SIMULATION OF BACTERIAL GENOME EVOLUTION UNDER

REPLICATIONAL MUTATIONAL PRESSURES

Paweł Bła

zej, Paweł Mackiewicz and Stanisław Cebrat

Department of Genomics, Faculty of Biotechnology, University of Wrocław, ul. Przybyszewskiego 63/77, Wrocław, Poland

Keywords:

DNA asymmetry, Mutational pressure, Genome evolution, Monte Carlo simulation.

Abstract:

Directional mutational pressure associated with DNA replication is one of the most signiﬁcant forces shaping

nucleotide composition and structure of bacterial chromosomes as well as inﬂuencing the evolution of their

genes. Here we introduced the model of bacterial genome evolution including two mutational pressures acting

in differently replicated DNA strands (called leading and lagging). The simulations were performed on the

population of protein coding genes from the Borrelia burgdorferi genome which shows a very strong compo-

sitional bias between the DNA strands. The simulated genomes were eliminated by selection because of: (i)

stop translation codon occurrence in their gene sequences and (ii) the loss of their coding signal which was

calculated according to the algorithm for recognition of protein coding sequences. This algorithm considers

three independent homogeneous Markov chains to describe transition between nucleotides separately for each

of three codon positions in a given DNA sequence. The negative selection for stop codons appeared much

stronger than the one based on the coding signal and led to elimination of more genomes from the population.

The genes were subjected both to the direct mutational pressure, characteristic of the strand on which they are

located and to the reverse pressure, characteristic of the opposite strand. Generally, the elimination of genomes

because of stop codons occurrence was the most frequent for the reverse pressure whereas the coding signal

selection eliminated the genome most often for the direct pressure. The leading strand mutational pressure

was more destructive for coding signal whereas the the lagging strand pressure generated more stop codons in

the gene sequences.

1 INTRODUCTION

A different mode of DNA strands’ replication makes

that one DNA strand named leading, is synthesized

continuously whereas the complementary one, i.e.

lagging strand is synthesized from Okazaki frag-

ments. In consequence, these strands are subjected to

different nucleotide substitution patterns (Frank and

Lobry, 1999); (Kowalczuk et al., 2001a); (Kowal-

czuk et al., 2001b); (Rocha and Danchin, 2001);

(Rocha et al., 2006). These various directional mu-

tational pressures lead to disparate nucleotide com-

position of the differently replicated strands, which

is very well pronounced in many bacterial genomes

and is called the DNA asymmetry (Lobry, 1996);

(Freeman et al., 1998); (Grigoriev, 1998); (McLean

et al., 1998); (Mrazek and Karlin, 1998); (Mack-

iewicz et al., 1999a); (Tillier and Collins, 2000a);

(Lobry and Sueoka, 2002). The asymmetry is de-

ﬁned as a deviation from the equality between com-

plementary nucleotides in a single DNA strand, i.e.

[G]=[C] and [A]=[T]. The strongest asymmetry asso-

ciated with replication is observed in the nucleotide

composition in the third codon position of protein

coding sequences, which indicates mutational cause

of this bias (McLean et al., 1998); (Mackiewicz

et al., 1999a); (Tillier and Collins, 2000a); (Lobry

and Sueoka, 2002). The leading strand is usually rich

in guanine and thymine whereas the lagging strand

shows excess of cytosine and adenine. It is assumed

that the mutation C–>T, which is the most com-

mon substitution observed in the leading strand, is

the main factor responsible for this bias (Frank and

Lobry, 1999) although analysis of several bacterial

genomes revealed that similar compositional biases

may result from different mutational patterns (Rocha

et al., 2006). The effect of the mutational pressure

is to some extent accepted by selection because the

DNA asymmetry is also visible in the codon usage of

genes and amino acid composition of coded proteins

(McInerney, 1998); (Lafay et al., 1999); (Mackiewicz

et al., 1999b); (Rocha et al., 1999).

It was also found that the distribution of genes be-

tween the differently replicated strands is not random.

Bła

zej P., Mackiewicz P. and Cebrat S..

SIMULATION OF BACTERIAL GENOME EVOLUTION UNDER REPLICATIONAL MUTATIONAL PRESSURES.

DOI: 10.5220/0003755900510057

In Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms (BIOINFORMATICS-2012), pages 51-57

ISBN: 978-989-8425-90-4

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

Genes (e.g. coding for ribosomal proteins) that are

essential for cell functioning tend to be coded in the

leading strand (Rocha and Danchin, 2003a); (Rocha

and Danchin, 2003b). Their conserved location in

the DNA strand implicates a strong constraint on

chromosomal rearrangements (see also (Mackiewicz

et al., 2001)). The location of genes on differently

replicated DNA strands is also related to their di-

vergence. Genes coded in the leading strand gen-

erally accumulate less substitutions than the lagging

strand genes whereas orthologs located on the dif-

ferently replicated strands in the compared genomes

show the highest divergence (Tillier and Collins,

2000b); (Rocha and Danchin, 2001); (Szczepanik

et al., 2001); (Mackiewicz et al., 2003b). Com-

puter simulation studies conﬁrmed these observations

and revealed additionally that the best survival strat-

egy for the majority of genes is switching between

DNA strands to change the direction of the muta-

tional pressure from time to time (Mackiewicz et al.,

2004); (Dudkiewicz et al., 2004); (Dudkiewicz et al.,

2005); (Mackiewicz and Cebrat, 2009). Exceptions

from this tendency are genes coding for ribosomal

proteins which do not proﬁt very much from switch-

ing the directional pressure. It is in agreement with

their extremely conserved positions on the prokary-

otic chromosomes (Mackiewicz et al., 2001); (Mack-

iewicz et al., 2003a).

The above-mentioned computer simulation in-

cluded amino acid composition as the selection con-

straint. Thus we have presented here the results of

other type of simulations in which selection for stop

codon occurrence and the algorithm for recognition of

protein coding sequences (Bła

zej et al., 2010); (Bła

zej

et al., 2011) were applied. This algorithm exploites a

speciﬁc way of genetic code degeneration and rela-

tions between mutational pressure and selection pres-

sure shaping the amino acid usage in the proteomes.

We used the algorithm to study how selection operat-

ing on the nucleotide level inﬂuences the elimination

of genes subjected to the directional mutational pres-

sure. We were also interested in changes of the coding

potential of genes under the mutational pressure dur-

ing simulation time.

2 METHODS

The presented simulations were performed on DNA

sequences of the Borrelia burgdorferi genome. This

genome is very suitable for studies of the directional

mutational pressure associated with replication be-

cause it shows very strong DNA asymmetry (McIner-

ney, 1998); (Mackiewicz et al., 1999c); (Lafay et al.,

1999), does not show a selection for synonymous

codon usage, and has the deﬁned mutational pres-

sure associated with replication for both DNA strands

(Kowalczuk et al., 2001b).

We considered population composed of 72

genomes (individuals). Each genome was rep-

resented by 475 genes annotated as protein cod-

ing sequences according to the NCBI database

(www.ncbi.nlm.nih.gov) record of the B. burgdorferi

genome. These genes were divided into two subsets:

1. sequences lying on the leading DNA strand (333

genes of the total length 356, 034 nt);

2. sequences lying on the lagging DNA strand (142

genes of the total length 173, 796 nt).

One Monte Carlo step (MCS) of the simulation

consisted of two stages:

1. The mutation process of gene sequence;

2. The selection process of individuals.

During the ﬁrst stage, a nucleotide of the genome

sequence was chosen for mutation according to the

Poisson process assuming one possible mutation per

genome on average. The selected nucleotide was then

substituted by another nucleotide with the probabil-

ity given in the one of two substitution matrices for

the leading or the lagging strand (see Tab. 1 and Tab.

2). The matrices describe the real mutational pressure

for the differently replicated DNA strands of the B.

burgdorferi genome. They were constructed empiri-

cally by the comparison of original genes with their

potential pseudogenes found in intergenic regions of

the B. burgdorferi chromosome (Kowalczuk et al.,

2001b).

Table 1: The substitution matrix describing mutational pres-

sure in the leading DNA strand. A nucleotide in the ﬁrst

column is substituted by a nucleotide in the ﬁrst row.

A T G C

A 0.81 0.10 0.07 0.02

T 0.07 0.87 0.03 0.03

G 0.16 0.12 0.71 0.01

C 0.07 0.26 0.05 0.62

Table 2: The substitution matrix describing mutational pres-

sure in the lagging DNA strand. A nucleotide in the ﬁrst

column is substituted by a nucleotide in the ﬁrst row.

A T G C

A 0.87 0.07 0.03 0.03

T 0.1 0.81 0.02 0.07

G 0.26 0.07 0.62 0.05

C 0.12 0.16 0.01 0.71

In our simulations, we have considered two ver-

sions of the mutational pressure acting on gene se-

BIOINFORMATICS 2012 - International Conference on Bioinformatics Models, Methods and Algorithms

quences. In the ﬁrst possibility (direct pressure), the

genes from a given DNA strand (e.g. leading) were

subjected to the matrix of the strand on which they

were lying (i.e. leading). In the second case (reverse

pressure), the genes were under the pressure charac-

teristic of the opposite strand (i.e. lagging in this ex-

ample). It mimics the inversion of the gene in chro-

mosome.

The selection was modeled by the gene ﬁnding al-

gorithm (called PMC) for prokaryotic genomes that

was previously described by (Bła

zej et al., 2010);

(Bła

zej et al., 2011); (Wa

nczyk et al., 2011). This al-

gorithm consists of three independent homogeneous

Markov chains which describe transition probabili-

ties between nucleotides separately for each of three

codon positions in a given DNA sequence. This al-

gorithm does not require a high chain order to work

properly and uses speciﬁc correlations in the nu-

cleotide composition observed in the ﬁrst, the second,

and the third codon positions, which are characteristic

of protein coding genes (Cebrat et al., 1997); (Cebrat

et al., 1998). Small sizes of matrices used by this al-

gorithm enable using only a few coding sequences for

its effective training.

If a given sequence was recognized by this algo-

rithm as a protein coding sequence in the ﬁrst reading

frame, the nucleotide substitution was accepted, oth-

erwise the individual with the lost coding signal was

eliminated and replaced by another individual from

the population. Another reason for the individual re-

placement was the appearing of a stop codon inside its

gene sequence. We have recorded several parameters

during the simulations, which were ﬁnally presented

as averages calculated over all genes and individuals

in the population.

3 RESULTS

Individuals could be eliminated from the population

because of two reasons: loss of the coding potential

by one of its gene or appearance of a stop codon in

one of its gene sequence. The accumulated numbers

of individuals eliminated because of these two selec-

tions during the simulation time are shown in Fig.

1 and Fig. 2, respectively. These numbers increase

in both cases but the elimination of genomes by stop

codons generation is much more frequent than by the

loss of the coding signal. It indicates that it is easier

to introduce a stop codon to the gene sequence than

to change its coding potential described by nucleotide

composition. Moreover, the increase in the accumu-

lated number of individuals eliminated because of the

selection against stop codons begins already from the

Figure 1: The accumulated numbers of individuals elimi-

nated because of the coding signal loss in their genes.

Figure 2: The accumulated numbers of individuals elimi-

nated because of the stop codon appearance in their gene

sequence.

start of simulations whereas the number for the cod-

ing signal selection grows signiﬁcantly only after 1

million MCS. Till this time, the elimination of indi-

viduals from the population by the coding signal loss

is negligible.

In addition, the frequency of elimination of in-

dividuals depends clearly on the applied mutational

pressure (Fig 1. and Fig 2). For example the accumu-

lated number of eliminated genomes because of the

coding signal loss grows faster when the direct pres-

sure (i.e. when genes are subjected to their own ma-

trix) is applied. The number is lower in the case of the

reverse pressure (i.e. when genes are under the oppo-

SIMULATION OF BACTERIAL GENOME EVOLUTION UNDER REPLICATIONAL MUTATIONAL PRESSURES

site matrix). If we consider the accumulated number

of genomes eliminated because of the stop codon ap-

perance we obtain the opposite situation. In this case

the number of eliminated genomes grows faster when

the reverse pressure is applied.

The superposition of these two differently acting

selections equalizes the potential differences between

the mutational pressures in the case of the number

of substitutions accumulated during simulations (Fig.

3). Thus the numbers are almost identical for the di-

rect and reverse mutational pressures.

Figure 3: The average numbers of mutations accumulated

during simulations.

The average protein coding signal (computed for

all individuals who were still alive in a given simu-

lation step) normalized by the average protein coding

signal at the beginning of the simulation was another

important feature which was considered during every

simulation step (Fig. 4 and Fig 5).

As one could expect the average coding signal de-

creases faster in the simulation without selection con-

straints than in the simulation when the selection was

applied (Fig 4). However, the differences are not very

large, which indicates that the applied selection keeps

the signal very poorly.

Interestingly, the similar differences in the coding

signal were observed between simulations with se-

lections when direct and reverse mutational pressures

were applied (Fig 5). Nevertheless, in the case of the

sequences subjected to the reverse pressure, the cod-

ing signal is better preserved than for the direct pres-

sure. It well agrees with the results presented in (Fig

1), which indicates that the direct pressure more often

inﬂuences the coding signal and eliminates individu-

als because of the signal loss than the reverse pres-

sure. Furthermore, it is in agreement with the com-

Figure 4: The comparison of the average coding signal in

the simulation run with selection (dashed line) and without

selection (solid line). The change of the average coding

signal was normalized by the average coding signal at the

start of the simulation.

Figure 5: The comparison of the average coding signal in

simulations with selections run under different mutational

pressures. The change of the average coding signal was

normalized by the average coding signal at the start of the

simulation.

parative genome analysis that orthologous sequences

located on the differently replicated DNA strands

evolve faster than orthologs at conserved positions

(Tillier and Collins, 2000b); (Rocha and Danchin,

2001); (Szczepanik et al., 2001); (Mackiewicz et al.,

2003b).

We also compared the accumulated number of

damaged genes subjected to different pressures for the

leading and lagging DNA strand separately (Fig. 6

BIOINFORMATICS 2012 - International Conference on Bioinformatics Models, Methods and Algorithms

Figure 6: The accumulated numbers of genes damaged be-

cause of the coding signal loss in the leading strand and in

the lagging strand.

Figure 7: The accumulated numbers of genes damaged be-

cause of the stop codon appearance in their gene sequence

in the leading strand and in the lagging strand.

and Fig. 7). Because the leading strand genes are

more than two times numerous than the lagging ones,

we normalized the numbers of damaged genes. It is

interesting that the numbers of genes eliminated be-

cause of their coding signal loss are the highest when

genes are under the mutational pressure typical of the

leading strand (Fig 6). It is in the case when the

leading strand pressure is direct for the leading strand

genes and reverse for the lagging genes. On the other

side, the highest accumulated numbers of genes elim-

inated by stop codons are in the situation when genes

are subjected to the mutational pressure from the lag-

ging strand (Fig. 7). It is in the case when the lagging

pressure is reverse for the leading strand genes and di-

rect for the lagging genes. However, the excess is less

pronounced in this case and is visible after 2.5 mil-

lion MCS. Till this time, the accumulated numbers of

damaged lagging genes are higher when they are un-

der the leading strand pressure than the lagging strand

pressure.

4 DISCUSSION

We have considered a simulation model of bacterial

genome evolution where algorithm for ﬁnding protein

coding signal and stop codon occurrence played a role

as selection criteria. The simulations were run under

different mutational pressures which were described

by various substitution matrices. The obtained re-

sults indicate that it is not indifferent to genes from

the differently replicated DNA strands to which mu-

tational pressure and selection they are subjected. The

leading strand mutational pressure is more destructive

for coding signal and therefore is more harmful for

genes from both DNA strands when selection for the

coding signal preservation is applied. On the other

hand, the pressure typical of the lagging strand elim-

inates more genes being under the selection against

stop codons because the lagging strand substitutions

generate such codons with higher frequency. Inter-

estingly, the results are in agreement with analysis

of nucleotide usage biases in four-fold degenerated

sites in codons from bacterial genes (Khrustalev and

Barkovsky, 2010).

In general, the elimination of genomes from the

population because of stop codon appearance is more

frequent than their elimination by the loss of the cod-

ing signal which is additionally delayed in the time

of simulation. Interestingly, the average number of

accumulated mutations is very similar in the applied

mutational pressures because of the superposition of

effects of these two selections.

It seems to be important to notice, that the effect

of selection pressure used in all simulations was ex-

clusively negative. The higher robustness of coding

sequences with enhanced coding signal after substi-

tutions could be considered as some hidden no direct

positive selection. Nevertheless, it would be interest-

ing to introduce the direct positive selection effect of

the increased coding signal.

We expect that the obtained results of simulations

should be very similar to those using other selection

algorithms predicting protein coding sequences based

on other coding measures, e.g. codon or dicodon

usage, because the measures used by the algorithm

applied here are strongly correlated with the others.

SIMULATION OF BACTERIAL GENOME EVOLUTION UNDER REPLICATIONAL MUTATIONAL PRESSURES

The presented model of bacterial genome evolution,

which was shown in the example of B. burgdorferi,

should give similar general results for other bacte-

rial genomes because their DNA asymmetry resem-

bles that from the species analyzed here.

REFERENCES

Bła

zej, P., Mackiewicz, P., and Cebrat, S. (2010). Using the

genetic code wisdom for recognizing protein coding

sequences. In Proceedings of the 2010 International

Conference on Bioinformatics & Computational Biol-

ogy (BIOCOMP 2010), pages 302–305.

Bła

zej, P., Mackiewicz, P., and Cebrat, S. (2011). Algo-

rithm for ﬁnding coding signal using homogeneous

markov chains independently for three codon posi-

tions. In Proceedings of the 2011 International Con-

ference on Bioinformatics and Computational Biology

(ICBCB 2011), pages 20–24.

Cebrat, S., Dudek, M., and Mackiewicz, P. (1998). Se-

quence asymmetry as a parameter indicating coding

sequence in saccharomyces cerevisiae genome. The-

ory in Biosciences, 117:78–89.

Cebrat, S., Dudek, M., Mackiewicz, P., Kowalczuk, M.,

and Fita, M. (1997). Asymmetry of coding ver-

sus non-coding strand in coding sequences of differ-

ent genomes. Microbial and Comparative Genomics,

2:259–268.

Dudkiewicz, M., Mackiewicz, P., Kowalczuk, M., Mack-

iewicz, D., Nowicka, A., Polak, N., Smolarczyk, K.,

Kriaga, J., Dudek, M., and Cebrat, S. (2004). Simu-

lation of gene evolution under directional mutational

pressure. Physica A, (336):63–73.

Dudkiewicz, M., Mackiewicz, P., Mackiewicz, D., Kowal-

czuk, M., Nowicka, A., Polak, N., Smolarczyk, K.,

Kiraga, J., Dudek, M., and Cebrat, S. (2005). Higher

mutation rate helps to rescue genes from the elimina-

tion by selection. Biosystems, 80:192–199.

Frank, A. and Lobry, J. (1999). Asymmetric substitution

patterns: a review of possible underlying mutational

or selective mechanisms. Gene, 238:65–77.

Freeman, J., Plasterer, T., Smith, T., and Mohr, S. (1998).

Patterns of genome organization in bacteria. Science,

279:1827.

Grigoriev, A. (1998). Analysing genomes with cumulative

skew diagrams. Nucleic Acids Res., 26:2286–2290.

Khrustalev, V. and Barkovsky, E. (2010). The probability

of nonsense mutation caused by replication-associated

mutational pressure is much higher for bacterial genes

from lagging than from leading strands. Genomics,

96:173–180.

Kowalczuk, M., Mackiewicz, P., Mackiewicz, D., Now-

icka, A., Dudkiewicz, M., Dudek, M., and Cebrat, S.

(2001a). DNA asymmetry and the replicational muta-

tional pressure. J. Appl. Genet., 42:553–577.

Kowalczuk, M., Mackiewicz, P., Mackiewicz, D., Now-

icka, A., Dudkiewicz, M., Dudek, M., and Cebrat,

S. (2001b). High correlation between the turnover of

nucleotides under mutational pressure and the DNA

composition. BMC Evol. Biol., 1:13.

Lafay, B., Lloyd, A., McLean, M., Devine, K., Sharp,

P., and Wolfe, K. (1999). Proteome composition

and codon usage in spirochaetes: species-speciﬁc and

DNA strand-speciﬁc mutational biases. Acids Res.,

27:1642–1649.

Lobry, J. (1996). Asymmetric substitution patterns in the

two DNA strands of bacteria. Mol. Biol. Evol., 13:,

660–665.

Lobry, J. and Sueoka, N. (2002). Asymmetric directional

mutation pressures in bacteria. Genome Biol., 3:58.

Mackiewicz, D. and Cebrat, S. (2009). To understand nature

- computer modelling between genetics and evolution.

In J. Miekisz and M. Lachowicz (eds), From Genetics

to Mathematics (Series on Advances in Mathematics

for Applied Sciences) Vol. 79, pages 1–33. World Sci-

entiﬁc.

Mackiewicz, D., Mackiewicz, P., Kowalczuk, M., Dud-

kiewicz, M., Dudek, M., and Cebrat, S. (2003a).

Rearrangements between differently replicating dna

strands in asymmetric bacterial genomes. Acta Mi-

crobiologica Polonica, 52:245–261.

Mackiewicz, P., Dudkiewicz, M., Kowalczuk, M., Mack-

iewicz, D., Kiraga, J., Polak, N., Smolarczyk, K.,

Nowicka, A., Dudek, M., and Cebrat, S. (2004). Dif-

ferential gene survival under asymmetric directional

mutational pressure. Lecture Notes in Computer Sci-

ence, 3039:687–693.

Mackiewicz, P., Gierlik, A., Kowalczuk, M., Dudek, M.,

and Cebrat, S. (1999a). Asymmetry of nucleotide

composition of prokaryotic chromosomes. J. Appl.

Genet., 40:1–14.

Mackiewicz, P., Gierlik, A., Kowalczuk, M., Dudek, M.,

and Cebrat, S. (1999b). How does replication-

associated mutational pressure inﬂuence amino acid

composition of proteins? Genome Res., 9:409–416.

Mackiewicz, P., Gierlik, A., Kowalczuk, M., Szczepanik,

D., Dudek, M., and Cebrat, S. (1999c). Mechanisms

generating long-range correlation in nucleotide com-

position of the borrelia burgdorferi genome. Physica

A, 273:103–115.

Mackiewicz, P., Mackiewicz, D., Kowalczuk, M., Dud-

kiewicz, M., Dudek, M., and Cebrat, S. (2003b). High

divergence rate of sequences located on different DNA

strands in closely related bacterial genomes. J. Appl.

Genet., 44:561–584.

Mackiewicz, P., Szczepanik, D., Gierlik, A., Kowalczuk,

M., Nowicka, A., Dudkiewicz, M., Dudek, M., and

Cebrat, S. (2001). The differential killing of genes

by inversions in prokaryotic genomes. J. Mol. Evol.,

53:615–621.

McInerney, J. (1998). Replicational and transcriptional se-

lection on codon usage in borrelia burgdorferi. Proc.

Natl. Acad. Sci. U.S.A., 95:10698–10703.

McLean, M., Wolfe, K., and Devine, K. (1998). Base com-

position skews, replication orientation, and gene ori-

entation in 12 prokaryote genomes. J. Mol. Evol.,

47:691–696.

Mrazek, J. and Karlin, S. (1998). Strand compositional

BIOINFORMATICS 2012 - International Conference on Bioinformatics Models, Methods and Algorithms

asymmetry in bacterial and large viral genomes. Proc.

Natl. Acad. Sci. U.S.A., 95:3720–3725.

Rocha, E. and Danchin, A. (2001). Ongoing evolution of

strand composition in bacterial genomes. Mol. Biol.

Evol., 18:1789–1799.

Rocha, E. and Danchin, A. (2003a). Gene essentiality deter-

mines chromosome organisation in bacteria. Nucleic

Acids Res., 31:5202–5211.

Rocha, E. and Danchin, A. (2003b). Essentiality, not ex-

pressiveness, drives gene strand bias in bacteria. Na-

ture Genetics, 34:377–378.

Rocha, E., Danchin, A., and Viari, A. (1999). Universal

replication biases in bacteria. Mol. Microbiol., 32:11–

16.

Rocha, E., Touchon, M., and Feil, E. (2006). Similar com-

positional biases are caused by very different muta-

tional effects. Genome Res., 16:1537–1547.

Szczepanik, D., Mackiewicz, P., Kowalczuk, M., Gierlik,

A., Nowicka, A., Dudek, M., and Cebrat, S. (2001).

Evolution rates of genes on leading and lagging DNA

strands. J. Mol. Evol., 52:426–433.

Tillier, E. and Collins, R. (2000a). The contributions of

replication orientation, gene direction, and signal se-

quences to base composition asymmetries in bacterial

genomes. J. Mol. Evol., 50:249–257.

Tillier, E. and Collins, R. (2000b). Replication orientation

affects the rate and direction of bacterial gene evolu-

tion. J. Mol. Evol., 51:459–463.

nczyk, M., Bła

zej, P., and Mackiewicz, P. (2011). Com-

parison of two algorithms based on markov chains ap-

plied in recognition of protein coding sequences in

prokaryotes. In Proceedings of the Seventeeth Na-

tional Conference on Applications of Mathematics in

Biology and Medicine, pages 118–123.

SIMULATION OF BACTERIAL GENOME EVOLUTION UNDER REPLICATIONAL MUTATIONAL PRESSURES