Biology Reference
In-Depth Information
also suggested the best approach for further experiments where results
were inconclusive. This program was followed by another, more sophis-
ticated program that allowed for experimental errors and assessed the
reliability of the fi nished sequence.
16
Dayhoff's choice of journal for
publishing this work—the
Journal of Theoretical Biology
—suggests
that she saw it as making a contribution to the organization and sys-
tematization of biological knowledge.
At about this time, in the early 1960s, Dayhoff began to collect com-
plete protein sequences. Her reasons were twofold. First, protein se-
quences were important in their own right, since they contained the key
information about how biology worked. Second, and perhaps more im-
portantly for Dayhoff, proteins contained information about evolution-
ary history. At the same time that Dayhoff was beginning her collection
efforts, Linus Pauling and Emile Zuckerkandl were developing a new
method of studying evolution, and the relationships between organisms,
using protein sequences as “documents of evolutionary history.”
17
Day-
hoff and others saw that such work would require both collections of
proteins and computer programs to perform the computationally in-
tensive tasks of sequence comparison and phylogenetic tree construc-
tion. Dayhoff and her colleagues at the NBRF scoured the published
literature for experimentally determined protein sequences and entered
those sequences on punched cards for computer processing. Although
the collection itself was a nontrivial task, it was never Dayhoff's ulti-
mate aim to be a botanist of sequences: “There is a tremendous amount
of information regarding evolutionary history and biochemical function
implicit in each sequence,” she wrote to a colleague, “and the number
of known sequences is growing explosively. We feel it is important to
collect this signifi cant information, correlate it into a unifi ed whole and
interpret it.”
18
Collection was a means to an end.
The fi rst edition of the
Atlas of Protein Sequence and Structure
,
published in 1965, listed some seventy sequences. Subsequent editions
contained not only the protein sequences themselves but also extensive
analyses performed by computer. These analyses included studies of the
evolution of specifi c protein families, the development of a model of
evolutionary change in proteins, an analysis of patterns in amino acid
alleles, simulation of protein evolution, and studies of abnormal hu-
man hemoglobins, ribosomal RNA, enzyme activity sites, and transfer
RNA.
19
The
Atlas
also provided phylogenetic trees and protein second-
ary (three-dimensional) structures. In the preface to the third edition
of the
Atlas
, Dayhoff and her collaborator Richard Eck outlined their
approach to the sequence collection problem:
