract Service found that just over two dozen academic libraries, which numerically constituted about 3 percent of the users, were burning up 1/3 of their computer's CPU time. Obviously, the program could not be sustained if it continued to grow. Thus, by January 1, 1985, a modified Academic Program was offered.

The new plan provides a grant of 90 percent of the online search costs to those academic institutions which subscribe to the printed Chemical Abstracts. The discount covers all hourly charges, search fees, online display costs, telecommunications charges, and offline prints. However, the search must be done on weekdays between 5:00 p.m. and 8:00 a.m. eastern time or between 8:00 a.m. and 1:00 p.m. Saturdays. There is no limit to the number of CAS Academic Program accounts which can be held by a given college or university. A minimum deposit of $200 is required, and each new Academic Program account may have a maximum of five passwords. Under the old program, only one person could search at a time since only one password was provided.

When Chemical Abstracts Service announced its Academic Pro-gram late in 1983, I saw an opportunity to broaden the scope of our online training efforts at Indiana University without losing control of the costs. Despite the fact that only one password was available under the original Academic Program, I believed that we could get multiple uses from it: training in our chemical information courses, online searching by chemistry and other librarians, and online searching by end-users. Therefore, many hours were spent in convincing both the library administration and chemists at I.U. that a unique opportunity was at hand. Despite the relatively abbreviated access hours, the academic program offered us for the first time a chance to search the full CA file and to know the cost before beginning each search. In a very real sense, therefore we had the capability of subscribing to an online version of Chemical Abstracts. By the end of June 1984, I had permission from the library administration to treat the twelve-month cost of the academic program as a subscription item and pay the entire $6,000 from our materials budget. Thus, no searches performed on the academic program password by undergraduates, graduate students, faculty, librarians or others would necessitate any additional charges to the user, provided no offline printing were involved.

Funds were taken from the Chemistry Library book budget to pay for the experiment. Furthermore, it was decided to train as many people as possible to do their own searches and to charge nothing for the searches. After the first eleven months of service, the program can be viewed as a huge success. Over 150 people have been trained to search the databases. One can judge the enthusiasm with which they have embraced online searching by the fact that 66 different people have had at least one search session between January 1 and May 25, 1985. Bear in mind that we paid only $6,000 for twelve months of searching on the old Academic Program. However, the true cost of the service expressed in terms of the normal rates in the first four months of 1985 alone was $33,558.94. Over 238 hours of computer connect time were used in only the first two months of that period on the Chemical Abstracts Bibliographic File (which corresponds to the printed CA) and over 40 hours on the Registry File (which is the structure searching component).

The Registry File is one of the truly great achievements in online searching. Anyone who has tried to search for a chemical compound using chemical nomenclature is aware of the complexities of such a search. The non-chemist is startled to learn that Glycerol, Glycerin, Osmoglyn, and 1,2,3-Propanetriol are all the same compound. With the Registry system, however, the user can in effect draw a picture of the compound and instruct the computer in Columbus, Ohio to search through a file of over seven million compounds to identify it or find compounds which are structurally related.

The answer which comes back is a picture of the compound and other information about it, including up to 50 variant names. The compound thus identified can then be searched in the Biblio-graphic File to identify methods of preparing it, physical properties, toxicity, uses, and other information about the substance. Once appropriate references have been found among the 6 1/2 million dating from 1967, a printout may be obtained which may even include the abstracts for citations entered into the database after 1975.

Our experience has shown that a user can be trained to do his own searches on the system in a lecture/demonstration session of about two hours each for the Bibliographic and Registry Files. While they may not have all of the skills of an experienced librarian or chemical information specialist, chemists clearly want to do their own searching. Following the training session, which consists of a one-hour lecture, followed by a one-hour demonstration, each participant has the opportunity to perform a guided bibliographic or structure search on a topic of his choice. After the search sessions, participants can sign up to do searches either on the library microcomputer or in a location outside the library.

Very little special instructional materials were developed locally to train the end-users. Instead, selected pages of handouts and training manuals provided by Chemical Abstracts Service were reproduced and given to each trainee. In addition, a simple log-on instruction sheet for the library microcomputer was developed to assist those who use library equipment.

Our experience at Indiana University with online instruction for end-users has been limited, but exciting to date. In some ways, the use of the CAS ONLINE files in our training programs represents a significant improvement. Obviously a better feel for an online search of a huge database can be had by searching a file of over 6 million references rather than a practice file of 100,000 citations which is available on some other systems. But there are other problems which will not go away. Despite the availability of some very good printed aids and textbooks on searching the Chemical Abstracts database, there is no good textbook available to aid in teaching the full range of computer based chemical information services which are available. Chemical Abstracts is not the only database which chemists should search. A recent article by Donald Hawkins lists at least 50 different databases relevant for chemistry.1

Librarians and teachers are faced with the continuing problem of acquiring or keeping in good repair adequate equipment to maintain an educational program. Despite the availability of computers and terminals in many offices and laboratories in our department, there is still a great demand placed on the library equipment by our newly trained end-users. Most of the searches performed by end-users this fiscal year have in fact been on our equipment.

During the remaining few days under our old Academic Program account, we plan to make heavy use of the CAS Online system in our training programs. We are doing this because the single most important consideration in training people is online search cost. The CAS Academic Program allowed us for the first time to get a handle on those costs. Under the new Academic Program, our end-users will have to be checked out and qualified for online searching much as is done now for those using NMR and other equipment in the Department of Chemistry. The meter will be running after July 1, and even with a 90 percent discount, our new searchers must develop cost-effective search habits.

At last year's St. Louis ACS Meeting I asked several of the vendors who offer online searching of Chemical Abstracts why they too do not make available to academic institutions the same type of reduced rate searching of the full files which CAS now has. The answer I consistently got was that they would very much like to do this, but that they are bound, by virtue of the CAS license agreement, to pay royalties to CAS for each use of a CA record which is not part of their educational files. Hence, in order to be sure of covering the potential royalties, they could not begin to meet the 90 percent discount which CAS allows for the new Academic Program option. A number of the vendors expressed the opinion that this is unfair competition from Chemical Abstracts Service.

There are many academic searchers who have devoted years to developing searching skills on vendor search systems other than STN International, the system for searching CAS ONLINE. Now the searchers find those vendors not only unable to offer the same type of attractive rates to academic subscribers as can STN International, but also not able to provide access to all of the data which Chemical Abstracts Service has on STN International. Yet other vendors and indeed other database producers have made efforts to support academic searchers in their training efforts. Last August, I surveyed the five major vendors who lease the Chemical Abstracts files for online searching in the United States. DIALOG, ORBIT, and BRS all have practice files of CA available for training purposes. DIALOG and ORBIT offer North American users instructional passwords for $15/hour including telecommunications costs. However, BRS charges $8/hour for CAST plus telecommunications charges which vary with the network used. The BRS program is offered to schools of library or information science, but other bona-fide training programs in information science may be appproved on an individual basis. Neither Questel norPergamon InfoLine offers practice files of CA at the present time, but the latter indicated that they are in the early stages of developing an educational package of reduced rates for academic institutions which want to use the Pergamon InfoLine system.

Another possibility which is beginning to become a reality is computer-assisted instruction for Chemical Abstracts searching. The Department of Information Studies at the University of Sheffield has developed two such packages, SIMCHEM and ESACHEM for learning the implementations of CA on Data-Star and the European Space Agency's Information Retrieval Services. Dr. Bert Ramsay of Eastern Michigan University is very interested in this approach. Computer-assisted instruction does offer some solutions to the problems of teaching end-users to do online searching. However, with the number of vendors, not to mention the number of databases, which are relevant for chemists, it will be a long time in the future before any type of instructional modules for some databases become available to teach the users to do their own searching. The Education Committee of the ACS Division of Chemical Information is presently developing training modules for searching Chemical Abstracts by computer. Two versions will be available by September--one for the Dialog and one for the STN International systems. Those interested in receiving copies may contact Arleen Somerville at the University of Rochester Library. Arleen also can supply you with a 23-page bibliography on chemical literature teaching aids which includes some material for online searching. On the road to competency on certain databases and vendor search systems, I hope that the users will be able to maintain an awareness of the vast array of computer-based information systems which they have at their disposal. Although some controls have to be instituted under the new Chemical Abstracts Service Academic Program in order to hold down the costs, the program still provides a substantial and much-needed discount to academic institutions. Other database producers and vendors should be encouraged to follow their lead.

Bibliography

(1)Hawkins, Donald T., "A Review of Online Physical Sciences and Mathematics Databases. Part 2: Chemistry." Database v. 8 no. 2 (June 1985): 31-41.

SAMPLE CAS ONLINE SEARCH

FOR INFORMATION ON THE TOXICITY OF GLYCEROL

The search is first performed in the Registry File, where a match is made. The answer is assigned the set number "l1". It is shown below.

RN 56-81-5

IN 1,2,3-Propanetriol (9CI)

MN Glycerol

SY Glycerin

SY Glycerine

SY Glyceritol

SY Glycyl alcohol

SY Trihydroxypropane

SY Glycerol (8CI)

SY Propanetriol (7CI)

SY Glyrol

SY Glysanin

SY Osmoglyn

SY 1,2,3-Trihydroxypropane

AR 30918-77-5

MF C3 H8 O3

CI COM, TSCA;

DR 78630-16-7, 71463-73-5, 75398-78-6, 8013-25-0, 30049-52-6,

29796-42-7, 37228-54-9;

OH

.

.

HOCH2CHCH2OH

The switch is made to the Bibliographic File as follows:

=> file ca

FILE 'CA' ENTERED AT 13:20:52 ON 26 FEB 85

COPYRIGHT 1985 BY THE AMERICAN CHEMICAL SOCIETY

Below is shown the strategy to search for information on the toxicity of glycerol, limiting the answers to those which are in English and not published as patents.

=> s l1 and toxic?

12719 L1

86512 TOXIC?

L2 123 L1 AND TOXIC?

=> s l2 and eng/la not p/dt

3776262 ENG/LA

1022823 P/DT L3 73 L2 AND ENG/LA NOT P/DT

The most recent answer in the database is displayed below with full bibliographic information and abstract.

=> d l3 bib abs 1

ANSWER 1

AN CA102(3):19118r

TI Toxic and osmotic effects of glycerol on human granulocytes

AU Armitage, W. John; Mazur, Peter

CS Oak Ridge Grad. Sch. Biomed. Sci., Univ. Tennessee

LO Oak Ridge, TN 37830, USA

SO Am. J. Physiol., 247(5, Pt. 1), C382-C389

SC 4-3 (Toxicology)

DT J

CO AJPHAP

IS 0002-9513

PY 1984

LA Eng

AB The addn. of glycerol [56-81-5] to granulocytes and its subsequent

diln. under various conditions were investigated to try to

distinguish between toxic and harmful osmotic effects of glycerol.

The lesion caused by glycerol at 0.degree. was expressed as a loss of

plasma membrane integrity (as visualized by fluorescein diacetate)

only after incubation (.gtoreq.1 h) at 37.degree.. This damage was

not ameliorated when osmotic stress was lessened by reducing the

rates of addn. and diln. of glycerol to keep the computed cell vol.

within 80-170% of isotonic cell vol. However, when osmotic stress

was reduced further by increasing the temp. of addn. and diln. of

glycerol from 0.degree. to 22.degree., the tolerance of the cells to

1M glycerol increased somewhat. Reducing exposure to glycerol to

.ltoreq.3 min at 0.degree. greatly increased survival, but this time

was too short to allow glycerol to equilibrate intracellularly.

Finally, the presence of extra impermeant solute (NaCl or sucrose) in

the medium to reduce the equil. cell vol. to 60% of isotonic cell

vol. enabled granulocytes to survive 30-min exposure to 1M glycerol

at 0.degree., but cells had to remain shrunken during the 37.degree.

incubation to prevent the loss of membrane integrity. Suspensions

that contained damaged granulocytes formed aggregates when incubated

at 37.degree., and these aggregates were responsible for a major

fraction of the obsd. loss in viability.

Note that the abstract contains computer coded symbols which are required for the print product. For comparison, the same abstract from the printed Chemical Abstracts is reproduced below.