holowin-b

HOLOWin (HOLOsynthon and Windows) : the aim of this program is to search for the key steps of a synthesis.
Targets and reactions are coded at a skeletal level. The first approach was developed from an idea of Professor Goverdhan Metha (Bangalore, India) as an option in the MARSEIL program. The idea was to seek if a given reaction can be used as a key step : “Recognizing that the major concern in the synthesis of a complex molecule is the assembly of its carbocyclic (or heterocyclic) framework, we addressed ourselves to the limited but fundamentally important question of selecting the key reaction to generate the gross skeleton of the target molecule.”
For example, if the target is :


	Since there are ketones in this target, one could think that they could be introduced by the oxy-Cope rearrangement :


		In a “traditionnal” backward approach the reaction should be coded :

	In HOLOWin this reaction is coded in a simplified form as we are interested by the formation of the ketone group :

		And the three following targets are given to the program :


	and for each target the oxy-Cope reaction is applied, which leads, among other solutions, to the interesting ones :


	With this appraoch the chemist can see very quickly if a given reacion has some chances to be used to built the skeleton of the target. The problem is not solved but the main direction is given. Below are some examples for the following targets of Taxol© :




	Some possible precursors :


	Solution 1 was published in 1989 (Tetrahedron, 1989, 45, 1985 pub. n°31) and was tested independently next year by another searchors (Zücker, Lupia, Syn. Let. 1990, 729)


	If we are interested by the formation of the double bond from the oxy-Cope rearrangement the reaction is coded by the following scheme :


		and the target is :


	And the solutions are :


	This approach is very interesting, the only drawback is that in some cases we are obliged to submit several structures for one target. So we decided to generalize it by coding targets and reactions at a skeletal level. All functions are removed.


	The oxy-Cope reaction is now coded :


	and for Taxol the target is :


	By this way all the possibilities are found, 72 for the Taxol, but it is difficult to visualize where the functions are located in the target. So we improved this approach by coding the reaction in two steps : backward and forward (this one being fully functionalized). Then in the program we added an option which allows to perform the forward reaction on the precursors.


	Example. For the Taxol a solution given by the backward reaction is :


	Then on this precursor it is possible to apply the forward reaction and the program shows the target which corresponds to this precursor :


	Some solutions for the Taxol skeleton :


	Precursors


	functionalized targets


	For the crinipellin framework, from radical cascade reaction, this appraoch led to several solutions, and among them these ones which were new :

They suggest a new radicalar cyclic pathway which was tested successfully by Prof. Curran from our suggestion (Aust. J. Chem, 1995, 48, 261-267) :

This approach was first developed as an option in the Marseil/SOS program (running on a Macintosh), then we improved it in the HOLOWin software for the Windows interface. In this version we added an important option : when a precursor is displayed on the screen it is possible to analyze it directly in a new window, then come back to the parent target. (In the previous version the precursors were saved on disk, then when all precursors have been displayed, the user could select one of them as a new target and analyzes it).
This approach is very interesting to dissect rapidly a target in one or two (or more) key steps. Here is a view of the screen which shows this two steps procedure for a possible strategy from Diels-Alder then from Robinson reaction :

In the forward direction the strategy should be :

Strategies proposed for Taxol from [2+2+2], Diels Alder and Robinson reactions :

Some other strategies proposed by Holowin for the Taxol framework from three reactions (Robinson, Diels-Alder, Michael) (they are shown in the forward direction) :

Related papers :

31 - How to search original key steps in a synthesis of complex structures by using a microcomputer as a pocket-like calculator.
G. Mehta, P. Azario, R. Barone and M. Chanon.
Tetrahedron, 1989, 45, 1985-1994.

35 - Microcomputer Assisted RetroSynthesis.
P. Azario, M. Arbelot, A. Baldy, R. Meyer, R. Barone and M. Chanon
New J. Chem., 1990, 14, 951-956.

38 - New computer-based approach for seeking a key step in the synthesis of complex structures. Application to Taxane and Crinipellin diterpenoid frameworks.
G. Mehta, R. Barone, P. Azario, F. Barberis, M. Arbelot and M. Chanon
Tetrahedron, 1992, 48, 8953-8962.

39 - La conception de stratégies de synthèse par ordinateur : le logiciel STRAKS
R. Barone, F. Barberis, M. Arbelot, A. Baldy and M. Chanon
Science Technique et Technologie, 1993, 25, 40-43

41 - Search of the key step of a synthesis with STRAKS program. Examples with the taxane skeleton.
R. Barone, F. Barberis, M. Arbelot, A. Baldy and M. Chanon.
Polish Journal of Chemsitry (article sur invitation), 1994, 68, 2191-2198

45 - Computer-Aided Organic Synthesis : The Holosynthon Concept and the HOLOWin software.
M. Chanon, R. Barone,F. Barberis and A. Martinez
J. Org. Chem. (Russ.) 1995, 31(9), 1408-1413. (Article sur invitation pour les 60 ans du Professeur A. Zefirov).

46 - HOLOWin : A Fast Way to Search for Tandem Reactions with Computer. Application to the Taxane Framework.
F. Barberis, R. Barone and M. Chanon.
Tetrahedron, 1996, 52(46), 14625-14630.

47 - Le Logiciel HOLOWin, une approche simple, rapide, originale pour rechercher des stratégies de synthèse par ordinateur.
R. Barone, F. Barberis and M. Chanon
Sciences Chimiques, Lettres des départements scientifiques du CNRS, 1997, 69, 14-16


	[Home] [Some dates] [Papers] [sos] [marseil] [conan] [holowin] [graal] [miscellaneous]