Artificial Cell Systems Based in Gene Expression Protein Effects - Advancing Artificial Intelligence Through Biological Process Applications - page 154

Figure 4. Crossover example

Figure 5. Types of mutations

promoter or a PCS) is added, similarly to the first
case of Figure 5; for another 20%, the existing
section is removed, similarly to the second case
of Figure 5; and finally, for the remaining 60% of
the opportunities, the value of one of the bits of
the section is randomly changed. The latter may
provoke not only the change of one of the values,
but also the change of the section type: if the bit
that identifies the section type is changed, the
information of that section varies. For instance,
if a promoter section turns into a PCS section, the
promoter sequence turns into the gene sequence,
as can be seen in the third case of Figure 5, and

constitutive and activation percentage values are
generated as in the case of Figure 6.

Changes of the Genome Evaluation

After reaching this development level and present-
ing the test set in (Fernandez-Blanco, Dorado,
Rabuñal, Gestal & Pedreira, 2007), the bottle-
neck of the model turned out to be the develop-
ment of the evaluation functions, since in every
new figure the development of the function was
time-consuming and not reusable. In order to
solve this problem, the evaluation function was