Computers are having a tremendous impact on chemical
research and the teaching of chemistry. This in turn is fostering
advances in chemical information science and technology (chemical
informatics). Technological changes have caused a re-examination
of the program to train chemical information specialists at Indiana
University, resulting in the modification of existing courses
and the creation of new courses and degree options. Plans for
their implementation and for the establishment of joint industry-academic
research projects are described in this paper.
KEYWORDS: chemistry education research librarianship
chemical informatics information science and technology
Gary Wiggins holds a BA in Chemistry and Russian
and an MA in Slavic Languages and Literature, plus MLS and Ph.D.
degrees in Library and Information Science. Since 1976, he has
been the Head of the Indiana University Chemistry Library, Bloomington,
IN 47405. (firstname.lastname@example.org)
What skills should a librarian or information specialist possess for success in today's rapidly changing information environment? Recently the Special Libraries Association made available on the Web a document entitled "Competencies for Special Librarians of the 21st Century."1
The authors of the report distinguished between:
Some of the skills outlined in the report can only be gained and practiced on the job. However, among the 24 competencies in the SLA study are those that should be imparted to students in programs that train tomorrow's information professionals, for example:
- expert knowledge of the content of information resources, including the ability to critically evaluate and filter them
- specialized subject knowledge appropriate to the business of the organization or client
- effective communication skills
- the skill of working well with others in a team.
It is difficult to gain the competencies needed to be an effective chemistry librarian or chemical information specialist. The discipline of chemical information science and technology is involved with everything from traditional library science to modern computer-based techniques for the storage and retrieval of chemical information. Nowadays, its practitioners also deal with heavily technology-based areas such as molecular modeling and computational chemistry. In recognition of the broadened scope of chemical information science, the American Chemical Society recently enhanced the title of the Journal of Chemical Information and Computer Sciences with the phrase "Includes Chemical Computation and Molecular Modeling."
There is an ever-increasing impact of the computer on chemical information. A recent article in Chemical & Engineering News notes "the growing interconnectedness of laboratory research activities with molecular modeling, computational chemistry, chemical analysis, database searching, and general information retrieval."2 In chemical information science, the single most noteworthy achievement of the last few decades was the development of software for the coding and retrieval of chemical structures. This has resulted in ways of searching and associating factual information about chemistry that utilize the universal language of chemistry, the chemical structure. Research in this area has moved beyond the 2-D depiction of molecules into the much more complex arena of the storage and retrieval of 3-D representations of chemical structures. Thus, the interactions that chemical substances themselves undergo in the real 3-D world are beginning to be mapped into information systems. The special techniques of chemical information science and related disciplines are being brought to bear on complex chemical problems, resulting in advances on a scale unthinkable even a decade ago. Today's chemistry librarian/chemical information specialist must keep up with such developments and incorporate them into the techniques available to meet chemists' information needs.
In light of the increased complexity of chemical
information searching today, with ever more knowledge of chemistry
needed to use the tools effectively, it was felt that the implementation
of new interdisciplinary graduate programs in chemistry/library
and information science and in chemistry /computer science should
be explored at Indiana University. Dual master's programs exist
already within the IU School of Library and Information Science
(SLIS) for areas such as Music, Public Affairs, Journalism, Law,
Art History, etc. They are aimed at individuals who seek to become
practicing librarians or information specialists, and who desire
to enhance their subject skills considerably beyond the level
of the bachelors degree, which is the minimum requirement for
entry into the 36 semester-hour MLS degree program or the new
42 semester-hour Master of Information Science (MIS) program.3
THE MLS/CHEMICAL INFORMATION SPECIALIST PROGRAM
For many years, the Department of Chemistry at Indiana University has participated with the School of Library and Information Science in a master's level program specializing in chemical information. Since its inception in 1969, over 40 students have completed the program. Those graduates have gone on to responsible positions at libraries and information centers, primarily in industry. Students in the program are required to take three chemical information courses in the Department of Chemistry.
The introductory course, C400 Chemical Information Sources and Services, is a one semester-credit-hour course. Enrollments in the class, which is offered only in the fall semester, are large, often in excess of 100. Most of the students are upper-level undergraduate chemistry majors, with a few graduate students from the specialized SLIS master's program. (At present, there is no chemical information course devoted to chemistry graduate students, although all incoming chemistry graduate students in the C500 Introduction to Research course are encouraged to attend several library lectures during their first semester.)
The other one semester-hour classes have undergone considerable redefinition in the past few years. C401 Computer Sources for Chemical Information, as the name implies, concentrates exclusively on major computer-based tools. For example, during the Spring 1996 semester, three weeks were devoted to the Beilstein CrossFire system, and three weeks were spent on structure searching in the STN Registry File. Enrollments in that class tend to be in the range of 10-20 students. Finally, C402 Current Topics in Chemical Information, another one-hour class that can be repeated one time, offers students the opportunity to concentrate on one or two topics which are explored in depth. During the Spring 1996 semester, the impact of the computer on the primary chemical literature (especially journals) and personal or in-house databases were the two themes of the course.
A recent evaluation of the program revealed a need for restructuring certain aspects of the curriculum, in particular, to provide more training with computers and computer-based retrieval techniques.4 In response to the survey of graduates of the Chemical Information Specialist program, more emphasis is being given in the three existing one-hour chemical information courses to topics of particular interest to industry, for example, patent searching. Practicing information specialists from industry now teach the section on patents in two of the courses. Beginning in 1996, the Chemistry Library receives the CD-ROM fulltext versions of all US patents. It also participates in the new Questel/ORBIT Academic Search Plan and the STN International Academic Program, which currently offer 90 percent and 80 percent discounts respectively on searches in selected databases. The Chemistry Library continues to fund the Beilstein CrossFireplusReactions database and has recently extended that license to include the Gmelin database. Thus, sophisticated structure and other search approaches to the literature covering millions of organic, organometallic, and inorganic compounds back to the 18th century are now available 24 hours a day for a fixed cost. As other software and databases prove essential to the continued success of the chemical informatics program, they will be considered for funding on the appropriate library budgets.
The Indiana University School of Library and Information Science has developed many new courses aimed at broadening the scope of experience of SLIS graduates in the total spectrum of information science. These include human-computer interaction, strategic intelligence, and the electronic information environment, among others. In light of these developments, a re-examination of the place of chemical information science in the graduate curriculum at Indiana University was undertaken in 1995. As a result, plans are now well underway with both SLIS and the Department of Computer Science for innovative new programs to provide additional options for those interested in chemical information science and technology. Those are described later in this paper.
The Department of Chemistry at Indiana University is in an enviable position with respect to chemical information instruction. In addition to a state-of-the-art chemistry library, with two professionally trained librarians having in excess of 40 years of post-MLS experience between them, IU can build on the successful MLS/MIS program for chemical librarians and information specialists. Since 1992, IU has been the site of the Clearinghouse for Chemical Information Instructional Materials.5 This program, sponsored by the American Chemical Society Division of Chemical Information and the Special Libraries Association Chemistry Division, endeavors to provide easy access to a wide spectrum of teaching tools for the chemical information area. Effective Internet tools have also been developed for use in the IU teaching program. "Chemical Information Sources from Indiana University: CIS-IU" is a guide to the main sources of Internet and other tools for solving chemical information problems. 6 "Chemical Information Sources and Services" includes the syllabus and lecture notes for C400.7 These are linked on the Web to appropriate sections of CIS-IU. Experiences in using the Web to teach chemical information courses at IU are described in a paper delivered at an online conference in the summer of 1996.8
Given the long experience of teaching in the chemical
information area at Indiana University, difficulties commonly
cited by others faced with teaching chemical information courses
are not a problem at Indiana University. Those include such things
as lack of effective teaching exercises; unavailability of key
search tools, textbooks, and user aids; unqualified librarians;
lack of a subject specialist in chemistry; etc. However, the program
as it currently exists has reached a plateau, whereas the field
of chemical information science and technology has erupted with
activity in many exciting new technological areas of relevance
to chemical information specialists. Hence, the idea for enhancing
the existing program with a stronger emphasis on chemical information
science and technology was conceived.
NEED FOR AN ADVANCED CHEMICAL INFORMATICS PROGRAM
In "Educating the Next Generation Scientist," Helen Gillespie points out the need to introduce into the chemistry curriculum modules dealing with information technology and computer topics ranging from networking issues to distributed processing to interfacing.9 The application of computer-based tools to the problems of chemistry is being carried out most intensively by those in the emerging field of chemical informatics (also known as "chemical information science and technology" or "computer chemistry"). One need only look at the increasing frequency of coverage of computer-based topics in the major news journal of the American Chemical Society, Chemical & Engineering News, to recognize that a revolution is underway in chemistry.10 Chemical informatics melds techniques from computational chemistry, molecular modeling, chemical analysis, database searching, and information retrieval, among others.
Herman Skolnik, then editor of the Journal of Chemical Information and Computer Sciences, stated more than a decade ago, "In view of the fact that chemical information science is a discipline of chemistry, one in which several thousand chemists are shaping meaningful careers, we should expect some graduate schools to conduct research in this discipline, preparing students for potential careers as chemical information specialists." 11
As chemical knowledge itself grows in magnitude and complexity, chemists, chemistry students, and librarians or information specialists who utilize chemical informatics techniques must have better backgrounds in computer science and information science. Thus, a high level of skill in three areas is increasingly necessary, both to use existing computer-based chemical information systems effectively and to develop new ones. It is surprising that at the end of 1996, more than a decade after Skolnik's call for a graduate research program in chemical informatics, there is still no graduate academic program in the U.S. where one can receive extensive focused training in that field. Certainly there is no place in the U.S. where long-term academic research is carried out in this area. This is in stark contrast to the situation in Europe where one finds major graduate programs at Sheffield and Leeds in England and at Nuremberg (Erlangen) and Munich in Germany. This observation is validated if one looks at the origin of the major advances in the field of chemical informatics in the last two decades. In 1980, the American Chemical Society Division of Chemical Information presented its first Skolnik Award. Named in honor of Herman Skolnik, the original editor of the ACS's respected Journal of Chemical Information and Computer Sciences, the award has gone to nineteen individuals since its inception, twelve of them Europeans. In the last ten years, it has been given to only two Americans.
In September 1995, an inquiry to CHEMIND-L (the Chemical Structure Indexing Discussion List) provoked many responses to a statement that there is no further academic research to be done in the area of chemical structure information.12 The inquirer suggested that all necessary research and development in the field is now performed by commercial companies and that the major research problems have all been solved. Several Skolnik Award winners were among the respondents. Michael Lynch of the University of Sheffield agreed that many of the chemical structure research problems have been solved and the solutions embraced by commercial interests, but, he asked, "By whom were the problems first identified and their solutions tackled? In this area, with few exceptions, the problems were identified in academia, and the first evidence of solutions produced there too." Another Skolnik Award winner, Sheffield's Peter Willett, stated that the pressures faced by researchers in the commercial sector are often at variance with the long-term nature of much academic research. Citing the example of methods for indexing chemical reaction databases, Willett emphasized that it took 14 years of work at Sheffield to identify an approach to the comparison of chemical reactant and product structures that could enable automatic reaction indexing. Others mentioned as current research problems: Markush reaction searching, delivery of powerful tools to the desktop, management of polymer structure representations, standards for data portability, the application of new computational paradigms (e.g., neural networks, expert systems, genetic algorithms), chemical structure MIME, molecular diversity, hybrid QM/MM (quantum mechanics/molecular mechanics) methods, and density functional theory. If there are so many research topics in just this one area of concern in chemical informatics--chemical structure information--surely the field as a whole is ripe with research opportunities. Obviously, there is a need in this country for a center of excellence in chemical informatics, one that could ultimately show that joint industry/academic chemical informatics research is not only possible, but profitable in the USA.
As the Indiana University Department of Chemistry was earning an international reputation in chemical research, computer-based techniques made major contributions to that success. In no small measure, activities in chemical informatics played a significant role. Services provided by the IU Molecular Structure Center, Quantum Chemistry Program Exchange (QCPE), Serena Software (producer of PCModel), the Chemistry Library, and related activities continue to enhance the chemistry research and teaching efforts at Indiana University. However, in order to remain competitive in the 21st century, the IU Department of Chemistry recognizes that a more focused approach to these techniques is required. Therefore, new programs in the multidisciplinary area of chemical informatics are being developed. Indiana University will be in a unique position to develop partnerships in chemical informatics research when the programs described below are implemented. The university has considerable strengths in chemistry, computer science, and library/information science, but a coordinated effort is required to put together programs that will attract graduate students with interdisciplinary interests in these fields. The new programs will supplement the existing School of Library and Information Science MLS/MIS-Chemical Information Specialist program and will help satisfy a market for graduates in new computer areas that has found the pool of potential recruits in the United States very sparse.
PLANNING FOR THE NEW PROGRAMS
Discussions about the proposed curricular innovations have been ongoing among the units involved and within the IUB Department of Chemistry since 1994. Continued partnering by the IU academic units and industry will ensure the success of the new curricula. As a measure of the interest in the field, a "Workshop on Chemistry, Computers, and Information" was held on the Bloomington campus on November 11, 1995, with participants from the interested units and attendees from both Bloomington and Indianapolis. The enthusiastic reception by the workshop participants alone would have provided ample motivation for creating the new programs. However, to test our assumptions, comments were solicited from selected people outside Indiana University in the fall of 1996. Listed below are excerpts from a few of the more than two dozen responses received.
"As information delivery technology develops,
computer science becomes an increasingly important part of the
information gathering and analysis process. Knowledge of the strengths
and weaknesses of the technology is of paramount importance for
developing our next generation of chemists."
--Director of Marketing and New Product Development of a major database producer
"Expertise in chemical information and computational
aspects of chemical structure handling is widely scattered in
the United States, with virtually no universities having critical
masses of people with such expertise."
--Professor of Chemistry and author of a major chemical information textbook
"Cutting-edge research in the area of chemical
information is not only exciting, but badly needed. In particular,
the full utilization of combinatorial chemistry is already challenging
traditional chemical sub-structure searching systems."
--Director of the Computer-Assisted Molecular Design Department at a major pharmaceutical company
During the planning of the new chemical informatics curricula, opinions were also sought in 1996 about the topics to be included in the new chemical informatics courses through questions directed at several relevant listserves and newsgroups. By this and other means, solid suggestions for the development of the programs were received from outside experts in the field. One of the foremost authorities in chemical information was a Visiting Scholar at Indiana University for 2 1/2 months in 1995/96. He taught several sections of the existing chemical information courses and consulted about the directions the new programs should take. In addition, the current editor of the Journal of Chemical Information and Computer Sciences visited the Department of Chemistry in October 1996 to present a seminar and to offer advice about the new programs. Building the program with input from such respected outside experts (a number of whom have agreed to affiliate with the program) will ensure its speedy acceptance by potential students, industrial research partners, and funding agencies.
By late 1996, the draft of topics for the new Chemistry graduate chemical informatics courses had been formulated, and the new course proposals written. In order to staff the new courses, an interdisciplinary team of educators and researchers was surveyed for interest and willingness to teach in the program. The response was quite positive.
The planning for the new programs is also drawing
on the experience gained by the IU Chemistry Library in forming
the interdisciplinary team that worked on a large multi-year biotechnology
informatics project. That three-year project was part of a $3
million U.S. Office of Education grant conducted by four midwestern
universities.13 The development of effective links
with colleagues in other geographic locations and the forging
of partnerships among diverse units at Indiana University during
the biotechnology project provided very relevant experience for
the new chemical informatics initiative. Although the USOE biotechnology
grant expired on October 1, 1996, work on the project is being
continued with outside funding. The biotechnology team will serve
as the basis for a complementary bioinformatics program being
created at IU.
THE PROPOSED CHEMICAL INFORMATICS PROGRAMS
As the anchor piece for the chemical informatics programs, the IU Department of Chemistry is defining two new three semester-hour graduate-level courses in chemical information science and technology. In order to attract a corps of graduate students to do research in this field, the department is working toward creation of a chemical informatics minor for the Chemistry Ph.D. and two new master's degree options for chemists in conjunction with the Department of Computer Science and the School of Library and Information Science. This will provide an environment where students from different disciplines can interact in productive interdisciplinary teaching and research endeavors. All students in the new programs will be required to work with chemistry research and teaching groups and/or to participate in externally funded chemical informatics projects. Procedures for approval of the new courses and degree programs have been initiated in the relevant departments, and it is expected that some aspects of the new programs will be in place during the 1997/98 academic year. When finalized, graduate students will have the following choices:
The minor would include the two new graduate courses. Eventually, it is hoped that this will grow into a major course of study in the Departmentof Chemistry. A major requires 18 semester hours in a given area of chemistry.
Students who enter with a minimum of a bachelor's degree in chemistry may choose the existing Chemical Information Specialist program. However, they may also opt for the dual master's degree in chemistry once the new joint SLIS-Chemistry program is in place.
This requires a minimum of a bachelor's degree in chemistry and is analagous to the existing SLIS MLS/MIS Chemical Information Specialist program. Over time, this may evolve into a dual master's program for students who prefer to concentrate on the computer side of chemical informatics.
A. The New Minor Courses in Chemical Informatics (C591-C592) and the Chemistry Ph.D. Minor.
The Chemistry Ph.D. student who elects to minor in chemical informatics would take 18 hours of chemistry in the major subject and 6 hours of chemical informatics courses. It is likely that the dissertation research project of such a student would draw heavily on the skills learned in the minor courses. As noted above, it is hoped that an interdisciplinary Ph.D. program in chemical informatics will evolve at Indiana University.
The two new Chemistry courses will cover topics such as:
This is a broad range of topics, and it is impossible
for one faculty member to do justice to all of them. Therefore,
faculty with relevant expertise from the three academic units
on the Bloomington campus will participate in team-teaching the
courses. Although most of the topics are at least touched on in
existing courses, the emphasis in the new courses will be on the
chemical information aspects of the topics. In some cases, visiting
faculty with particular expertise who could add perspectives from
the chemical/pharmaceutical industry or the chemical database/software
industries will be invited to participate in teaching. It is anticipated
that students in all three of the new programs, as well as some
students who will be enrolled in the existing MLS/MIS program
would take these courses. Furthermore, since the courses are unique
in graduate curricula in the United States, it is likely that
a market could be found for them in the distance education arena.
B. SLIS/Department of Chemistry Joint Master's Degree (60 or 66 hours).
Including the two chemical informatics courses, students in the Chemistry/SLIS dual master's programs will be expected to take:
The requirement for a chemistry master's thesis would
be waived, as it is in most of the other existing joint programs
which SLIS has with departments in the College of Arts and Sciences
and with other schools at IU. (Note that, at the present time,
the requirement of 24 hours of graduate chemistry courses is the
same number required for chemistry Ph.D. students at Indiana University.)
C. Computer Science Master's in Chemical Information Systems (30-33 hours).
In addition to the two chemical informatics courses (6 hours), students in the Computer Science master's program will be expected to take:
As noted above, students who enter this program must have the minimum requirement of a bachelor's degree in chemistry.
A cadre of students with interests and skills in
chemical informatics, who work closely with existing chemistry
research and teaching groups in solving their information problems,
will enhance the chemistry research and teaching efforts at IU.
By bringing together faculty and students from the three distinct
units-the School of Library and Information Science, Computer
Science, and Chemistry-interesting interdisciplinary research
projects can be conducted. These research activities will foster
links between IU and corporations in the chemical/pharmaceutical
and software/database sectors that have significant interest in
chemical informatics. Complementing the existing SLIS MLS/MIS
degrees and the proposed bioinformatics program, the new chemical
informatics programs will produce graduates who are uniquely equipped
to solve the complex, critical information problems of the chemical
and pharmaceutical industries.