Encoding searches problems

Another set of early studies came from the work of Judson and coworkers [35, 36], which emphasized using GAs for search problems on small molecules and peptides, especially cyclic peptides. A dihedral angle representation was used for the peptides with values encoded as binary strings, and the energy function used the standard CHARMM force field. Mutations were implemented as bit flips and crossovers were introduced by a cut-and-paste of the strings. The small size of the system enabled a detailed investigation of the various parameters and policies chosen. In Ref. [37], a comparison between a GA and a direct search minimization was performed and showed the advantages and weaknesses of each method. As many concepts are shared between search problems on small peptides and complete proteins, these studies have contributed to subsequent attempts on full proteins. 

Let s begin with the simple case of binary encoding of the solution of a typical search problem for the maximum of a function. The quality of the solution is defined by the... 

It is only since the early 1990s that EAs have been applied to problems in computational chemistry. Yet, in that short space of time, they have had remarkable success in almost every complex search problem within the field. This success must be due to the problem independence of EAs. Given a problem, all that is required to implement an EA is appropriate encoding and fitness functions. It is hoped that the many applications discussed here are sufficient to demonstrate the potential of this powerful new tool for the investigation of a wide range of optimization problems in chemistry and biology. 

The CAS Substructure Search System provides a high degree of flexibility, with several levels of AND, OR, and NOT Boolean logic. Encoding searches however suffered from the following problems ... 

Signal peptide identification, like DNA intron/exon sequence discrimination, involves the two related problems of signal peptide discrimination (search for content) and cleavage site recognition (search for signal). It is well suited to neural network methods for several reasons. The functional units are encoded by local, linear sequences of amino acids rather than global 3-dimensional structures (Claros et al., 1997). The ambiguity of... 

The flexibility permitted in fragment definition is necessary in order to make full use of the encoded 2 D chemistry in the CSD. However, this flexibility, together with certain specific conventions used in the database, can lead to apparent misconceptions of the search process. Some of these problem areas are discussed in detail in Section 3.7.4 below. 

A special type of bond-type assignment problem concerns tautomerism. Some chemical databases provide specific tautomeric bond types, or analyse a query for either the keto or enol constructions and add the missing one (as an. OR. option) automatically. In the CSD, only the single tautomeric form identified by the crystal structure determination is encoded in the connectivity tables. The software has not yet been upgraded to accommodate searches for both forms, hence the onus is on the user to encode both representations if that is the desired goal of a query. 

In terms of coding method, it is necessary to depend on the situation of practice to set up the appropriate encoding and decoding scheme in order to overcome the shortcomings of the binary code. When variables are real vectors, we can directly use code by real number to inject heuristic information relevant to the field of problem in order to increase the search ability of GA. 

---