Fundamentals of Medicinal Chemistry

to assess the components of the library. A key problem with very large com-

binatorial libraries of mixtures is the large amount of work required to screen

these libraries.

Deconvolution is a method, based on the process of elimination, of reducing

the number of screening tests required to locate the most active member of a

library consisting of a mixture of all the components. It is based on producing

and biologically assaying similar secondary libraries that contain one less build-

ing block than the original library. It is emphasized that the biological assay is

carried out on a mixture of all the members of the secondary library. If the

secondary library is still as active as the original library the missing building

block is not part of the active structure. Repetition of this process will eventually

result in a library that is inactive, which indicates that the missing building block

in this library is part of the active structure. This procedure is carried out for

each of the building blocks at each step in the synthesis. Suppose, for example,

one has a tripeptide library consisting of a mixture of 1000 compounds. This

library was produced from 10 different amino acids (A

1

–A

10

) using two syn-

thetic steps, each of which involved 10 building blocks (Figure 6.15). The

formation of a secondary library by omitting amino acid A

1

from the initial

set of amino acids but reacting these nine with all 10 amino acids in the first and

second steps would produce 900 compounds. These compounds will not contain

amino acid residue A

1

in the first position of the tripeptide. If the resulting

library is biologically inactive the active compound must contain the residue at

position one in the tripeptide. However, if the mixture is active the process must

be repeated using A

1

but omitting a different amino acid residue from the

synthesis. In the worst scenario it would mean that the 900 member library

would have to be prepared ten times in order to determine first residue of

the most active tripeptide. Repeating this process of omission, combinatorial

Preparation of the original library

The preparation of the first group
of secondary libraries to find the
first residue in the peptide.

The preparation of the second
group of secondary libraries to find
the second residue in the peptide.

The preparation of the third group
of secondary libraries to find the
third residue in the peptide.

10 100

100