close

Вход

Забыли?

вход по аккаунту

?

Book Review Three-Dimensional Chemical Similarity Searching. By C. Pepperrell

код для вставкиСкачать
BOOKS
Searching Questions
Guardians of Science. Fairness and
Reliability of Peer Review. By H.-D.
Dunirl. VCH Verlagsgesellschaft,
Weinheiin/VCH Publishers, New
York. 1993. 118 pp., hardcover
DM 78.00, $ 50.00.-ISBN
3-52729041-911-56081 -751-8
The subtitle “Fairness and Reliability
of Peer Review” well describes what this
book is about. The author, H.-D. Daniel,
is one of Germany’s
leading experts in
this area and is thus
well qualified to examine this question.
The data on which
the study is based
relate to the refereeing of one year’s
output of .4ngeiiwirl/e
Chcwic~.
Manuscripts submitted for “Angewandte” are usually reviewed by two independent referees whose
recommendations are then evaluated by
the editor. To that extent the procedure is
essentially the same as that normally applied to proposals submitted to the German Research Association (Deutsche
Forschungsgemeinschaft, DFG), and
thus the results are also relevant to the
situation there. The author analyzes all
the observations according to the rules of
the art of statistics. He begins by trying to
quantify the degree of agreement between
the two specialist reviewers. If the recommendations ;ire classified on a scale of
1 to 4. ranging from acceptance without
change through two levels of revision to
rejection. the statistical agreement seems
surprisingly poor. However, if the recommendations are treated as being in agree-
(
-
Thib section contains book reviews and a list of
new b o t i k b receiced by theeditor. Book reviews are
written by invitation from the editor. Suggestions
for hooks to be reviewed and for book reviewers
a r e welcome. Publishers should send brochures or
(hettei-) hooks t o Dr. Ralf Baumann. Redaktion
Angewindte <‘hmmie. Postfach 10 I 161, D-69451
Weinheini. Federal Republic orGermony. Theeditor rcscrves the right of selecting which books will
be revieaed Uninvited books not chosen for
revieh \$ill not he returned.
ment when they differ by only one point,
the degree of consensus turns out to be
satisfactory. To examine the “correctness” of decisions to reject, Daniel uses an
interesting approach: he traces the fate of
manuscripts that have been rejected and
subsequently published elsewhere, and
compares their frequency of citation with
that of papers appearing in “Angewandte”. As expected, the frequencies for
the former are appreciably lower than
those for the latter. Even Daniel does not
know whether this might be simply because papers in “Angewandte” are generally cited more often (the journal has a
high “impact factor”), and therefore one
cannot conclude very much about the individual papers from this analysis. However, in the absence of any better approach this is certainly a useful piece of
analysis. The book contains many other
statistical analyses which are worth studying by everyone who is concerned with
refereeing, either actively or passively, in
other words every scientist engaged in research or university teaching.
There are two aspects to refereeing procedures. First there is the “correctness” of
the decisions. As an insider I can confirm
that, at least in the case of German referees, recommendations are made in good
faith and to one’s best ability. Secondly,
however, there must be confidence within
the scientific community that this is so.
Since both the journals and the D F G preserve the anonymity of referees so far as
possible (for very good reasons), great importance attaches to books and articles
such as this which show, at least statistically, that the procedures are above reproach. However, I was somewhat surprised that Daniel gives no hint of
recognition that, in the nature of things.
every referee report must contain an element of subjectivity. All attempts to correct this by quantitative measures merely
shift the problem to somewhere else.
Daniel restricts himself to aspects that can
be analyzed quantitatively. Thus he takes
no account of the message that may be
read “between the lines”. We learn nothing here about the reviewer who writes a
damning report but cannot bring himself
to recommend rejection, nor about the
one who gives a report that is intended to
be favorable but qualifies it with sarcastic
comments. In such cases the editor needs
to be able to weigh up the situation and to
know the peculiarities of his reviewers. I
am disturbed by the many cases that are
cited of manuscripts that were initially rejected (by a wide variety ofjournals, mostly well-known) but were later honored by
Nobel prize awards. Unfortunately. refereeing systems probably suffer from a certain degree of conservatism. and have difficulty in dealing with anything which
conflicts too much with current thinking---even when, as is usually the case, all
those involved make great efforts to avoid
this and value originality above all else.
In such an important matter as refereeing it is not surprising that there have already been many attempts at improving
the system. One chapter lists every such
proposal that I am aware of. and analyzes
the experiences that have resulted from
these in various places. For me personally
this was the most interesting part of the
book, showing as it does that there is
probably no alternative to continuing
with the present system. Here the ethical
guidelines laid down by the American
Chemical Society, which are reproduced
in the book. deserve to be observed generally.
To complete this scholarly work Daniel
includes a comprehensive bibliography of
published work on the subject of refereeing together with extensive notes. The
whole work is written in fluent English:
11 8 pages that are well worth reading.
M a nf ied Mahnig
Deutsche Forschungsgemeinschaft
Bonn (FRG)
Three-Dimensional Chemical Similarity Searching. By C. Pepperrell.
Research Studies Press, Taunton,
UK, 1994. XV, 304pp., hardcover
E 57.50.--ISBN 0-86380-145-5
For well over a century chemists have
relied on the two-dimensional structural
formula for the display and communication of chemical information. This mode
of representation of structure has certainly been the traditional mainstay for the
BOOKS
storage, manipulation. and retrieval of
structural information in computerized
systems. The Cheniicul Abstructs database
alone comprises more than ten million
such structures. The widespread and continuing use of the two-dimensional formula is, of course, a legacy of the past-it was
bequeathed to us by pioneers such as
Kekule and Crum Brown who developed
these formulas in the 1860s and then went
on to show how they could be applied in
the rationalization of chemical phenomena. Although this representation has
served us well. the question now being
raised is whether it has outlived its usefulness. Two-dimensional formulas are really
quite unsuitable for computerized storage
systems as the formulas need to be redrawn in the same format and orientation
before structures can be compared. The
comparing of structures and assessment
of similarities has become very important
during the past decade, following on from
the realization that molecular similarity
studies can be profitably exploited in the
early stages of molecular design. The design of pharmaceutical drugs and agrochemicals in particular has benefited. In
addition, a host of other modern chemical
problems involve investigation of the stereochemistry or chirality of molecular
species. It is thus hardly surprising that
chemists are now beginning to take a serious look at the representation of chemical
species by three-dimensional formulas.
The book reviewed here may be regarded as an important and innovative step in
this direction. It comes from the information studies group at the University of
Sheffield, arguably the world’s leading
group in the development of new approaches to the management of three-dimensional structural data. The author has
chosen to focus on the techniques that
may be used for searching three-dimensional structural databases, with special
emphasis on similarity searching. After
pointing out the limitations of two-dimensional searching, the author analyzes the
role and effectiveness of four different
three-dimensional similarity methods,
namely the distance distribution method,
the use of individual distances, atom mapping, and the maximal common substructure method. It is concluded that in overall terms the best of these is the atom
mapping method, an approach based on a
localized description of the three-dimensional environment of each atom in a molecule. When two molecules are to be compared, atoms with similar local environments are mapped on to each other. and
the similarities between atoms so mapped
are combined together to yield a global
measure of the similarity of the two mole360
(‘
cules. The rest of the book. including a
30-page appendix giving the results of
sample calculations, is devoted to a discussion of the atom mapping method.
Among other things, the author considers
how the method may be optimized and
made faster in practice, and describes a
variety of elaborations that broaden its
scope and applicability. In fact, with
appropriate modifications, the method is
applicable not only to relatively small
molecules but also to the largest of macromolecules.
In presenting all this material in one
volume the author has performed a very
useful service to many different kinds of
chemists. Structure-directed searching is
now the most common type of query
made to chemical structural databases
and the demand is likely to grow considerably in the future. This work not only addresses the increasingly pressing problem
of the storage and manipulation of threedimensional data but also shows how similarity notions may be successfully incorporated into the searching of such data.
Even though in its current form the
method cannot cope with very large databases, it is certainly able to handle medium-sized databases containing up to
50000 structures. Moreover. research at
present under way in several laboratories
around the world will undoubtedly push
up this limit substantially in the foreseeable future. The atom mapping method
has many exciting potential applications
and will feature in the design of molecular
species for a variety of applications. The
author is to be congratulated on bringing
out such a useful and succinct guide to this
rapidly evolving domain.
Dennis H . Rouvray
University of Georgia
Athens, G A (USA)
The World of Physical Chemistry. By
K . J. Laidler. Oxford University
Press, Oxford, 1993. 488 pp., hardcover E 55.00.-ISBN 0-19-855597-0
The title of this book should really have
been extended by adding the subtitle “A
Historical Discourse”. It contains a history of ideas in physical chemistry. from the
beginnings when this classification did not
yet exist, up to about the middle of the
present century. To make for lighter reading the author intersperses the account
with biographical notes about actors who
have played important roles on this stage,
although some of these have had little
recognition. Some additional short biographies are collected together in an ap-
VCH ~,rlf1g,\ge.\ellr~/7ufr
mhH, 0-694.71 Wiwhrim, lYY5
0.77O-U833;95;0303-03611B 10.00
pendix, and there are plenty of literature
references.
Three introductory chapters explain the
background to the development of physical chemistry within the overall framework of the sciences, and discuss the nature of the problems that it addresses and
the methods of working. The latter are
illustrated by taking as examples two
scientists with very different styles. H.
Eyring and R. Norrish, and by outlining
the development of the most important
centers of physical chemistry research in
England, Scotland, and North America.
One chapter is concerned with the sources
from which early physicochemical studies
originated, and describes the beginnings
of the journals that are now the leaders in
this field. The heart of the book consists of
the seven chapters outlining the development of the sub-disciplines of thermodynamics, the kinetic and statistical theory
of matter, chemical spectroscopy, electrochemistry, chemical kinetics, colloid and
interface chemistry, and quantum chemistry.
The author’s treatment is firmly linked
to the individual scientists and their contributions to the development of the subject. In the course of this he also mentions
some little known facts. For example, I
had not appreciated how wide-ranging
was the work of D. L. Chapman, who not
only contributed to the theory of the diffuse double layer by developing an early
forerunner of the Debye-Huckel theory of
electrolyte solutions, but also made important contributions to the understanding of reactions in the gas phase. Even
more surprising is the reappraisal of the
work of Agnes Pockel who, as a housewife
in Braunschweig, carried out studies in
her kitchen on the behavior of films of
organic substances, the results of which
were published thanks to the support of
Lord Rayleigh. In her experiments she
used an apparatus that already contained
the essential elements of the Langmuir
balance for measuring the surface tension
of a film.
The reader will find that many recent
developments are mentioned only very
briefly or are even completely absent.
However. to include all these would have
made a larger book than was intended.
The treatment is undoubtedly much colored by personal background, and an author from outside the English-speaking
world would have given a different emphasis in many respects. However, that in
itself adds to one’s interest in reading this
well-written book. In an age when the
principles of physical chemistry are nearly
always treated in a deductive way in textbooks. and the student is seldom told
+. 3 / 0
Angew. Chem. Int. Ed. Enyl. 1995, 34, N o . 3
Документ
Категория
Без категории
Просмотров
0
Размер файла
310 Кб
Теги
pepperrell, dimensions, chemical, similarity, book, three, searching, review
1/--страниц
Пожаловаться на содержимое документа