Biomedical Engineering Reference
In-Depth Information
The cellular system parameters were fixed
as follows:
Pmtr are the promoter sequences of the
gene
Act is the gene activation percentage
Protein lifetime: 3 cycles
Cons indicates the constitutive genes
20x20 development environment
Seq is the protein that encodes the gene
Division-inducing proteins: 1000 (north),
1100 (south), 1010 (west), and 1110 (east).
Special indicates if the gene has any associ-
ated behaviour
Apoptosis-inducing protein: 0000 string
Length Vertical Bar
5% and 10% communication limit values
and 10% apoptosis limit value
The first test sought to minimize the number of
genes needed to build a 5 cells vertical bar located
at the center of the environment. To achieve this,
we established a template that considered the
number of sections and the distance of the figure
centroid to the center as penalisations. The evalu-
ation function is shown in equation (2).
We found an individual with three genes (see
Table 1) that gave rise to a 5 sections structure.
The first gene works as the clock signal of the
system, because it is constitutive and does not
have any inhibiting promoters. The second gene
induces the structure to grow upwards; it will
only be activated during the first development
stages of the structure because it has a very high
activation percentage. Finally, the third gene is
easily activated and acts as a regulator, because
it will not allow the activation of the gene that
induces the growth.
Figure 8 shows the final stage of the tissue.
After the changes of the cycles 2, 4, 6 and 8,
the tissue has become stable. The error of this
solution is 0.0205 and mainly due to the fact that
the structure is not at the center and only grows
upwards. Between the options of minimizing the
number of genes and obtaining a more center-close
solution, the system opted for a DNA with only
3 genes, of which only one has a special action
grown N.
We also carried out experiments to select the
sequences that were identified as division as well
as apoptosis sequences. We tested randomly se-
lected sequences and sequences with the biggest
possible Hamming distance between them. The
result of the tests showed that the specification of
the sequences did not have any influence on the
development of the tests.
Growing Tests
We carried out several tests to reach a better
understanding of how the proposed model func-
tions and to extract some conclusions. The tests
involve the search of DNA strands whose associ-
ated performance might be the growth of simple
structures with a minimum number of genes.
Cellular bars growth with different lengths and
orientations is the structure that most simplifies
the complexity, the visual study of the structure,
and the analytical study of the genes. The choice
of this simple structure accelerates the search for
results, since it simplifies the evaluation of the
structures, making it quicker and more efficient,
as we could see in (Goldberg 1989).
The following tables show the genes of every
test, every row representing a gene. The ab-
breviations of the headings have the following
meaning:
N refers to the order number within the
gene's DNA
Search WWH ::




Custom Search