Application of Genetic Algorithms to Combinatorial Synthesis:  A Computational Approach to Lead Identification and Lead Optimization^,

Abstract

A genetic algorithms (GA) based strategy is described for the identification or optimization of active leads. This approach does not require the synthesis and evaluation of huge libraries. Instead it involves iterative generations of smaller sample sets, which are assayed, and the “experimentally” determined biological response is used as an input for GA to rapidly find better leads. The GA described here has been applied to the identification of potent and selective stromelysin substrates from a combinatorial-based population of 206 or 64 000 000 possible hexapeptides. Using GA, we have synthesized less then 300 unique immobilized peptides in a total of five generations to achieve this end. The results show that each successive generation provided better and unique substrates. An additional strategy of utilizing the knowledge gained in each generation in a spin-off SAR activity is described here. Sequences from the first generations were evaluated for stromelysin and collagenase activity to identify stromelysin-selective substrates. GlyProSerThr-TyrThr with Tyr as the P1‘ residue is such an example. A number of peptides replacing Tyr with unusual monomers were synthesized and evaluated as stromelysin substrates. This led to the identification of Ser(OBn) as the best and most selective P1‘ residue for stromelysin.

Genetic optimization of combinatorial libraries

“…1 While this ''substructure chromosome'' focuses on chemical functionality, it will also carry some information about physical properties such as lipophilicity, because these are known to be related to substructure-based descriptions. The advantage compared to using lists of reagents as in the work of Weber et al 10 and Singh et al 9 is that parents and children do not need to originate from the same reaction nor do they have to share the same scaffold.…”

“…Weber et al 10 and Singh et al 9 have shown that a model consisting simply of a list of reagents which lead more often to active compounds than to inactives, is surprisingly effective. It is noteworthy that this approach does not use any structureactivity model.…”

A Genetic Algorithm for Solving Combinatorial Problems and the Effects of Experimental Error – Applied to Optimizing Catalytic Materials

“…As the ratio of A to B2 increases from 10% to 20% the cross-links linearly reduce back the optimum result. The optimum result has 90% of its components zero and yet it can be seen from Figure 6 that the fastest convergence is achieved when q is 3, which means that the probability of a component being zero is 3 4. By allowing more nonzero components the GA has more chance of discovering the catalyst A.…”

Evolutionary algorithms in computer-aided molecular design

“…With the increasing use of automated synthesis and assay procedures, this kind of application should become even more rapid and effective [197]. More recently, a GA method has been used by Singh and co-workers to guide the combinatorial synthesis of stromelysin substrates [198]. Here again, the fitness function used to evaluate the population members was an experimentally derived rather than computed value.…”