Properties of the Standard Genetic Code and Its Alternatives Measured

by Codon Usage from Corresponding Genomes

Małgorzata Wnetrzak, Paweł Bła

zej and Paweł Mackiewicz

Department of Bioinformatics and Genomics, Faculty of Biotechnology, University of Wrocław,

Fryderyka Joliot-Curie 14a, 50-383 Wrocław, Poland

Keywords:

Alternative Genetic Code, Codon Usage, Error Minimization, Genetic Code, Mutation, Optimization.

Abstract:

The standard genetic code (SGC) and its modiﬁcations, i.e. alternative genetic codes (AGCs), are coding

systems responsible for decoding genetic information from DNA into proteins. The SGC is thought to be

universal for almost all organisms, whereas alternative genetic codes operate mainly in organelles and some

speciﬁc microorganisms containing usually reduced genomes. Previous analyzes showed that the AGCs mini-

mize the consequences of amino acid replacements due to point mutations better than the SGC. However, these

studies did not take into account the potential differences in codon usage between the genomes on which given

codes operate. The previous analyzes assumed a uniform distribution of codons, even though we can observe

signiﬁcant codon bias in genomes. Therefore, we developed a new measure involving codon usage as an addi-

tional parameter, which allowed us to assess the quality of a given genetic code. We tested our approach on the

SGC and its 13 alternatives. For each AGC we applied an appropriate codon usage characteristic of a genome

on which this code operates. This approach is more reliable for testing the impact of codon reassignments

observed in the AGCs on their robustness to point mutations. The results indicate that the AGCs are generally

more robust to point mutation than the SGC, especially when we consider the codon usages characteristic

of their corresponding genomes. Moreover, we did not ﬁnd a genetic code optimal for all considered codon

usages, which indicates that the alternative variants of the SGC evolved in speciﬁc conditions.

1 INTRODUCTION

There are many alternative genetic codes (AGCs)

which have emerged from the standard genetic

code (SGC). They are used mainly in mitochondrial

genomes (Abascal et al., 2012; Boore and Brown,

1994; Sengupta et al., 2007) but also in plastid (Cor-

tona et al., 2017; Janouskovec et al., 2013), some

nuclear genomes (Hoffman et al., 1995; Sanchez-

Silva et al., 2003; Santos et al., 1993) and bacterial

genomes (Bove, 1993; Campbell et al., 2013; Mc-

Cutcheon et al., 2009). Recent ﬁndings of AGCs in

various protists suggest that the number of these codes

can be strongly underestimated (Heaphy et al., 2016;

anek et al., 2017; Z

ahonov

a et al., 2016). Except for

the quite large genomes of ciliates, the AGCs operate

typically in small genomes encoding a limited number

of proteins. The small genome size facilitates the evo-

lution of genetic code variants because the potential

codon reassignments may not cause such a harmful

effect in synthesized proteins as in the case of large

nuclear genomes, in which even a single change in

the codon meaning may affect thousands of proteins

(Massey and Garey, 2007). Therefore, such reassign-

ments cannot be easily accepted.

The NCBI database includes currently 33 AGCs:

www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi.

The main differences between them and the SGC can

be classiﬁed into three categories: (i) reassignment of

a codon encoding one of the 20 typical amino acids

or stop translation signal, (ii) loss of codon meaning

induced by the disappearance of this codon from the

genome, and (iii) assignment of new amino acids

such as selenocysteine and pyrrolysine (Sengupta

et al., 2007). The most common are the reassign-

ments of stop codons to sense codons, e.g. codon

UGA to tryptophane (Bła

zej et al., 2019). Changes

in sense codons occur less frequently and mainly in

mitochondrial genomes.

Furthermore, it was shown that the same reassign-

ments have often evolved independently in different

phylogenetic lineages (Sengupta et al., 2007). The

main mechanisms involved in the evolution of the

AGCs are: (i) deletion of tRNA genes, (ii) duplication

of tRNA genes and their mutations, (iii) editing and

post-transcriptional base modiﬁcations of tRNAs, (iv)

Wnetrzak, M., Bła

zej, P. and Mackiewicz, P.

Properties of the Standard Genetic Code and Its Alternatives Measured by Codon Usage from Corresponding Genomes.

DOI: 10.5220/0008981000440051

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 3: BIOINFORMATICS, pages 44-51

ISBN: 978-989-758-398-8; ISSN: 2184-4305

mutations in genes encoding translational release fac-

tors, or (v) loss of codons subjected to a strong muta-

tional pressure (Sengupta and Higgs, 2005; Sengupta

et al., 2007). Horizontal gene transfer could also have

played a certain role in the AGCs evolution (Devoto

et al., 2019).

It is commonly believed that the AGCs emerged

through neutral evolution in small populations sub-

jected to genetic drift and strong mutational pres-

sure leading to tiny AT-rich genomes (Freeland et al.,

2000; Sengupta et al., 2007; Swire et al., 2005). How-

ever, other hypotheses concerning the evolution of

the alternative genetic codes have also been proposed.

They assume that: (i) codon changes associated with

the deletion of tRNA genes are driven by selection

to minimize the genome size and the time of repli-

cation (Andersson and Kurland, 1995), (ii) the reas-

signment of codon AUA from isoleucine to methion-

ine results in accumulation of methionine at the inner

membrane of animal mitochondria, which plays anti-

oxidant and cytoprotective role (Bender et al., 2008),

(iii) codon ambiguity can facilitate phenotypic diver-

sity and adaptability, which may help, e.g. yeasts, to

cope with stressful environments (Santos et al., 1999;

Gomes et al., 2007), (iv) mitochondrial genetic codes

may have evolved to reduce protein synthesis costs by

reassigning amino acids that are less expensive in syn-

thesis (Swire et al., 2005), (v) some changes in the ge-

netic code were accepted because they minimized ef-

fects of point mutations at the amino acid level (Kur-

naz et al., 2010; Morgens and Cavalcanti, 2013). All

these hypotheses suggest that the changes in the SGC

leading to the AGCs evolved as an adaptive trait.

The latter explanation seems interesting because

the same hypothesis, postulating that the code evolved

to minimize the effects of amino acid replacements

and errors during translation, was put forward for the

SGC (Epstein, 1966; Haig and Hurst, 1991; Freeland

et al., 2003; Goodarzi et al., 2005). Thus, common

principles could have governed the evolution of ge-

netic codes in general.

To verify this hypothesis, the SGC was compared

with: (i) possible theoretical genetic codes differing

from the universal code in one, two, or three codon

assignments, as well as (ii) its alternatives, regarding

the minimization of the harmful effects of amino acid

replacements in synthesized proteins (Bła

zej et al.,

2018; Bła

zej et al., 2019). The results indicated that

the codon reassignments observed in the AGCs gen-

erally improve their robustness to amino acid replace-

ments in comparison with the SGC and such natural

reassignments are often almost as good as the best

theoretical ones.

However, the cost function used to assess the

properties of the genetic codes took into account only

the sum of changes in the polarity of encoded amino

acids induced by single-point mutations in all the

codons. This function did not include the effects of

any mutational pressure or codon usage but assumed

a simple model, in which each codon occurs with

the same frequency

and the probability of any nu-

cleotide mutation is

and does not depend on the

codon position. Such approach was useful only in

ﬁnding the general tendencies of genetic codes re-

garding the error minimization hypothesis.

Because the codon bias is an important factor

inﬂuencing the ﬁnal mutational effect, it should be

taken into account. Therefore, in this work, we imple-

mented the codon frequencies observed in genomes

into the cost function formula. Then, we calculated

the cost values for the AGCs and for the SGC, in-

cluding the codon usage from the genomes of the or-

ganisms which use the alternative codes. The results

show that in almost all cases the AGCs tested on the

codon frequencies characteristic of the organisms that

use these codes outperform the SGC in terms of ro-

bustness to amino acid replacements.

2 METHODS

The examined alternative genetic codes were

downloaded from the NCBI taxonomy web page:

www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi.

From the whole set of described genetic codes we

chose 13 codes, which differ from the SGC by at

least one codon assignment and for which we were

able to easily obtain the codon frequencies of the

respective genomes that use these codes. The data

were extracted from the Codon Usage Database,

www.kazusa.or.jp/codon, (Nakamura et al., 2000)

and appropriate references (Perseke et al., 2011;

Swart et al., 2016) (Table 1). As we mentioned in

the Introduction section, we investigated the quality

of the SGC and its selected alternatives including the

respective codon usages in the applied cost function.

We tested all possible combinations of the SGC and

13 chosen genetic codes with 13 codons usages,

which gave us 14 × 13 = 182 cost values in total.

In order to test the properties of the genetic

codes, including speciﬁc codons usage observed in

the genomes on which a given code operate, we in-

troduced a new formula for the cost function F. This

parameter combines the differences between the prop-

erties of amino acids encoded by pairs of codons

< i, j > varying in one nucleotide substitution and

the probability of choosing codons < i, j > computed

from the given codon usage. According to these re-

Properties of the Standard Genetic Code and Its Alternatives Measured by Codon Usage from Corresponding Genomes

Table 1: The genetic code variants studied in this work and

the list of selected genomes from which we extracted re-

spective codon usages. All the codes are numbered (No.)

according to the notation in the NCBI taxonomy web page:

www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi.

No. Genetic code Genome

4 The Mold, Protozoan Mitoch. Aspergillus nidulans

5 Invertebrate Mitoch. Ascaris suum

6 Hexamita Nuclear Oxytricha falla

9 Flatworm Mitoch. Astropecten polyacanthus

10 Euploid Nuclear Euplotes octocarinatus

12 The Alternative Yeast Nuclear Candida albicans

13 Ascidian Mitoch. Halocynthia roretzi

16 Chlorophycean Mitoch. Spizellomyces punctatus

22 Scenedesmus Mitoch. Scenedesmus obliquus

23 Thraustochytrium Mitoch. Thraustochytrium aureum

24 Pterobranchia Mitoch. Rhabdopleura compacta

27 Karyorelict Nuclear Parduczia

28 Condylostoma Nuclear Condylostoma magnum

strictions, we deﬁned F for a given genetic code and

codon usage, in the following way:

F =

∑

<i, j>∈D

P(< i, j >)[ f (i) − f ( j)]

, (1)

where D is the set of pairs of codons that differ in

one nucleotide substitution, whereas P(< i, j >) is the

probability of choosing the pair < i, j >, calculated

according to the total probability formula:

P(< i, j >) = P(i) · P( j|i) (2)

where P(i) is the probability of selecting the codon

i and P( j|i) is the conditional probability of choos-

ing the codon j, when we know that i has been se-

lected (the frequency of the codon j divided by the

sum of frequencies of all the codons differing from i

by one nucleotide substitution). Moreover, f (i) and

f ( j) are the polarity values of the amino acids, com-

monly used in the study of the genetic code optimal-

ity, (Woese, 1973) encoded by the codons i and j, re-

spectively. Therefore, F represents the total weighted

sum of the squared differences between physicochem-

ical properties of amino acids encoded by the codon

pairs differing in one nucleotide substitution. What

is more, all the single substitutions that lead to non-

sense mutations, i.e. replacements of any amino acid

by stop translation signal, are included in the calcula-

tion as the maximum of squared difference computed

for all possible changes between the chosen amino

acid property. Thus small F values indicate that the

given code shows a tendency to minimize the conse-

quences of amino acid replacements, whereas large

values mean that the code is poorly optimized in this

respect.

Similarly to (Bła

zej et al., 2019), we calculated

the normalized percentage difference Pd between the

values of the function F for the SGC and the tested

code test for a ﬁxed codon usage. This difference is

deﬁned by

Pd(test,SGC) =

F(test) − F(SGC)

F(SGC)

· 100% . (3)

Clearly, Pd takes values in the range from −100 to

+∞. Particularly, negative values of Pd suggest that

the SGC is less robust to the consequences of point

mutations than the test code.

We would also like to point up the relationship

between F deﬁned here and several quality measures

proposed by other authors (Di Giulio, 1989; Haig and

Hurst, 1991; Freeland and Hurst, 1998; Santos and

Monteagudo, 2010; Bła

zej et al., 2016). The key dif-

ference lies in including the effect of different codon

frequencies on the potential costs of changes in amino

acid properties. In this case, for each codon i we as-

sume that this change is proportional to the probabil-

ity of choosing the second codon of the pair, j, from

the set of all codons differing from i in one nucleotide.

This requirement seems to be more realistic in com-

parison to the previous studies assuming that all pos-

sible pairs of codons D are equally probable.

Differences in the F values between the SGC and

its alternatives for various codon usages were as-

sessed statistically using the t-Student test because

the variables fulﬁlled the normal distribution require-

ment, as tested in the Shapiro-Wilk test. The re-

sulted p-values were corrected using the Benjamini-

Hochberg method to control the false discovery rate.

Differences between the compared groups were con-

sidered signiﬁcant when the p-value was smaller than

0.05. The analyzes were performed in R package

3.5.1 (A language and environment for statistical

computing, R Core Team 2018, R Foundation for Sta-

tistical Computing, Vienna, Austria).

3 RESULTS

At the ﬁrst stage of our study we calculated the val-

ues of the cost function F for the individual AGCs and

compared them with the result for the SGC. In this ap-

proach, we applied the codon usage corresponding to

the genome on which a selected alternative code op-

erates. In the Table 2, we presented the F function

values calculated for the chosen genetic codes and

the selected codon usages. It is clear that the qual-

ity of the SGC is strongly dependent on the assumed

codon usage because the F values change from 5.06

(codon usage 23) to 11.02 (codon usage 6). Because

the quality of the SGC depends on the codon usage,

we checked if the AGCs are better or worse than the

SGC regarding the codon usages gathered from their

BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms

corresponding genomes. Therefore, for every non-

standard genetic code, we calculated the values of the

cost function F and the measure Pd, which allowed us

to determine the difference in performance between

the SGC and the AGCs (Table 2). The calculated val-

ues of the Pd measure indicate that 11 out of 13 AGCs

perform from 11% to almost 39% better than the SGC

for the codon usages corresponding to the genomes on

which they operate. Only code 12 has its cost value

comparable with the SGC, and in just one case (code

23) the SGC outperforms the alternative genetic code

but only by less than 7%.

Table 2: The values of the F function calculated for the stan-

dard genetic code (F(SGC)) and its alternatives (F). For

each non-standard genetic code, we computed also the Pd

value as a measure of its distance from the SGC, which is

our reference point. The ﬁrst column (Code) refers to the

record number of the considered alternative code according

to the annotation in the NCBI database. The second column

(Codon usage) includes the codon usages numbered accord-

ing to the numbers of alternative genetic codes operating on

genomes from which a given codon usage was extracted.

Code Codon usage F F(SGC) Pd

4 4 6.016 7.123 -15.54

5 5 5.086 5.733 -11.28

6 6 6.759 11.018 -38.65

9 9 4.136 6.255 -33.87

10 10 7.554 8.550 -11.64

12 12 6.580 6.558 0.334

13 13 4.30 6.027 -28.58

16 16 6.007 7.937 -24.31

22 22 4.895 7.784 -37.10

23 23 5.406 5.059 6.85

24 24 4.145 5.492 -24.51

27 27 6.621 9.823 -32.59

28 28 6.509 9.370 -30.52

Another interesting question which arose during

this investigation concerns the genetic code optimal-

ity in terms of minimizing the function F for ev-

ery considered codon usage. In order to answer this

question, we calculated the cost values of the cho-

sen genetic codes for every codon usage considered

in this study. Figure 1 presents the box-plots of these

values. Generally, 10 AGCs show smaller median

F values than the SGC (numbered as 1 in the ﬁg-

ure). However, the difference is statistically signif-

icant only for the code 28, i.e. the Condylostoma

Nuclear Code (p=0.017). In turn, 3 alternative codes

have the median F value greater than the SGC but

the difference is statistically signiﬁcant only for the

code 23, i.e. the Thraustochytrium Mitochondrial

Code (p=0.004), which substantially stands out from

the others and shows the greatest variation in terms

of F values. Nevertheless, 11 codes have the min-

imum value of the cost function smaller than that of

1 4 5 6 9 101213162223242728

Genetic code

Figure 1: The box-plots of the cost function F values cal-

culated for every considered codon usage and the given ge-

netic code numbered on the x-axis. The standard genetic

code is indicated as 1, whereas other numbers refer to its

alternative versions. The thick horizontal line indicates the

median, the box shows the range between the ﬁrst and third

quartiles (IQR, the inter-quartile range) and the whiskers

determine the range without outliers for the assumption of

1.5 IQR.

the SGC. Only the codes 12 and 23 have the minimum

slightly greater than the SGC.

We also checked which codon usages generate the

smallest and the largest cost values for the SGC and

each of the AGCs. The results are presented in the

Table 3. It is interesting that only three out of 13

AGCs reached the minimum of the cost function F for

their respective codon usage. These genetic codes are:

Scenedesmus Mitochondrial Code (code 22), Thraus-

tochytrium Mitochondrial Code (code 23) and Pter-

obranchia Mitochondrial Code (code 24). What is

more, the codon usages characteristic of the organ-

isms using the codes 23 and 24 are also the best solu-

tions in terms of cost minimization for other genetic

codes, including the SGC. Generally, the minimal F

values are very similar to each other because they

range from 3.84 (Condylostoma Nuclear Code, code

28) to 5.41 (Thraustochytrium Mitochondrial Code,

code 23). The SGC does not seem to be well opti-

mized in terms of the minimal F value because there

are 11 AGCs that have this value smaller. In the

case of the AGCs, the minimal F values do not de-

viate substantially from the cost values calculated for

the codon usages typical of the corresponding codes.

The differences are from 0 to 3.4 (Euploid Nuclear

Properties of the Standard Genetic Code and Its Alternatives Measured by Codon Usage from Corresponding Genomes

Code, code 10). These results agree with our observa-

tions presented in the Table 2, where we showed that

the AGCs are mostly better adapted than the SGC to

their respective codon usages. It suggests that an op-

timization process must have taken place between the

AGCs and the speciﬁc codon usages of the genomes

on which these codes operate.

Table 3: The codon usages for which the given genetic

codes reach the minimal and maximal cost values (F min

and F max). CU min - the codon usage for which F is

minimized; CU max - the codon usage for which F is max-

imized. The codon usages were numbered according to the

alternative genetic codes operating on the genomes from

which the given codon usage was extracted.

Code CU min F min CU max F max

1 23 5.05940 6 11.01860

4 24 4.15692 6 10.76410

5 24 4.05795 6 10.57890

6 23 4.84787 10 8.17872

9 24 4.06393 6 10.62270

10 24 4.14932 6 10.76050

12 23 5.08569 6 11.09380

13 24 4.06806 6 10.57490

16 22 4.55405 6 9.99753

22 22 4.89575 6 11.26960

23 23 5.40601 4 18.08380

24 24 4.14572 6 10.72980

27 23 4.84787 10 8.17872

28 24 3.84902 10 7.16863

In the case of maximization of F, there is no code

that is the least optimized for its speciﬁc codon us-

age (Table 3). The codon usage that produced the

maximal cost values for 12 out of 14 genetic codes

is typical of Oxytricha falla (codon usage 6). In this

case, the cost values varied from nearly 10 (code 16)

to 11.27 (code 22). For alternative genetic codes 6,

27 and 28, the worst codon usage proved to be that

of Euplotes octocarinatus (codon usage 10), with the

cost values ranging from 7.17 (code 28) to 8.18 (codes

6 and 27). The codon usage from Aspergillus nidu-

lans (codon usage 4) occurred the worst for Thraus-

tochytrium Mitochondrial Code (code 23). Its cost

function reached the largest cost value equal to 18.08.

Among the tested codes, the Condylostoma

Nuclear Code (code 28) minimized the F function for

almost all considered codon usages (Table 4). The

one exception was the Chlorophycean Mitochondrial

Code (code 16) that minimized the cost function

best for the codon usage typical of the mitochondrial

genome of Scenedesmus obliquus. The smallest of

these cost values is 3.85 and was reached for the code

28 and the codon usage 24 (corresponding to Rhabdo-

pleura compacta mitochondrial genome). The largest

value was 7.17, which was reached for the code 28

Table 4: The codon usages (Codon usage) and the genetic

codes (Code) which showed the minimum cost value (F) for

these usages. The codon usages were numbered according

to the alternative genetic codes operating on the genomes

from which the given codon usage was extracted.

Codon usage Code F

4 28 5.45865

5 28 4.71020

6 28 6.40637

9 28 3.85415

10 28 7.16863

12 28 6.32379

13 28 4.19715

16 28 5.59880

22 16 4.55405

23 28 4.82071

24 28 3.84802

27 28 6.48476

28 28 6.50975

and the codon usage 10 (corresponding to Eu-

plotes octocarinatus genome).

4 DISCUSSION

The main goal of this work was to test the quality

of the alternative genetic codes by using a more re-

alistic measure, which includes the codon usage and

the probabilities of codon substitution characteristic

of the genomes on which these coding systems oper-

ate. This approach can be justiﬁed because we found

out that minimizing the costs of amino acid replace-

ments by the genetic codes strongly depends on the

codon usage. The AGCs were generally more ef-

fective than the SGC in minimization of the conse-

quences of point mutations but the differences were

not statistically signiﬁcant in most cases, when many

codon usages were tested. However, the AGCs mini-

mized the mutational costs substantially better (up to

39%) than the SGC for the codon usage correspond-

ing to the genomes on which these codes operate. The

cost function values calculated for the codon usages

speciﬁc for the AGCs were very similar or identical

with the minimal values found for any codon usage

and the given code. On the other hand, we could not

ﬁnd any genetic code optimal for every considered

codon usage.

These results indicate that the tested AGCs are

generally optimized to the codon usages correspond-

ing to the genomes on which they function. These

codes evolved in speciﬁc conditions of the given

genomes, which could have promoted their optimiza-

tion. In contrast, the SGC evolved very early dur-

ing life evolution in primordial organisms that easily

transferred genetic information between themselves,

BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms

which could have made this code universal for many

organisms characterized by various codon biases.

Therefore, it could not have been fully optimized dur-

ing its evolution, as indicated also by simulation stud-

ies based on evolutionary algorithms (Bła

zej et al.,

2018; Bła

zej et al., 2016; Massey, 2008; Novozhilov

et al., 2007; Santos and Monteagudo, 2011; San-

tos and Monteagudo, 2017; Wnetrzak et al., 2018),

whereas the minimization of mutation errors could

have occurred by the direct optimization of the mu-

tational pressure around the established genetic code

(Dudkiewicz et al., 2005; Mackiewicz et al., 2008;

Bła

zej et al., 2013; Bła

zej et al., 2017; Bła

zej et al.,

2015). It is also possible that the codon usage was

tuned in response to the changes in the genetic code.

The lack of the full SGC optimization could have also

resulted from its evolution driven by the expansion of

biosynthetic pathways of amino acids (Wong, 1975;

Di Giulio, 1999; Wong et al., 2016; Di Giulio, 2017).

The presented results may have an importance

for designing artiﬁcially modiﬁed organisms with al-

ternative codes, which produce peptides or proteins

with non-natural amino acids (Xie and Schultz, 2006;

Chin, 2014), by indicating that codon usage can sub-

stantially change the performance of the codes and

should be taken into account in the design. We are

planning to expand out model by including other

types of selection and codon measures to analyze the

optimality of the genetic codes.

ACKNOWLEDGEMENTS

We are very grateful to ﬁve anonymous reviewers for

their insightful comments and remarks, which signif-

icantly improved the manuscript. This work was sup-

ported by the National Science Centre, Poland (Nar-

odowe Centrum Nauki, Polska) under Grant number

2017/27/N/NZ2/00403.

REFERENCES

Abascal, F., Posada, D., and Zardoya, R. (2012). The evo-

lution of the mitochondrial genetic code in arthropods

revisited. Mitochondrial DNA, 23(2):84–91.

Andersson, S. and Kurland, C. (1995). Genomic evolution

drives the evolution of the translation system. Bio-

chemistry and Cell Biology, 73(11-12):775–787.

Bender, A., Hajieva, P., and Moosmann, B. (2008). Adap-

tive antioxidant methionine accumulation in respira-

tory chain complexes explains the use of a deviant ge-

netic code in mitochondria. Proc Natl Acad Sci U S A,

105(43):16496–16501.

Bła

zej, P., Mackiewicz, D., Grabinska, M., Wnetrzak, M.,

and Mackiewicz, P. (2017). Optimization of amino

acid replacement costs by mutational pressure in bac-

terial genomes. Scientiﬁc Reports, 7:1061.

Bła

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

(2013). Using evolutionary algorithms in ﬁnding of

optimized nucleotide substitution matrices. In Genetic

and Evolutionary Computation Conference, GECCO

’13, Amsterdam, The Netherlands, July 6-10, 2013,

Companion Material Proceedings, pages 41–42.

Bła

zej, P., Miasojedow, B., Grabinska, M., and Mack-

iewicz, P. (2015). Optimization of mutation pressure

in relation to properties of protein-coding sequences

in bacterial genomes. PLoS One, 10:e0130411.

Bła

zej, P., Wnetrzak, M., Mackiewicz, D., Gagat, P., and

Mackiewicz, P. (2019). Many alternative and theoret-

ical genetic codes are more robust to amino acid re-

placements than the standard genetic code. Journal of

Theoretical Biology, 464:21–32.

Bła

zej, P., Wnetrzak, M., and Mackiewicz, P. (2016). The

role of crossover operator in evolutionary-based ap-

proach to the problem of genetic code optimization.

Biosystems, 150:61–72.

Bła

zej, P., Wnetrzak, M., and Mackiewicz, P. (2018). The

importance of changes observed in the alternative ge-

netic codes. Proceedings of the 11th International

Joint Conference on Biomedical Engineering Systems

and Technologies - Volume 4: BIOINFORMATICS,

pages 154–159.

Boore, J. L. and Brown, W. M. (1994). Complete DNA

sequence of the mitochondrial genome of the black

chiton, Katharina tunicata. Genetics, 138(2):423–43.

Bove, J. M. (1993). Molecular features of mollicutes. Clin

Infect Dis, 17 Suppl 1:S10–S31.

Campbell, J. H., O’Donoghue, P., Campbell, A. G.,

Schwientek, P., Sczyrba, A., Woyke, T., S

oll, D., and

Podar, M. (2013). UGA is an additional glycine codon

in uncultured SR1 bacteria from the human micro-

biota. Proc Natl Acad Sci U S A, 110:5540–5545.

Chin, J. W. (2014). Expanding and reprogramming the ge-

netic code of cells and animals. Annu Rev Biochem,

83:379–408.

Cortona, A. D., Leliaert, F., Bogaert, K. A., Turmel, M.,

Boedeker, C., Janouskovec, J., Lopez-Bautista, J. M.,

Verbruggen, H., Vandepoele, K., and Clerck, O. D.

(2017). The plastid genome in cladophorales green

algae is encoded by hairpin chromosomes. Current

Biology, 27(24):3771–3782.

Devoto, A. E., Santini, J. M., Olm, M. R., Anantharaman,

K., Munk, P., Tung, J., Archie, E. A., Turnbaugh, P. J.,

Seed, K. D., Blekhman, R., Aarestrup, F. M., Thomas,

B. C., and Banﬁeld, J. F. (2019). Megaphages infect

Prevotella and variants are widespread in gut micro-

biomes. Nature Microbiology, 4:693–700.

Di Giulio, M. (1989). The extension reached by the mini-

mization of the polarity distances during the evolution

of the genetic code. J Mol Evol, 29(4):288–93.

Di Giulio, M. (1999). The coevolution theory of the origin

of the genetic code. J Mol Evol, 48(3):253–5.

Properties of the Standard Genetic Code and Its Alternatives Measured by Codon Usage from Corresponding Genomes

Di Giulio, M. (2017). Some pungent arguments against

the physico-chemical theories of the origin of the ge-

netic code and corroborating the coevolution theory. J

Theor Biol, 414:1–4.

Dudkiewicz, A., Mackiewicz, P., Nowicka, A., Kowalezuk,

M., Mackiewicz, D., Polak, N., Smolarczyk, K., Ba-

naszak, J., Dudek, M. R., and Cebrat, S. (2005).

Correspondence between mutation and selection pres-

sure and the genetic code degeneracy in the gene

evolution. Future Generation Computer Systems,

21(7):1033–1039.

Epstein, C. J. (1966). Role of the amino-acid ”code” and of

selection for conformation in the evolution of proteins.

Nature, 210(5031):25–8.

Freeland, S. J. and Hurst, L. D. (1998). The genetic code is

one in a million. J Mol Evol, 47(3):238–248.

Freeland, S. J., Knight, R. D., Landweber, L. F., and Hurst,

L. D. (2000). Early ﬁxation of an optimal genetic

code. Mol Biol Evol, 17(4):511–8.

Freeland, S. J., Wu, T., and Keulmann, N. (2003). The

case for an error minimizing standard genetic code.

Origins of Life and Evolution of the Biosphere, 33(4-

5):457–477.

Gomes, A. C., Miranda, I., Silva, R. M., Moura, G. R.,

Thomas, B., Akoulitchev, A., and Santos, M. A.

(2007). A genetic code alteration generates a pro-

teome of high diversity in the human pathogen Can-

dida albicans. Genome Biol, 8(10):R206.

Goodarzi, H., Najafabadi, H. S., Hassani, K., Nejad, H. A.,

and Torabi, N. (2005). On the optimality of the ge-

netic code, with the consideration of coevolution the-

ory by comparison of prominent cost measure matri-

ces. J Theor Biol, 235(3):318–25.

Haig, D. and Hurst, L. D. (1991). A quantitative measure

of error minimization in the genetic-code. J Mol Evol,

33(5):412–417.

Heaphy, S. M., Mariotti, M., Gladyshev, V. N., Atkins, J. F.,

and Baranov, P. V. (2016). Novel ciliate genetic code

variants including the reassignment of all three stop

codons to sense codons in Condylostoma magnum.

Mol Biol Evol, 33:2885–2889.

Hoffman, D. C., Anderson, R. C., DuBois, M. L.,

and Prescott, D. M. (1995). Macronuclear gene-

sized molecules of hypotrichs. Nucleic Acids Res,

23:1279–1283.

Janouskovec, J., Sobotka, R., Lai, D.-H., Flegontov, P.,

Kon

ık, P., Komenda, J., Ali, S., Pr

sil, O., Pain, A.,

Oborn

ık, M., Luke

s, J., and Keeling, P. J. (2013). Split

photosystem protein, linear-mapping topology, and

growth of structural complexity in the plastid genome

of Chromera velia. Molecular Biology and Evolution,

30(11):2447–2462.

Kurnaz, M. L., Bilgin, T., and Kurnaz, I. A. (2010). Certain

non-standard coding tables appear to be more robust

to error than the standard genetic code. J Mol Evol,

70(1):13–28.

Mackiewicz, P., Biecek, P., Mackiewicz, D., Kiraga, J.,

Baczkowski, K., Sobczynski, M., and Cebrat, S.

(2008). Optimisation of asymmetric mutational pres-

sure and selection pressure around the universal ge-

netic code. Computational Science - Iccs 2008, Pt 3,

5103:100–109.

Massey, S. E. (2008). A neutral origin for error minimiza-

tion in the genetic code. J Mol Evol, 67(5):510–516.

Massey, S. E. and Garey, J. R. (2007). A comparative ge-

nomics analysis of codon reassignments reveals a link

with mitochondrial proteome size and a mechanism

of genetic code change via suppressor tRNAs. J Mol

Evol, 64(4):399–410.

McCutcheon, J. P., McDonald, B. R., and Moran, N. A.

(2009). Origin of an alternative genetic code in the

extremely small and GC-rich genome of a bacterial

symbiont. PLoS Genetics, 5(7):e1000565.

Morgens, D. W. and Cavalcanti, A. R. (2013). An alterna-

tive look at code evolution: using non-canonical codes

to evaluate adaptive and historic models for the origin

of the genetic code. J Mol Evol, 76(1-2):71–80.

Nakamura, Y., Gojobori, T., and Ikemura, T. (2000). Codon

usage tabulated from the international DNA sequence

databases: status for the year 2000. Nucleic Acids Res,

28:292.

Novozhilov, A. S., Wolf, Y. I., and Koonin, E. V. (2007).

Evolution of the genetic code: partial optimization of

a random code for robustness to translation error in a

rugged ﬁtness landscape. Biol Direct, 2:24.

Perseke, M., Hetmank, J., Bernt, M., Stadler, P. F., Schlegel,

M., and Bernhard, D. (2011). he enigmatic mito-

chondrial genome of Rhabdopleura compacta (Pter-

obranchia) reveals insights into selection of an efﬁ-

cient tRNA system and supports monophyly of Am-

bulacraria. BMC Evolutionary Biology, 11:134.

anek, T.,

Zihala, D., Sokol, M., Derelle, R., Klime

s, V.,

Hradilov

a, M., Zadrob

ılkov

a, E., Susko, E., Roger,

A. J.,

Cepi

cka, I., and Eli

s, M. (2017). Nuclear ge-

netic codes with a different meaning of the UAG and

the UAA codon. BMC Biology, 15:8.

Sanchez-Silva, R., Villalobo, E., Morin, L., and Torres, A.

(2003). A new noncanonical nuclear genetic code:

Translation of UAA into glutamate. Current Biology,

13(5):442–447.

Santos, J. and Monteagudo, A. (2010). Study of the genetic

code adaptability by means of a genetic algorithm. J

Theor Biol, 264(3):854–865.

Santos, J. and Monteagudo, A. (2011). Simulated evolution

applied to study the genetic code optimality using a

model of codon reassignments. BMC Bioinformatics,

12.

Santos, J. and Monteagudo,

A. (2017). Inclusion of the ﬁt-

ness sharing technique in an evolutionary algorithm

to analyze the ﬁtness landscape of the genetic code

adaptability. BMC Bioinformatics, 18(1):195.

Santos, M. A., Cheesman, C., Costa, V., Moradas-Ferreira,

P., and Tuite, M. F. (1999). Selective advantages cre-

ated by codon ambiguity allowed for the evolution of

an alternative genetic code in Candida spp. Molecular

Microbiology, 31:937–947.

Santos, M. A., Keith, G., and Tuite, M. F. (1993). Non-

standard translational events in Candida albicans me-

diated by an unusual seryl-tRNA with a 5’-CAG-3’

(leucine) anticodon. The EMBO Journal, 12:607–616.

BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms

Sengupta, S. and Higgs, P. G. (2005). A uniﬁed model of

codon reassignment in alternative genetic codes. Ge-

netics, 170(2):831–40.

Sengupta, S., Yang, X., and Higgs, P. G. (2007). The mech-

anisms of codon reassignments in mitochondrial ge-

netic codes. J Mol Evol, 64(6):662–88.

Swart, E. C., Serra, V., Petroni, G., and Nowacki, M.

(2016). Genetic codes with no dedicated stop codon:

Context-dependent translation termination. Cell,

166:691–702.

Swire, J., Judson, O. P., and Burt, A. (2005). Mitochon-

drial genetic codes evolve to match amino acid re-

quirements of proteins. J Mol Evol, 60(1):128–39.

Wnetrzak, M., Bła

zej, P., Mackiewicz, D., and Mackiewicz,

P. (2018). The optimality of the standard genetic code

assessed by an eight-objective evolutionary algorithm.

BMC Evolutionary Biology, 18:192.

Woese, C. R. (1973). Evolution of the genetic code. Natur-

wissenschaften, 60(10):447–59.

Wong, J. T. (1975). A co-evolution theory of the genetic

code. Proc Natl Acad Sci U S A, 72(5):1909–12.

Wong, J. T., Ng, S. K., Mat, W. K., Hu, T., and Xue, H.

(2016). Coevolution theory of the genetic code at age

forty: Pathway to translation and synthetic life. Life

(Basel), 6(1):E12.

Xie, J. M. and Schultz, P. G. (2006). Innovation: A chemical

toolkit for proteins - an expanded genetic code. Nat

Rev Mol Cell Biol, 7(10):775–782.

ahonov

a, K., Kostygov, A. Y.,

Sev

ıkov

a, T., Yurchenko,

V., and Eli

s, M. (2016). An unprecedented non-

canonical nuclear genetic code with all three termi-

nation codons reassigned as sense codons. Curr Biol,

26:2364–2369.

Properties of the Standard Genetic Code and Its Alternatives Measured by Codon Usage from Corresponding Genomes