370 CATALYZING INQUIRY
been used, and yet joint authorship or due credit withheld. In one story, a respected biologist held
regular discussions with an excellent mathematician colleague. The biologist assigned a theoretical
project on a topic already worked out by the mathematician to an in-house physics postdoctoral
fellow instead of pursuing joint work with the mathematician. The results of this in-house work fell
short of what was possible or desirable but displaced other serious attempts at theoretical analysis of
an interesting problem.
This example suggests a view of mathematics and computer science that is ancillary and peripheral
to the “real” substance of biology. The fact that computing and mathematics have developed powerful
tools for the analysis of biological data makes it easy for biologists to see the computer scientist as the
data equivalent of a lab technician. However, although programming is an essential dimension of most
computer scientists’ backgrounds, it does not follow that the primary utility of the computer scientist is
to do programming. Algorithm design, to take one example, is not programming, but because algo-
rithms must be implemented as a computer program, it is easy to confuse the two.
In other cases known to the committee, the expertise of biological scientists has been denigrated by
those in computing. For example, computer scientists sometimes view a successful biological experi-
ment as one that “merely” produces more data and do not appreciate the fundamental creative act
required to devise the appropriate experiment. This attitude suggests a view of biology in which the
“real” science resides in the creation of a theory or a computational model, and data are merely what is
needed to populate the model.
What accounts for such attitudes? The committee believes one contributing factor is not much
different than loyalty to one’s discipline. Professionals in one discipline quite naturally come to believe
that the ways in which they have learned to see their discipline have inherent advantages (if they did
not, they would not be part of the discipline), and challenges to the intellectual paradigms they bring to
the subject may well be met with a certain skepticism.
A second point to consider is that interdisciplinary work is not necessarily symmetric. This is
especially true in the mix of academic research activity vis a vis applied or technical support activity.
That is, it is often possible to identify one field as being the side where research advances are occurring
and the other as applying some kind of support. In some cases, Ph.D.-level research in computer science
can be enriched by what is routinely taught in undergraduate classes in biology, and vice versa.
For example, individuals pursuing cutting-edge research in database design may be interested in
finding data models to exercise their design. They are interested in finding domain experts to help them
better understand the complexities of a certain interesting problem domain, such as biology, but these
database researchers see the data and the insights coming from the biologist as helping to define the
problem, but as having little to do with finding the solution. Similarly, biologists may be investigating
a new topic in biology and need quantitative or logistical or algorithmic help to accomplish the research,
but they feel the real intellectual contribution—to biology—comes from their insights on the biological
side.
The primary exception to these scenarios is where a research group in computer science gets teamed
up with a research group in biological science. In such instances, the relationship can be truly symmetri-
cal. Both parties benefit from a symbiotic relationship. Both yield practical value to the other, while
gaining theoretical value for themselves. Both operate at an equivalent level of intellectual contribution.
Both gain an equivalent level of real research coming out of the activity.
10.3.2.5 Attitudinal Issues
Biology laboratories are increasingly dependent on various forms of information technology. High-
throughput instrumentation generates large volumes of data very quickly. Computer-based databases
are the only way to keep track of a biological literature that is growing at exponential rates. Computer
programs are increasingly needed to assemble and understand biological data derived from experi-
ments or resident in databases.