вход по аккаунту


Organic Reactions Classified by Neural Networks Michael Additions FriedelЦCrafts Alkylations by Alkenes and Related Reactions.

код для вставкиСкачать
Ro\ri. _I Moll. M. Spieles. G Bach. S Dihne, J. Kriwaneh. M. Lish.
.I P r d r < ' / i w i . 1995. 337. 203-208: b) A. Pawlik. U. De Rossi. S. Dihne.
unpublished rcsults
T V Drache\,i. V. I . Novodcrezhkin. A. P. Razjivin. Chcnr. PIirs. 1995, IY4.
223 2 3 5 , W Kiihlbraiidt. Nomi-c 1995. 374. 497V498.
I Goto. 7: Kondo. ,4ii!y!v. C/iein. 1991. i03. 17-33: A y q r i l . C/IIW./ i l l . Ed
E q / 1991. .?(I. 17 13.
E , Ci. McRac. M . Kdsha i n Phi..sicii/ P ~ o c iir RudfrrrkJfi ~ i o i o g r(Eds:
L Augeiistcln R Mason. B Rosenberg). Academic Press, New York. 1964.
p. 23.
E (>.McRac.. M Kasha. J C / i ~ v ~P /i I. I . J1958.
28. 721 --722.
.I K. Maurus. G Bird. J P/ij.\ C%nn. 1972. 76. 2982 2986
E. Daltroi/cr Ci Scheihe. K . Gschuind. F. Haimerl. P h o r a , ~Sci. Dig 1974.
a ) U. De
I S . 141 450
F.D S x \ a .
(; R Olin. J A i m C/rcw. Soc. 1977, 99. 4848-4850.
K. R. Wilsoii R. E. Pincock. Con. J (%em 1977. 53. 889-894.
S. F. Mason i i i <'fr<.ii/ur D i d i i o i m Priiicip/e.s oi7d Ap/dic(lri~~ii\
(Eds: K . Nahanishi. N. Bero\a. R . W. Woody). VCH. Weinheim. 1994. p. 48.
S. Kit-stein. H Mohwald. J. C h m Plirs. 1995. 103. 826-833.
V Bu\\. ..Iup,ik. < ' / i ~ i 1991.
103. 889-890: Aligerl.. Chrrli /ii/.Ed. Eiig/. 1991.
models this feature of the human brain.['] In this network the
basic processing units, the artificial neurons. are arranged in two
dimensions and thus "maps" are obtained of the analyzed information.
Clearly, an enormous range of variations exists for organic
reactions. and a map reflecting this broad spectrum of possibilities would be large indeed. To illustrate the potential of o u r
method we will therefore limit the discussion to a set of reactions
having an important feature in common: the reaction center. the
set of atoms and bonds directly involved in the bond rearrangement during reaction. The reaction scheme chosen is the addition of a C-H bond to a C = C bond (Table 1 ) and includes
such important reaction types as Michael additions, FriedelCrafts alkylations by alkenes, and free radical additions to alkenes.
30. X69 x70
D G,irgiulo. F Dcrguini. N. Bcrovn. K. Nakanishi. N. Harada. J. ,4iii. Chrw
So <.1991. I l l . 7046-7047
I.Tinoco. R d i t i i Hrs. 1963. 20. 133 - 139
Table 1 Electronic variables for chardcteriLing the reaction centers of the reactions
discussed [a].
Electronic variables
1 2
Lingran Chen and Johann Gasteiger*
Organic reactions are influenced by many factors: the structures of the starting materials, reagents, and catalysts as well as
reaction conditions such as temperature, solvent, pressure, and
light. Each of these factors can be considered as a separate
coordinate spanning a multidimensional space, and in this sense
a chemical reaction is an event in this type of multidimensional
space. Chemists have largely gained their knowledge on organic
reactions from observations on a series of reactions; common
and differentiating features were sought in order to assign reactions to classes. which were frequently named after the principal
discovers, for example Wittig reaction, Michael addition, and
Beckmann rearrangement. Clearly. such a one-dimensional
classification scheme can only insufficiently account for the variety of observations on chemical reactions.
We will show here that a set of chemical reactions can be
projected by a self-organizing neural network into a two-dimensional map. A reaction consists of a point in such a map, and the
distance between two points reflects how similar two reactions
are. Different types of similarities between reactions can be represented by projection into different directions of the twodimensional map.
The human brain generates two-dimensional maps in the visual. auditory, and somatosensory cortex from corresponding
sensory information obtained from the environment. The selforganizing neural network method developed by Kohonenl']
Pi-of .I Gasteigcr. Dr. 1.Chen
lnstitut fur Organische Chemie der Universitiit Erlangen-Nurnberg
Niigelsbachstiasse 25. D-91052 Erlangen (Germany)
Fax Int. code +(9131)85-6566
e-inail. gasteigcr,<
This Hark u:i\ \upported by the Alexander-von-Humboldt-Stiftung (scholarship to L C . ) We fhank MDL Information Systems providing us with ISlS
Host iind. togethcr with Fachinform~itionszentrumChemie, Berlin. for giving
HCCCS\ 10 the ChernInform-RX reaction data base.
+ H-C-C--C
Organic Reactions Classified by
Neural Networks: Michael Additions,
Friedel- Crafts Alkylations by Alkenes,
and Related Reactions**
+ H-C
[a] For definitions of the symbols refer to the text
A set of 120 reactions was obtained by a search with the
in the 1991 version of the
reaction retrieval system ISIS
ChemInform-RX data
As the changes in the structures
of the starting materials are the most decisive influences on a
chemical reaction, we concentrated in this investigation only on
these structural influences. The question is then. how should the
structures be coded? Clearly, lists of functional groups around
the reaction center cannot be the best method. because this list
would be quite extensive. Actually, chemists have already tackled the problem of having to compare diverse functional groups
and have generalized the influence of functional groups with
concepts such as partial charges and inductive/field. resonance,
and polarizability effects. Methods for the empirical calculations of these effects developed in our
were used to
calculate the influence of functional groups on the atoms of the
reaction center. Specifically, G-and n-electronegativities, 1, and
xn,were considered at atoms C-1 and C-3, the total charges. yi,,,
on atoms C-2 and C-3, and the effective polarizability, r*i, on
C-3 (see Table 1 ) .
These seven variables were used to describe each individual
reaction of the data set and used as input to the Kohonen neural
network. This network projects a seven-dimensional space into
a map consisting of 12 x 12 neurons. Each neuron has as many
weights as there are input variables for each object; in our case
there are seven. A reaction s will be mapped into that neuron c
that has weights w,,most similar to the input variables .Y,~of the
reaction considered [Eq. (a)]. After each input of a reaction the
weights of all neurons are adjusted such as to make them more
similar to the input variables. However, this adjustment is
largest for the winning neuron c and decreases with increasing
distance of a neuron from this central neuron. Reactions that
have similar electronic variables will thus be mapped into the
same or adjacent neurons. In our case quite a few neurons obtained several reactions (up to five), indicating a high degree of
similarity for those reactions. A number of neurons (78) did not
obtain any reaction at all.
To visualize the results of the self-organization of the 120
reactions in the Kohonen network, the reactions were inspected
by chemists and assigned to appropriate reaction types, which
were identified by symbols (Table 2). The neurons were labeled
with these symbols to indicate which reaction was projected into
which neuron. The results are shown in Figure 1.
Table 2. Assigned reaction types. associated symbols (see Fig. I ) . and the number
of reaction instances
Reaction typc
No. of
Reaction indcx
Michael addition
1 - 14. 24 -84
misassigned reaction center
Friedel -Crafts alkylation by alkenes 18
15-71. 86 92. 115. 116
electron transfer reaction
photochemical addition of a n acyl
radical to n n electron-poor olefin
photoinduced fl~lkylation
95 -97. 101
H' transfer reaction
98 - 100
reaction with special mechanism
hydride abstraction reaction
103 105
Nazarov reaction
106- 110
allylation of an 0x0 ester and
subsequent anti-Markovnikov
addition of HBr[a]
pdiladiuiii-catalyzed hexatrienolate
condensation reaction
photoinitiated radical addition to
crown ethers
118 120
Quite a few neurons obtained several reactions, which were
always of the same type, indicating the power of the Kohonen
network in perceiving similarities of different reactions. Only
one neuron, neuron (1 2,l) stored conflicting information: it
contained both a condensation reaction and a reaction that had
been coded in the data base with a wrong reaction center.
Michael additions comprised the bulk of the reactions, and
consequently the Kohonen network reserved the largest area for
this reaction type. The more specialized reaction types and those
with only a few members were pushed to the edges of the map.
It is quite remarkable that the reaction types assigned by a
chemist were also perceived by a Kohonen network, which used
the specified physicochemical variables of the reaction center as
criteria for assigning reactions; in the Kohonen network the
individual reactions of one type were collected in the same region of the map.
Even within the area of one reaction type the site in which a
reaction is located contains a lot of chemical information. Figure 2 shows the part of the map of Figure 1 containing Michael
[a] Two-step reaction,
Fig 2. Detailed analysis of thr cluster of Michael additions in the Kohonen feature
map i n Fig. I . The neurons marked wwth 22 and 3 2 were mapped by Michael
additions 111 which the rcncting H C bonds are activated by two or three strongly groups. respectively. The neurons marked with * indicate
Michael additions in which the reacting C=C bonds are activated by two strongly
clectron-withdrawing groups The neurons marked with S were mapped by special
Michael additions.
F~~ 1 . Kohonen feature map obtained for thc classification of 120 rc:iction\. White
empty neurons: boxes with a n x denote conflict neurons The symbols are defined in Table 2.
additions. Additional labels give further information on the individual reactions. Reactions that have only one (no label), two
(2Z), o r three (32) electron-withdrawing groups at the reacting
H-C bond are quite well separated. By the same token, reactions that have two strongly electron-withdrawing groups (*) at
the C=C bond are separated from those that have only one (no
label). An important advantage of a two-dimensional map is
that structural variations at both reacting bonds, the C-H and
the C=C bonds. can be indicated simultaneously. Furthermore,
also within a reaction type such as Michael additions the more
special examples (S) are found at the edges of the map.
We will now discuss some of the special Michael additions in
more detail. Scheme 1 shows the reactions mapped into neurons
(6,1), (3,7), (3.12). and (12.12). Whereas in most Michael addi-
bond is activated by groups
exert a - effect'
the three reactions mapped into neuron (6,l) show that the
(6, 1):
R = Et, iPr, iBu
fact, all three of these reactions were mapped into empty neurons at the edges of the map.
The two-dimensional Kohonen feature maps can show the
relationships of the chemical reactions under investigation,
point out the major reaction types, present their scopes, and
indicate also unusual reactions. Thus, they allow the chemist to
order observations on chemical reactions in an intuitively more
appealing and chemically more significant manner.
Receibed: October 6. 1995 [ZX4501E]
German version: Arigeii. Chivii. 1996. IOX. 844- 846
Keywords: Kohonen maps Michael additions
works . reaction classification
R = Me, Ph
( 3 , 12):
(12. 12):
Scheme I . Michael additions mapped into neurons (6.1), (3.7). (3.12) and (12,12).
The bonds denoted h) dashed lines are those broken or made in the reaction.
carbanion initiating a Michael addition can also be stabilized by
three groups exerting a - I effect.['' The two reactions in neuron
(3,7) are quite unique: the CH, group reacting is activated by an
ester group that exerts its influence through conjugation across
a double bond.'"' A CH, group also reacts in the reaction
stored in neuron (3.12), but it is activated by an ortho-nitro
substituent on the phenyl group.["] Clearly, a common feature
in the last two reactions is that a CH, group is activated by an
electron-withdrawing group in conjugation; however, this effect
is transmitted across different x systems. Thus, it is gratifying
that these two special Michael additions end up in similar regions of the map (neuron (3,7) and (3,12)).
The reaction i n neuron (12,12) is the only Michael addition""] in this set of reactions in which a H -CSpbond reacts; all
other cases involve a H -CSp3bond. This reaction thus extends
the scope of Michael additions and is stored at the borderline of
the area of this reaction type.
One of the exciting features of the Kohonen network is that
in classifying the reactions. the network automatically gains
chemical knowledge from those reaction instances. This.
"trained" network can then be used to predict reaction types of
the unknown reactions. To illustrate this point. the 120 reactions
of the previous data set were divided into two groups. The
60 reactions with odd indices (see Table 2) were used as the
training set. the other 60 reactions as test set. The result is quite
impressive: The reaction types of 95 YO(57 instances) of the
reactions in the test set were correctly predicted. The reason for
two undecided cases and one wrong case is very simple: these
three special reactions are not represented in the training set. In
. neural
[ l ] a) T Kohonen, Biol C'dxwi. 1982, 43, 59- 69; b) S~~//-Orxrrri/_irrion
arid A z o ciutrw MmcrrJ. 3rd ed.. Springer, Berlin. 1989.
[2] J. Zupan. J. Gasteiger. Nrirrul "VPIU orks for Ch~wmsI,\ Ari ltrlroducl;on. VCH.
Weinheim. 1993, J. Gastetger. J. Zupdn, A r i g w . Chern 1993. 105. 510-536;
Arigraii. Chtm?. h l . Ed. En$ 1993, 32. 503 -527
[3] ISIS Host IS avaikahle from MDL Information System\. Inc.. San Leandro.
[4] a ) A. Parlow. C. Wetske, J. Gasteiger. J Clieiri. / t i / . ('uiriptf. Srr. 1990, 30.
400 -402: b) Chemlnform-RX data base: Produced hy F:ichinformiitionszentrum Chemie. Berlin. Germany. available from MDL Information Systems
Inc.. San Leandro. CA. USA.
[5] J. Gasteiger. M Marsili, 7?~rru/i~drufi
1980. 36. 3219- 3 2 3
[6] J. Gasteiger. H. Saller. A i i g w . C'/icrii. 1985. 97. 699- 701 . Aiijiciv. C/i(,i?i.In/
Ed. €rig/. 1985. -74, 687 ~689.
[7] M G . Hutchings. J. Gasteiger. Pictro/ici/roii L C I C1983.
74. 2541k2544.
[X] J Gastetger. M. G. Hutchings. J CIimi. Socc P(,rkiri T,.or7\. 7 1984. 559- 564.
[Y] 0. A. Kolyamshin, V. V. Kormachev. Y N. Mttrasov. Z/7. Oh.rlid. K/irni. 1990.
60. 2170-2171.
[lo] D. Seebach. U. MiDlitz. P. Uhlmann, Clinii. fh.
1991. i l l . 1x45- 1852.
1111 D. M. Floyd. R. V. Moquin, K S. Atwal. S. Z. Ahmed. S H . Spergel. J. Z.
Gougoutas. M F. Malley. J. Org. C h ~ m ?1990.
55. 5572 5579.
[12] G. Griffiths, H. Mettler. L. S Mills. E Previdoli, He/). ( ' h i . Acrrr 1991, 74.
3OY- 314.
Schellman Motif in Dehydrooligopeptides:
Crystal and Molecular Structure of Boc-ValAPhe-Leu-Phe-Ala-APhe-Leu-OMe**
Kanagalaghatta Ramabhatta Rajashankar,
Suryanarayanarao Ramakumar, Tapas Kumar Mal,
Ratan Mani Jain, and Virander Singh Chauhan*
In the last few years it has become increasingly apparent that
peptides containing z,fl-didehydrophenylalanine residues
(APhe) exhibit preferential secondary structural features both in
the solid state and in
Introduction of APhe
residues in peptide sequences results in / h u n s [ i 5 1and helical
Recently we observed a novel [j-bend ribbon
structure in a pentapeptide containing two APhe residues!"' In
longer peptides containing several APhe residues 3, ,-helical
structures d ~ m i n a t e . [ ~l -o ]~ In
. a recent example an a-helical
Prof. V. S. Chauhan, T. K Mal. Dr. R. M. Jain
International Center for Genetic Engineering and Biotechnolog!,
NII Campus. Aruna Asaf Ali Marg
New Delhi-1 10067 (India)
Fax: Int. code +(11)6862316
K. R. Rajashankar. Prof. S. Ramakumar
Department of Physics, Indian Institute of Science
Bangalore-56001 2 (India)
We thank the Deparrment of Science and Technology (DST). India, lor financial support. the Department of Biotechnology (DBT). India. for access lo
facilities. and Prof. K. K Tiwari for encourasement. K . R. R. thanks Council
of Scientific and Industrial Research (CSIR). India. for a fellowship.
Без категории
Размер файла
516 Кб
classifier, reaction, michael, alkylation, organiz, neural, network, additional, related, friedelцcrafts, alkenes
Пожаловаться на содержимое документа