Genes of the photosynthetic systems I and II are isolated for 45 cyanobacteria genomes. A frequency dictionary is built for each gene, to which a point in the 64-dimensional space of triplets is assigned. The photosystem gene structure in this space is considered. The genes are found to be clustered, depending on their belonging to the forward and reverse strands. Moreover, the points belonging to the forward and reverse strands form two perpendicular planes. The genes are grouped according to the type of bacteria within the main clusters. The values of the gene GC-content are distributed along the gradient.
Photosynthesis, the conversion of solar energy into biomass, is one of the most fundamental
processes on the Earth. Only photoautotrophic organisms such as cyanobacteria and plants can use
photons to break down water molecules into hydrogen and molecular oxygen. The photosynthetic
system of cyanobacteria, in contrast to purple and green bacteria, consists of two subsystems:
photosystem I and photosystem II. The functions of these systems are mutually complementary. The
primary function of photosystem II is to generate a strong oxidant, which initiates the oxidation of
water and transfers its electrons to a membrane carrier. The primary function of photosystem I is to
saturate these low-level electrons with energy in order to reduce NADP+. Since the energy of the total
process is too great within the framework of one reaction center, there appeared two photosystems in
the course of evolution, where different parts of this reaction occur. Their specific functions determine
the features of their structure. Thus, photosystem I is symmetric, i.e. there two branches of electron
transport, which makes it much faster. In contrast, photosystem II is asymmetric and has only one
working branch, which slows down the transport of electrons, but makes it more controllable.
Recently, significant progress has been made in determining the spatial structures of the photosystem
of various cyanobacteria. However, they continue to attract the interest of researchers. The natural
location of photosystem I (PSI), photosystem II (PSII), cytochrome (Cyt) b6f, and ATP synthase
within thylakoid membranes at the molecular level was visualized. An inhomogeneous distribution of
these four photosynthetic complexes was revealed and their dynamic features in a dense membrane
environment were determined [
In the present study, frequency dictionaries are used to represent genes as points in the frequency space of triplets. Let us describe in more detail the procedure for constructing dictionaries and identifying their structuredness. The genetic sequences of the length L , consisting of the alphabet symbols A,C,G,T , are considered. In our case, the genes of the photosystem act as genetic sequences. For each of the sequences, a frequency dictionary of thickness 3 is compiled. The frequency dictionary of thickness 3 implies a list of all the triplets 1 2 3 of consecutive nucleotides with an indication of the frequencies of these triplets. There can be 64 triplets in total. The frequency f is the ratio of the number of copies n of a given triplet to the total number of all the triplets N , where N is the sum of all n : f n
N
The triplets within a fragment are defined as follows: they do not intersect, but the combination of all the triplets completely covers the entire sequence in the gene. In other words, the reading frame in the construction of the dictionary is shifted by three nucleotides instead of one. In this case, the dictionary specifies the mapping of the genome into a 64-dimensional metric space. Two genes are considered to be close if the corresponding points in the 64-dimensional space are close in the sense of the Euclidean metric.
Thus, each gene is assigned a point in the 64-dimensional space of triplets. The following parameters are associated with each point: the name of the gene, the name of the species to which the gene belongs, the gene strand type (forward or reverse), the GC - content of the gene. The data view is built in the space of the first three principal components, calculated for the 64-dimensional space of triplets based on the obtained set of points in the VidaExpert program (http://bioinfoout.curie.fr/projects/vidaexpert/). The projections of the space on the plane of the first and second as well as second and third principal components are considered. (1) 3. Results and discussion
45 genomes of cyanobacteria from the EMBL bank have been examined, and the genes of the photosynthetic system have been annotated. The genes belonging to the forward and reverse strand have been found to form two clusters similar in the number of points belonging to them. That is, the number of the genes in the forward and reverse strand is approximately equal. Moreover, the points belonging to the same strand can be well approximated by an embedded plane. In addition, the planes corresponding to the points of each strand are perpendicular (Figure 1).
At the same time, the genes of the same type do not form dense clusters, but they are extended along with the clusters of the forward and reverse strands (Figure 2).
On the contrary, the genes belonging to cyanobacteria of the same species form clusters within the
strands. Figure 3 shows the genes related to Gloeobacter kilaueensis (ID in the EMBL-bank
CP003587) and Gloeobacter violaceus (BA000045), are denoted by crimson, Synechocystis
(AP012205, AP012276, AP012277, AP012278, BA000022, CP003265) are marked in purple, Nostoc
(CP003552, BA000019, CP003548, CP001037) are indicated in turquoise, Prochlorococcus marinus
(CP000551, CP00552, CP000553) are indicated in light green.
The distribution of the values of the gene GC -content has also been analyzed in the space of the first
three principal components. It has been found that the values of the gene GC -content are located in
the ascending order from the bottom (points indicated in green) upwards (points indicated in red)
(Figure 4). The average value points are marked in yellow. There is a gradient distribution in the
space of the frequencies of triplets for the values of the gene GC -content in the cyanobacteria
photosystem. This type of distribution is common in complete genomes. In particular, it is found in
the genomes of chloroplasts [
The spatial structure of the genes in the photosynthetic systems of cyanobacteria in the space of triplet frequencies is similar to the structures previously studied for the complete genomes of chloroplasts, mitochondria, and bacteria in terms of the spatial distribution of the values of the gene GC -content. However, the difference from the spatial structure of the complete genomes is that a pronounced clustering of the genes of the forward and reverse strands is observed for the genes of the cyanobacteria photosystem. In addition, the points have been found to be grouped in the frequency space of triplets according to the types of organisms corresponding to them rather than according to the type of genes.