Single Nucleotide polymorphisms and their relationship to codon usage bias in the Pacific oyster Crassostrea gigas.

Abstract

DNA sequence polymorphism and codon usage bias were investigated in a set of 41 nuclear loci in the Pacific oyster Crassostrea gigas. Our results revealed a very high level of DNA polymorphism in oysters, in the order of magnitude of the highest levels reported in animals to date. A total of 290 single nucleotide polymorphisms (SNPs) were detected, 76 of which being localised in exons and 214 in non-coding regions. Average density of SNPs was estimated to be one SNP every 60 bp in coding regions and one every 40 bp in non-coding regions. Non-synonymous substitutions contributed substantially to the polymorphism observed in coding regions. The non-synonymous to silent diversity ratio was 0.16 on average, which is fairly higher to the ratio reported in other invertebrate species recognised to display large population sizes. Therefore, purifying selection does not appear to be as strong as it could have been expected for a species with a large effective population size. The level of non-synonymous diversity varied greatly from one gene to another, in accordance with varying selective constraints. We examined codon usage bias and its relationship with DNA polymorphism. The table of optimal codons was deduced from the analysis of an EST dataset, using EST counts as a rough assessment of gene expression. As recently observed in some other taxa, we found a strong and significant negative relationship between codon bias and non-synonymous diversity suggesting correlated selective constraints on synonymous and non-synonymous substitutions. Codon bias as measured by the frequency of optimal codons for expression might therefore provide a useful indicator of the level of constraint upon proteins in the oyster genome.

Full text

Page 1

Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive publisher-authenticated version is available on the publisher Web site

1
Gene
Volume 406, Issues 1-2, 30 December 2007, Pages 13-22
doi:10.1016/j.gene.2007.05.011
© Copyright © 2007 Elsevier B.V. All rights reserved

Archimer
Archive Institutionnelle de l’Ifremer
http://www.ifremer.fr/docelec/


Single Nucleotide polymorphisms and their relationship to codon usage
bias in the Pacific oyster Crassostrea gigas

C. Sauvage1, N. Bierne2, S. Lapègue1* and P. Boudry1

1 Laboratoire de Génétique et Pathologie - IFREMER - La Tremblade, France
2 Département de Biologie Intégrative, Institut des Sciences de l’Evolution de Montpellier UMR 5554
CNRS-UMII, Station Méditerranéenne de l’Environnement Littoral, Sète, France
* Correspondance :
Dr Sylvie Lapègue, LGP, Station IFREMER, 17390 La Tremblade, France
Tél : +33 (0)5 46 76 26 30 ; Fax : +33 (0)5 46 76 26 11
email : slapegue@ifremer.fr





Abstract:

DNA sequence polymorphism and codon usage bias were investigated in a set of 41 nuclear loci in
the Pacific oyster Crassostrea gigas. Our results revealed a very high level of DNA polymorphism in
oysters, in the order of magnitude of the highest levels reported in animals to date. A total of 290
single nucleotide polymorphisms (SNPs) were detected, 76 of which being localised in exons and 214
in non-coding regions. Average density of SNPs was estimated to be one SNP every 60 bp in coding
regions and one every 40 bp in non-coding regions. Non-synonymous substitutions contributed
substantially to the polymorphism observed in coding regions. The non-synonymous to silent diversity
ratio was 0.16 on average, which is fairly higher to the ratio reported in other invertebrate species
recognised to display large population sizes. Therefore, purifying selection does not appear to be as
strong as it could have been expected for a species with a large effective population size. The level of
non-synonymous diversity varied greatly from one gene to another, in accordance with varying
selective constraints. We examined codon usage bias and its relationship with DNA polymorphism.
The table of optimal codons was deduced from the analysis of an EST dataset, using EST counts as a
rough assessment of gene expression. As recently observed in some other taxa, we found a strong
and significant negative relationship between codon bias and non-synonymous diversity suggesting
correlated selective constraints on synonymous and non-synonymous substitutions. Codon bias as
measured by the frequency of optimal codons for expression might therefore provide a useful indicator
of the level of constraint upon proteins in the oyster genome.


Keywords: SNP; Genetic diversity; Codon bias; Crassostrea gigas

Page 2

Introduction
Single Nucleotide Polymorphisms (SNPs) are the most abundant sequence variations
encountered in a genome (Cho, Mindrinos et al., 1999; Picoult-Newberg, Ideker et al., 1999;
Griffin and Smith, 2000). Although moderately sparse in the human genome with one SNP
per kb (Sachidanandam, Weissman et al., 2001), they sometimes reach high densities in
some highly diverse species such as in some insects (e.g. one SNP every 125 bp in the
Aedes mosquito genome, Morlais and Severson, 2003) or some crops (e.g. one SNP every
104 bp in the maize genome, Tenaillon, Sawkins et al., 2001). With the development of DNA-
based marker assays and high-throughput genotyping technologies, SNPs have become
markers of choice for large scale mapping and genotyping (Rafalski, 2002; Black, Baer et al.,
2001). For example, it has had a great impact on the generation of genetic maps, for the
analysis of genetic diversity, trait mapping and diagnostics. They are especially useful for
association studies because of their high frequency in the genome, and they are genetically
more stable than microsatellite markers. SNPs are therefore ideally suited for the generation
of high-density genetic maps (Cho, Mindrinos et al., 1999; Nairz, Stocker et al., 2002) and
have number of advantages for population genetics studies (Vignal, Milan et al., 2002).
However, only few SNP markers have been developed in marine bivalves and more
generally in marine invertebrates.
Marine molluscs are recognised to reveal one of the highest level of allozyme
polymorphism within the animal kingdom (average heterozygosity 15-30%, Ward, Skibinski
et al., 1992; Bazin, Glemin et al. 2006). The extreme heterozygosity of marine molluscs is at
first best explained by large effective population sizes expected for species with high
fecundities, extensive larval dispersal, dense populations and broad distribution. However,
numerous studies have questioned the neutrality of allozyme variation in marine molluscs.
Popular examples of direct selection on some allozyme loci come from the marine bivalves
literature (Koehn, Newell et al., 1980; Karl and Avise, 1992; Riginos, Sukhdeo et al., 2002),
although they have sometimes been criticized (McDonald, 1996; Bierne, Daguin et al., 2003).
The observation that multilocus heterozygosity is frequently correlated with fitness-related
traits in these species (review in David, 1998) was initially taken as evidence for
overdominance at allozymes (Mitton, 1993). However, a consensus now emerged that the
correlation is in fact best explained by the indirect effect of deleterious mutations on neutral
marker variations (David, 1998), this neutral alternative introduced two requisites that have a
bearing on our understanding of the population genetics of marine bivalves: (i) a high genetic
load and (ii) the existence of particular population structures generating a strong variance in
individual inbreeding (Bierne, Tsitrone et al., 2000). Both have received support. The genetic
load has been quantified and was estimated to be extremely high (Bierne, Launey et al.,
1998; Launey and Hedgeock, 2001). Several authors have challenged the idea that marine
bivalves occur in large, homogeneous, randomly mating populations. Hedgecock’s
sweepstakes hypothesis proposes that large variation in reproductive success lead to
effective population sizes several orders of magnitude below census numbers (Hedgecock,
1994). The slight, unpredictable but significant genetic differentiation observed at small
spatial scales (Johnson and Black, 1984; David, Perdieu et al., 1997) is also consistent with
the fragmentation of marine populations into small, transient reproductive groups. Finally,
Bazin (2006) reported that mitochondrial diversity was not higher in marine molluscs than in
other taxonomic groups in sharp contrasts with allozyme diversity. Bazin (2006) also reported
a positive correlation between allozyme diversity and nuclear DNA diversity at a large
taxonomic scale, suggesting that these two types of variation share the same information,
namely population size. However, their meta-analysis nonetheless revealed that
quantification of nuclear DNA diversity remains uncommon in many invertebrate groups,
including molluscs, precluding the use of DNA polymorphism for comparison with
mitochondrial diversity. Taking into consideration the doubts accumulated on the neutral
status of allozymes and on the effective population size in marine bivalves, one might
enquire thorough assessments of silent DNA polymorphism in these taxa.
Here, we describe the characterization of SNPs in coding and non-coding sequences
of the Pacific oyster Crassostrea gigas. We used primer sequences designed in a set of 41
ESTs and direct sequencing of PCR products. The validity of detected SNPs was
2

Page 3

ascertained by multiple sampling of the same allele in a sample of related individuals. This
sampling strategy allowed us to remove technical artefacts of PCR and sequencing that
generate singleton substitutions. We describe levels of diversity observed in coding and non-
coding regions. We also undertake an investigation of codon usage bias and its relationship
with DNA polymorphism. It is now widely accepted that selection acts on synonymous
codons to improve translation in many organisms (Ikemura, 1992; Akashi, 1994) but not
always (Duret and Mouchiroud, 1999). Variation in the effectiveness of selection on
synonymous codons between species is likely to reflect variation in effective population sizes
rather than variation in coefficients of selection (Li, 1987; Bulmer, 1991; Akashi, 1995; Cutter,
Baird et al., 2006). Indeed, genomes with the strongest variation in codon usage correspond
to species recognised to have large effective population sizes such as bacteria, yeast or
insects (Ikemura, 1982; Merkl, 2003; Akashi, 1995) while natural selection does not appear
to play a role in Mammals (Urrutia and Hurst, 2001). Although selection on synonymous
codon use has been known for more than two decades, it remains unclear whether codon
usage primarily affects the elongation rate or the fidelity of protein synthesis (Akashi, 2001;
Duret, 2002). However, the latter hypothesis has received evidences in Drosophila
melanogaster, Caenorhabditis elegans and Escherichia coli (Akashi, 1994; Marais and Duret,
2001; Stoletzki and Eyre-Walker, 2007). A negative correlation between the rate of non-
synonymous substitution and codon bias have been described in a number of species
(Stoletzki and Eyre-Walker, 2007; Pal, Papp et al., 2001; Betancourt and Presgraves, 2002)
and seems best explained by the effect of selection on the accuracy of translation (Bierne
and Eyre-Walker, 2006; Stoletzki and Eyre-Walker, 2007). As a consequence, some authors
have recently proposed that codon bias might sometimes be used as an indicator of the
selective constraint acting on a protein (Stoletzki and Eyre-Walker, 2007; Plotkin, Dushoff et
al., 2006) as others proposed to use codon bias as a measure of the level of gene
expression (Sharp and Li, 1987; Karlin and Mrazek, 2000; Coghlan and Wolfe, 2000). We
here provide results suggesting that codon bias could provide useful information about the
level of constraint upon a gene in the oyster genome.
2. Material and Methods
2.1 Biological material and DNA extraction
Twenty four two year-old oysters (Crassostrea gigas) were used in this SNP
discovery experiment. They were produced during the French MOREST program that
studied the genetic basis of the summer mortality phenomenon as described in Degremont
(2007).These 24 individuals belongs to the third stage of selection for the resistance to the
summer mortality and were selected to initiate a genetic linkage mapping experiment to
detect quantitative trait loci (QTL) of aquacultural interest. These oyster are related with each
others as following (figure 1): there are two replicates of four groups composed of three
individuals (one male and two females), which are brothers and sisters. They are related by a
semi-brother relationship to individuals of the three other groups. All individuals that compose
the eight groups are the progeny issued from a cross of two individuals randomly taken from
eight pools (G2) of individuals, named A, O, G, J, K, Q, R and AC. These eight pools are
themselves the progeny of an ancestral couple of oysters (G1) selected for their resistance to
summer mortality phenomenon (Degremont, Ernande et al., 2007).

DNA was extracted from samples of mantle tissue following the Promega DNA Wizard
cleanup Kit recommendations. The quality of DNA was first checked on a 1.5% agarose gel
(15V/cm; 40 min) and secondly by quantification using an Eppendorf biospectrophotometer.
This allowed to get 24 equal concentration of DNA (100µg/mL) prepared for the amplification
step.
2.2 Primer Design and Amplification
Primers were designed using the Primer3 software (Rozen and Skaletsky, 2000) from
Expressed Sequence Tags (EST) developed in the Pacific oyster and retrieved from the
Genbank database (http://www.ncbi.nlm.nih.gov/). Few additional unpublished ESTs were

3

Page 4

also added in our experiment (Huvet, Boulo and Renault, Pers. Com.). Primers produced
fragments from 250 to 550 bp. PCR conditions were standardized by the use of a touchdown
PCR protocol. The same reaction mix was used for each pair of primers and amplification
was performed with three ranges of annealing temperatures (60 to 50°c, 65 to 55°c and 82 to
74°c). The mix was composed for each sample of 0.3U Taq polymerase (New England
Biolabs), 10mM of provided buffer, 1mM MgCl2, 2mM of dNTP (Eurogentec), 10µM of each
primer and 100ng of genomic DNA, in a final reaction volume of 48µL. Pairs of primers with
the same optimal range of annealing temperature were grouped and ran on a Perkin-Elmer
ABI 2700 PCR machine (Applied Biosystems) as following: initial denaturation step for 5 min
at 94°C, then 2 cycles of 1 min denaturation at 94°c, for every subsequent 2 cycles, the
annealing temperature was decreased by 1 degree Celsius during 1 min, and an extension
step at 72°c of 1 min. At the lowest annealing temperature (50, 55 or 74°c), 25 cycles at 94°c
for 30s, 30s annealing and 1 min extension at 72°c were applied. A quantity of 5µL of PCR
products was purified with 2µL of the ExoSAP-IT enzyme (Amersham Biosciences) to
remove non-incorporated dNTPs and primers according to the manufacturer manual. Then,
purified PCR products were sequenced in a single direction using the foward primer with the
ABI Prism BigDye v3 Terminator Cycle sequencing Kit (Applied Biosystems) and the
sequences were analysed on an ABI 3100 Avant genetic analyser (Applied Biosystems).
2.3 Nucleotide diversity
Sequences were edited, corrected by hand if needed and aligned using Seqscape
v2.1 software (Applied Biosystems) using the KB basecaller algorithm. After a Blast
homology search (http://www.ncbi.nlm.nih.gov/BLAST/), the sequences that did not
correspond, at least partly to the EST used to design primers, were removed. Multiple
sequences alignment was exported to BioEdit v7.0.5. (Hall, 1997) and aligned with the
corresponding EST sequence in order to infer intronic regions. Open Reading Frames (ORF)
were inferred by trying the three frames and finding the longest ORF together with the help of
orthologous sequence from GenBank. Regions outside the ORF but matching the EST
sequence were considered to be UTRs. However, UTRs were found for only 4 loci (Table 1)
and the majority of non-coding DNA sequences described in the present study belongs to
introns. We expected to find identical alleles by descent (i.e. the same sequence) several
times because individuals of the sample are related. As a consequence, singleton
substitutions were inferred to be technical artefacts arisen during PCR and sequencing
steps. Although, SNPs were characterised by direct sequencing of PCR products, this
sampling strategy allowed us to obtain a valid characterisation of these markers. SNPs were
identified as transitions or transvertions for both coding and non-coding regions. For SNPs
occurring in coding sequences, variations were classified as synonymous or non-
synonymous changes. The analysis of genetic diversity was conducted with DnaSP 4.0
(Rozas and Rozas, 1995). We computed the average number of SNPs per site (Π) which is
the number of polymorphic sites divided by the length of the sequence. Calculations were
conducted independently for non-coding (Πnc), synonymous (Πs), and non synonymous (Πns)
substitutions. In order to compute Πs and Πns, we estimated the number of synonymous and
non-synonymous sites in a coding sequence with the method of Nei and Gojobori (1986).

2.4 Synonymous codon use
In order to investigate codon usage bias, the table of optimal codons was deduced
from the analysis of an EST dataset using EST counts as a rough assessment of gene
expression (Duret and Mouchiroud, 1999). The dataset was composed of 8800 EST
sequences which were available for C. gigas in Genbank (http://www.ncbi.nlm.nih.gov/) and
the Marine Genomics Europe Network of Excellence databases. Open reading frames
(ORFs) were identified by choosing the longest possible translation into amino acid
sequence. Sequences with an ORF smaller than 100 codons (300bp) were removed from the
dataset to prevent the occurrence of wrong ORFs and to have enough codons to compute
the frequency of synonymous codons. It is likely that some EST annotations are wrong as a
result of artifactual frameshift substitutions in single-pass sequences. However, our
procedure should result in removing a part of the coding region from the analysis rather than

4

Page 5

misinferring UTRs as coding regions. We therefore argue that sequence quality problems
could introduce a statistical noise but should not bias the analysis. We built clusters of ESTs
corresponding to the same gene by using the algorithm developed by Bazin, Duret et al.
(2005) to construct the Polymorphix database. Two sequences were clustered into the same
family if they shared fragment longer than 300pb with similarity superior to 95%. Only the
longest ORF of a cluster was used to compute tables of codon usage. The frequency of
synonymous codon per amino acid was computed with the program CODONW (Peden,
1999). A synonymous codon was inferred as “optimal” when its frequency significantly
increased with EST counts (Duret and Mouchiroud, 1999). This was tested by non-
parametric Spearman’s correlation in JMP v5.0© (SAS Institute Inc.). Once optimal codons
were inferred, we used the frequency of optimal codons (Fop; Ikemura, 1985) as a measure
of codon bias.
3. Results
3.1 Types and distribution of polymorphism

A total of 84 EST sequences were chosen for primer design. Fifteen of these ESTs
were chosen on the basis of their putative function in relation to summer mortality. Forty one
ESTs (51%) amplified a clear genomic DNA fragment. For 31 of the 41 amplifying loci (72%),
the size of the fragment obtained was longer than the size expected from the EST sequence
because of the presence of introns. Therefore, we suspect that the presence of several
and/or very long introns could partly explain the rate of failure of PCR amplification.

total of 290 SNP markers identified in both coding and non-coding DNA. Around ¾ of SNPs
(69%) were localised in non-coding regions (Table 1). The level and the repartition of this
polymorphism in each locus is shown in the figure 2. We found only 4 monomorphic locus
which represented only 5% of amplified loci. The average density of SNPs in coding regions
was estimated to 1 every 60 bp and 1 every 40 bp in non-coding ones. An average number
of 6.7 SNPs were detected in each amplicon with a minimum of 0 (monomorphic sequence)
and a maximun of 30 SNPs within a 241 bp (1 SNP every 9bp) entirely non-coding sequence
of the Integrin gene.
Approximately, a total of 10.5 kb of the pacific oyster genome were sequenced with a
Our data show a global ts/tv ratio (1.3) in the Pacific oyster, which is similar to the ratio
observed in Drosophila (1.5, Moriyama and Powell, 1996) or humans (1.4, Brookes, 1999).
ts/tv was equal to 2.1 for synonymous changes while it was 0.9 and 1.2 for non-synonymous
and non-coding changes respectively (exact t Fisher test, t=0.264). With twice more
possibilities of transvertions than transitions, ts/tv should ideally be equal to ½, if substitutions
appeared randomly in DNA. However, it is well-known that a mutational bias favours
transition over transversion. Because of the structure of the genetic code, synonymous
changes are more often transitions than transversions. As much of the diversity is composed
of synonymous substitutions, the ts/tv ratio increases in coding sequences (ts/tv =1.34). The
same phenomenon is observed in the mosquito Aedes aegypti where ts/tv in coding
sequences is similar. (Morlais and Severson, 2003). The mutational bias for transition is
therefore more accurately estimated from non-coding sequences and was estimated to be
ts/tv =1.2 in C. gigas.

Diversities were similar at synonymous (Πs=0.035) and non-coding positions (Πnc=0.038)
and were not significantly different (per substitution t test = -0.55, P=0.29). Although the level
of polymorphism in coding and non coding regions was positively correlated as expected for
neutral substitutions sharing correlated genealogies, the correlation was not significant
(Spearman’s ρ= 0.3; P=0.16). As the two kinds of silent substitutions gave the same
information, they have been combined into a single indice (Πsi). The non-synonymous to
silent polymorphism ratio (Πn/Πsi) was 0.16 on average. However, Πn/Πsi varied greatly
from one gene to another.
The number of SNPs per site was calculated for both coding and non-coding regions.
5

Page 6

3.2 Codon Usage Bias
Table 2 presents Spearman’s correlation coefficients between the frequency of
synonymous codons per amino acid and EST counts. As an illustration, we explain the
results obtained at the Phenylalanine amino acid. Phenylalanine is a two-fold degenerate
codon, which means that two different codons (UUU and UUC) can code for this amino acid.
The average frequency of this amino acid among all others is 0.05 in the Pacific oyster
genome and it does not vary significantly with expression levels as measured by EST counts.
Ideally, each of the two codons of the phenylalanine should be equally represented by a
frequency of 0.5. However, as expected in a GC-poor genome, the UUU codon is over-
represented on average in lowly expressed genes (~60% in the category of genes with a
single EST). On the other hand, the frequency of the UUC codon increases with expression
level and the situation is reversed in highly expressed genes for which the UUU codon is
under-represented (~40% in the category of genes with 5 or more ESTs). The same trends
were observed for other amino-acid: although -A and –U ending codons are more frequent
on average as expected in a GC-poor genome, -G or -C ending optimal codons are more
frequently observed (Table 2). As a consequence, GC-content at the third coding position is
also correlated to expression levels (Spearman’s ρ = 0.087, p<0.0001) although less than
Fop (Spearman’s ρ = 0.13, p<0.0001). As a consequence, codon usage is paradoxically less
balanced in lowly expressed genes that are GC-poor than in highly expressed genes that
reach an intermediate GC-content. This result highlights that codon usage bias is difficult to
establish without an assessment of expression levels (Duret and Mouchiroud, 1999). In the
case of the genome of C. gigas, a simple analysis of codon usage without expression data
could have resulted in the inference that GC-poor genes are more biased and that -A and -U
ending codons are optimal for translation while it is indeed the reverse –translational
selection primarily favours -G or -C ending codons although rare in a GC-poor genome but
the effectiveness of selection is not sufficient to result in a strong enrichment in GC-content
which remains intermediate in biased genes. The results we obtained in C. gigas are similar
to those recently reported in some GC-poor Nematode genomes in which translational
selection nonetheless tends to favour -G or -C ending codons (Cutter, Baird et al., 2006).
We analysed whether genetic diversities would be correlated with codon bias.
Synonymous diversity could have been expected to decrease with codon bias because of
selection for codon usage. However, selection on synonymous substitutions is often too
small to result in a detectable correlation (Bulmer, 1991; McVean and Charlesworth, 2000).
In Drosophila for instance, although synonymous divergence decrease with codon bias
(Sharp and Li, 1987) synonymous polymorphism does not vary with codon bias with a
dataset size comparable to this study (Bierne and Eyre-Walker, 2006). On the other hand,
we detected a significant correlation between non-synonymous diversity and Fop (Figure 3;
Spearman’s ρ= -0.47, P=0.005) or between Πn/Πsi and Fop (Spearman’s ρ= -0.45, P=0.02).
4. Discussion
4.1 High Nucleotide diversity
The level of DNA polymorphism observed in this study is one of the highest ever
observed to date, with one SNP every 40 bp in non-coding regions. To our knowledge, the
highest levels of DNA polymorphism reported in the animal kingdom were found in the
nematod Caenorhabditis remanei (Cutter, Baird et al., 2006) and the sea squirts Ciona
savignyi (Small, Brudno et al., 2007) with one SNP every 20 bp. Insect species are also
recognised to often be highly polymorphic. The density of SNPs was reported to be one
every 50bp in Drosophila (Shapiro, Huang et al., 2007) and one every 125 bp in Mosquitoes
(Morlais and Severson, 2003). Our results therefore confirm previous reports on allozyme
diversities which revealed that marine invertebrates are amongst the most diverse animal
species (Ward, Skibinski et al., 1992; Solé-Cava and Thorpe, 1991). Furthermore, DNA
sequence polymorphism provides the opportunity to compare different categories of
substitutions on which selection is expected to act differently. Synonymous and non-coding
6

Page 7

(silent) substitutions that do not result in a change in the protein can be used as a neutral
reference to infer the level of purifying selection that act on non-synonymous (replacement)
substitutions. A theoretical expectation is that selection should more efficiently remove
deleterious mutations in species with large population sizes (Kimura, 1983). The non-
synonymous to silent polymorphism ratio (Πn/Πsi) should therefore be inversely correlated to
the effective population size. In addition, unlike divergence data, polymorphism data should
not be affected by adaptive evolution (McDonald and Kreitman, 1991) and Πn/Πsi is
expected to be a good measure of the proportion of non-synonymous substitutions that are
effectively neutral (Smith and Eyre-Walker, 2002). We found that non-synonymous
substitutions contributed substantially to the polymorphism observed in C. gigas:
Πn/Πsi=16%. This value is much lower than in humans (0.22, Bustamante, Fledel-Alon et al.,
2005), but it is fairly higher to what has been reported in Drosophila (0.12, Shapiro, Huang et
al., 2007) and much higher than in Caenorhabditis remanei (0.09, Cutter, Baird et al., 2006)
and Ciona savignyi (0.07, Small, Brudno et al., 2007). Therefore, purifying selection does not
appear to be as strong as it could have been expected for a species with a large effective
population size. However, this result at the molecular level concurs with the observation of a
high genetic load in this species (Launey and Hedgecock, 2001) and would require further
attention. Unfortunately, without any data on the frequency of substitutions in natural
populations, it is difficult to investigate further the potentially deleterious nature of
segregating non-synonymous polymorphisms (e.g. Eyre-Walker, Woolfit et al., 2006).
Some polymorphisms that affect biological functions can occur outside of coding
regions, in promoters or other regulating regions. Non-coding DNA contains important
sequences for regulatory functions and can be constrained to some extent. Detectable levels
of constraint on non-coding DNA have already been reported in Drosophila where non-
coding diversity is lower than synonymous diversity (Moriyama and Powell, 1996);
(Andolfatto, 2005). The non-coding sequences we studied were in the close vicinity of genes,
mostly in introns. However, contrary to what was reported in Drosophila we observed a
similar level of diversity between the two kinds of silent substitutions. We conclude that
synonymous and non-coding substitutions are equally (un)constrained in the oyster genome.
Recombination might be an interesting parameter to investigate further in the future in
order to reconcile the possible discrepancy observed between high diversity and relaxed
purifying selection. Indeed, intriguing results have been reported in marine bivalves. Guo and
Allen (1996) reported the existence of a single recombinational hot spot in the proximal
region of each chromosome arm in Mulinia lateralis. Hubert and Hedgecock (2004) detected
chromosomal rearrangements in C. gigas and suggested that blocks of the genome may
retain linkage disequilibria for longer than might be expected in putatively large, well-mixed
populations.
4.2 Pattern of Codon Usage Bias
The pattern of codon usage within a genome results from a balance among selection,
random genetic drift and mutation (Bulmer, 1991). Our results in Crassostrea gigas first
suggest that a mutational bias tends to enrich the genome in A and T because the genome is
GC-poor on average (GC3=43%). Secondly, using a moderately large EST database (8800
ESTs) we nonetheless were able to easily detect synonymous codons that significantly
increase with expression levels measured with EST counts (Duret and Mouchiroud, 1999).
Ideally, one may verify that optimal codons correspond to the most abundant tRNAs in the
cell (Ikemura, 1985). Unfortunately, tRNA concentrations in the cell or tRNA gene copy
numbers are not available for C. gigas to date and we had to deal with expression levels
alone. In addition, it should be noted here that expression levels inferred from EST data
seem to be affected by gene length (Munoz, Bogarad et al., 2004). It can potentially be a
problem because gene length and codon bias are correlated in a number of species,
positively in Prokaryotes (Eyre-Walker, 1996) but negatively in Eukaryotes (Moriyama and
Powell, 1998; Duret and Mouchiroud, 1999). Although puzzling the correlation is not
7

Page 8

necessarily the result of a direct link of causality (Duret and Mouchiroud, 1999). It seems
difficult to imagine that a gene length effect could bias the inference of optimal codons for
two reasons: (i) Longer genes are more energetically costly to translate. It seems expected
that they could be underrepresented in the category of highly expressed genes. The fact that
expression levels measured with EST data miss the correlation does not imply that other
variables such as codon bias cannot be studied with EST counts. (ii) Although a significant
correlation between codon bias and gene length has been reported in some genomes, the
correlation is always small, accounting for <5% of the total variance in codon bias (Moriyama
and Powell, 1998).
Most of these codons that we inferred as being optimal for translation were -G or -C
ending codons. As a consequence selection for translational efficiency tends to enrich highly
expressed genes in G and C (GC3=49% for genes with five or more ESTs) when compared
to lowly expressed genes (GC3=43% for genes with one EST). A difference of 6% in GC-
content (or 5% in Fop) between highly and lowly expressed genes may be considered as
small. However, EST count is a very rough assessment of expression level and the inherent
statistical noise of an analysis of a moderately small EST dataset prevents from using this
difference as a reliable estimate of selection intensity. Nonetheless, using a much larger EST
database (114665 ESTs) in Drosophila melanogaster (data from Hey and Kliman, 2002), one
may remark that genes with one EST are already GC-rich (GC3=64%) and that highly
express genes are not tremendously richer in GC (GC3=70% for genes with more than 50
ESTs). Conversely, the simple fact that a significant effect is detected with a small EST
dataset could have been taken as evidence that translational selection is strong in C. gigas.
We observed a strong and significant negative correlation between the frequency of
optimal codons and non-synonymous diversity. Several hypotheses have been proposed to
explain the correlation but an agreement recently emerged to propose that this is a
consequence of correlated selective constraints on synonymous and non-synonymous
substitutions (Bierne and Eyre-Walker, 2006; Stoletzki, Welch et al., 2005) possibly as a
consequence of selection for translational robustness (Drummond, Bloom et al., 2005).
Indeed, Akashi (1994), Marais and Duret (2004) and Stoletzki and Eyre Walker (2007)
provided evidence for selection on translational accuracy in Drosophila, C. elegans and E.
coli respectively. Codon usage bias in C. gigas could therefore be due, at least partly, to
selection for translational accuracy: a synonymous codon tends to be replaced by a more
preferred one to reduce the fitness cost of a misincorporation (minimize the costs of
proofreading and/or the cost of producing incorrect proteins). However, the confirmation of
translational selection for accuracy in C. gigas would require further tests (see Stoletzki and
Eyre-Walker, 2007). Whatever the exact cause of the correlation, our results nonetheless
provide evidence that codon bias as measured by the frequency of optimal codons for
expression might be used as a useful indicator of the selective constraint acting on a protein
in the oyster genome (Stoletzki, Welch et al., 2005).
Ackowledgements
The authors are very indebted to Erick Bazin for the procedure allowing clustering
EST sequences in gene families and thank two anonymous referees for their constructive
remarks. This work was partly funded by the European project Aquafirst (Combined genetic
and functional genomic approaches for stress and disease resistance marker assisted
selection in fish and shellfish, FP6), the Marine Genomics Europe Network of Excellence
(FP6), the French national programme GIS “Génomique Marine” (The Bivalvomix Project,
coord. N. Bierne), the Bureau des Ressources Génétiques (BRG) and the Region Poitou-
Charentes (CPER). The authors would like to thank also Nicole Faury, Tristan Renault,
Viviane Boulo and Arnaud Huvet who provided us some of the primer sequences used in this
paper.
8

Page 9

References
Akashi, H., 1994. Synonymous Codon Usage in Drosophila melanogaster: Natural Selection and
Translational Accuracy. Genetics. 136, 927-935.
Akashi, H., 1995. Inferring weak selection from patterns of polymorphism and divergence at "silent"
sites in Drosophila DNA. Genetics. 139, 1067-76.
Akashi, H., 2001. Gene expression and molecular evolution. Curr Opin Genet Dev. 11, 660-6.
Andolfatto, P., 2005. Adaptive evolution of non-coding DNA in Drosophila. Nature. 437, 1149-52.
Bazin, E., Duret, L., Penel, S. and Galtier, N., 2005. Polymorphix: a sequence polymorphism
database. Nucl. Acids Res. 33, 481-484.
Bazin, E., Glemin, S. and Galtier, N., 2006. Population size does not influence mitochondrial genetic
diversity in animals. Science. 312, 570-2.
Betancourt, A.J. and Presgraves, D.C., 2002. Linkage limits the power of natural selection in
Drosophila. Proc Natl Acad Sci U S A. 99, 13616-20.
Bierne, N., Launey, S., Naciri-Graven, Y. and Bonhomme, F., 1998. Early effect of inbreeding as
revealed by microsatellite analyses on Ostrea edulis larvae. Genetics. 148, 1893-906.
Bierne, N., Tsitrone, A. and David, P., 2000. An inbreeding model of associative overdominance
during a population bottleneck. Genetics. 155, 1981-90.
Bierne, N., Daguin, C., Bonhomme, F., David, P. and Borsa, P. (2003). Direct selection on allozymes
is not required to explain heterogeneity among marker loci across a Mytilus hybrid zone. Mol
Ecol 12(9): 2505-10.
Bierne, N. and Eyre-Walker, A., 2006. Variation in synonymous codon use and DNA polymorphism
within the Drosophila genome. J Evol Biol. 19, 1-11.
Black, W.C.t., Baer, C.F., Antolin, M.F. and DuTeau, N.M., 2001. Population genomics: genome-wide
sampling of insect populations. Annu Rev Entomol. 46, 441-69.
Brookes, A.J., 1999. The essence of SNPs. Gene. 234, 177-86.
Bulmer, M., 1991. The Selection-Mutation-Drift Theory of Synonymous Codon Usage. Genetics. 129,
897-907.
Bustamante, C.D., Fledel-Alon, A., Williamson, S., Nielsen, R., Hubisz, M.T., Glanowski, S.,
Tanenbaum, D.M., White, T.J., Sninsky, J.J., Hernandez, R.D., Civello, D., Adams, M.D.,
Cargill, M. and Clark, A.G., 2005. Natural selection on protein-coding genes in the human
genome. Nature. 437, 1153-7.
Cho, R.J., et al. 1999 Genome-wide mapping with biallelic markers in Arabidopsis thaliana. Nature
Genetics. 23, 203-7.
Coghlan, A. and Wolfe, K.H., 2000. Relationship of codon bias to mRNA concentration and protein
length in Saccharomyces cerevisiae. Yeast. 16, 1131-45.
Cutter, A.D., Baird, S.E. and Charlesworth, D., 2006. High nucleotide polymorphism and rapid decay
of linkage disequilibrium in wild populations of Caenorhabditis remanei. Genetics. 174, 901-
13.
David, P., 1998. Heterozygosity-fitness correlations: new perspectives on old problems. Heredity. 80
(5), 531-7.
David, P., Perdieu, M.A., Pernot, A.F. and Jarne, P., 1997. Spatial and temporal population genetic
structure in the marine bivalve Spisula ovalis. Evolution. 51, 1318-1322.
Dégremont, L., Ernande, B., Bedier, E. and Boudry, P., 2007. Summer mortality of hatchery-produced
Pacific oyster spat (Crassostrea gigas). I. Estimation of genetic parameters for survival and
growth. Aquaculture. 262, 41-53.
Drummond, D.A., Bloom, J.D., Adami, C., Wilke, C.O. and Arnold, F.H., 2005. Why highly expressed
proteins evolve slowly. Proc Natl Acad Sci U S A. 102, 14338-43.
Duret, L., 2002. Evolution of synonymous codon usage in metazoans. Curr Opin Genet Dev. 12, 640-
9.
Duret, L. and Mouchiroud, D., 1999. Expression pattern and, surprisingly, gene length shape codon
usage in Caenorhabditis, Drosophila, and Arabidopsis. Proc Natl Acad Sci U S A. 96, 4482-7.
Eyre-Walker, A., 1996. Synonymous codon bias is related to gene length in Escherichia coli: selection
for translational accuracy? Mol Biol Evol. 13, 864-72.
Eyre-Walker, A., Woolfit, M. and Phelps, T., 2006. The distribution of fitness effects of new deleterious
amino acid mutations in humans. Genetics. 173, 891-900.
Griffin, T.J. and Smith, L.M., 2000. Single-nucleotide polymorphism analysis by MALDI-TOF mass
spectrometry. Trends Biotechnol. 18, 77-84.
Guo, X. and Allen, S.K., Jr., 1996. Complete interference and nonrandom distribution of meiotic
crossover in a mollusc, Mulinia lateralis (Say). Biol Bull. 191, 145-8.
Hall, T., 1997: Bioedit (http://www.mbio.ncsu.edu/BioEdit/bioedit.html).
9

Page 10

Hedgecock, D., 1994: Does variance in reproductive success limit effective population sizes of marine
organisms? In: Beaumont, A.R. (Ed.), Genetics and Evolution of Aquatic Organisms.
Chapman and Hall, London, pp. 122-134.
Hey, J. and Kliman, R.M., 2002. Interactions between natural selection, recombination and gene
density in the genes of Drosophila. Genetics. 160, 595-608.
Hubert, S. and Hedgecock, D., 2004. Linkage maps of microsatellite DNA markers for the Pacific
oyster Crassostrea gigas. Genetics. 168, 351-62.
Ikemura, T., 1982. Correlation between the abundance of yeast transfer RNAs and the occurrence of
the respective codons in protein genes. Differences in synonymous codon choice patterns of
yeast and Escherichia coli with reference to the abundance of isoaccepting transfer RNAs. J
Mol Biol. 158, 573-97.
Ikemura, T., 1985. Codon usage and tRNA content in unicellular and multicellular organisms. Mol Biol
Evol. 2, 13-34.
Ikemura, T., 1992: Correlation between codon usage and tRNA content in microorganisms, in Transfer
RNA in Protein Synthesis. In: Hatfiel, D.L., Lee, B.J. and Pirtle, R.M. (Eds.). CRC, Boca
Raton, FL, pp. 87-111.
Johnson, M.S. and Black, R., 1984. Pattern beneath chaos: the effect of recruitment on genetic
patchiness in an intertidal limpet. Evolution. 38, 1371-1383.
Karl, S.A. and Avise, J.C., 1992. Balancing selection at allozyme loci in oysters: implications from
nuclear RFLPs. Science. 256, 100-2.
Karlin, S. and Mrazek, J., 2000. Predicted highly expressed genes of diverse prokaryotic genomes. J
Bacteriol. 182, 5238-50.
Kimura, M., 1983. The neutral theory of molecular evolution. New York, Cambridge University Press.
Koehn, R.K., Newell, R.I. and Immermann, F., 1980. Maintenance of an aminopeptidase allele
frequency cline by natural selection. Proc Natl Acad Sci U S A. 77, 5385-9.
Launey, S. and Hedgecock, D., 2001. High genetic load in the Pacific oyster Crassostrea gigas.
Genetics. 159, 255-65.
Li, W.H., 1987. Models of nearly neutral mutations with particular implications for non-random usage
of synonymous codons. J Mol Evol. 24, 337-45.
Marais, G., Domazet-Lošo, T., Tautz, D. and Charlesworth, B., 2004. Correlated Evolution of
Synonymous and Nonsynonymous Sites in Drosophila. Journal of Molecular Evolution. 59,
771-779.
Marais, G. and Duret, L., 2001. Synonymous codon usage, accuracy of translation, and gene length in
Caenorhabditis elegans. J Mol Evol. 52, 275-80.
McDonald, J.H., 1996. Detecting non-neutral heterogeneity across a region of DNA sequence in the
ratio of polymorphism to divergence. Mol Biol Evol. 13, 253-60.
McDonald, J.H. and Kreitman, M., 1991. Adaptive protein evolution at the Adh locus in Drosophila.
Nature 351, 652-654.
McVean, G.A.T. and Charlesworth, B., 2000. The Effects of Hill-Robertson Interference Between
Weakly Selected Mutations on Patterns of Molecular Evolution and Variation. Genetics. 155,
929-944.
Merkl, R., 2003. A survey of codon and amino acid frequency bias in microbial genomes focusing on
translational efficiency. J Mol Evol. 57, 453-66.
Mitton, J. B., 1993. Theory and data pertinent to the relationship between heterozygosity and fitness in
The Natural History of Inbreeding and Outbreeding—Theoritical and Empirical Perspectives.
N. W. THORNHILL. Chicago, University of Chicago Press. pp 17-41.
Moriyama, E.N. and Powell, J.R., 1996. Intraspecific nuclear DNA variation in Drosophila. Mol Biol
Evol. 13, 261-77.
Moriyama, E.N. and Powell, J.R., 1998. Gene length and codon usage bias in Drosophila
melanogaster, Saccharomyces cerevisiae and Escherichia coli. Nucleic Acids Res. 26, 3188-
93.
Morlais, I. and Severson, D.W., 2003. Intraspecific DNA variation in nuclear genes of the mosquito
Aedes aegypti. Insect Mol Biol. 12, 631-639.
Munoz, E.T., Bogarad, L.D. and Deem, M.W., 2004. Microarray and EST database estimates of
mRNA expression levels differ: the protein length versus expression curve for C. elegans.
BMC Genomics. 5, 30.
Nairz, K., Stocker, H., Schindelholz, B. and Hafen, E., 2002. High-resolution SNP mapping by
denaturing HPLC. Proc Natl Acad Sci U S A. 99, 10575-80.
Nei, M. and Gojobori, T., 1986. Simple methods for estimating the numbers of synonymous and
nonsynonymous nucleotide substitutions. Mol Biol Evol. 3, 418-26.
Pal, C., Papp, B. and Hurst, L.D., 2001. Highly expressed genes in yeast evolve slowly. Genetics. 158,
927-31.
10

Page 11

Peden, J.F., 1999. CodonW: Analysis of Codon Usage, University of Nottingham.
Picoult-Newberg, L., Ideker, T.E., Pohl, M.G., Taylor, S.L., Donaldson, M.A., Nickerson, D.A. and
Boyce-Jacino, M., 1999. Mining SNPs from EST databases. Genome Res. 9, 167-74.
Plotkin, J.B., Dushoff, J., Desai, M.M. and Fraser, H.B., 2006. Estimating selection pressures from
limited comparative data. Mol Biol Evol. 23, 1457-9.
Rafalski, A., 2002. Applications of single nucleotide polymorphisms in crop genetics. Curr Opin Plant
Biol. 5, 94-100.
Riginos, C., Sukhdeo, K. and Cunningham, C.W., 2002. Evidence for selection at multiple allozyme
loci across a mussel hybrid zone. Mol Biol Evol. 19, 347-51.
Rozas, J. and Rozas, R., 1995. DnaSP, DNA sequence polymorphism: an interactive program for
estimating population genetics parameters from DNA sequence data. Comput Appl Biosci. 11,
621-5.
Rozen, S. and Skaletsky, H., 2000. Primer3 on the WWW for general users and for biologist
programmers. Methods Mol Biol. 132, 365-86.
Sachidanandam, R., et al. 2001. A map of human genome sequence variation containing 1.42 million
single nucleotide polymorphisms. Nature. 409, 928-33.
Shapiro, J.A., Huang, W., Zhang, C., Hubisz, M.J., Lu, J., Turissini, D.A., Fang, S., Wang, H.Y.,
Hudson, R.R., Nielsen, R., Chen, Z. and Wu, C.I., 2007. Adaptive genic evolution in the
Drosophila genomes. Proc Natl Acad Sci U S A. 104, 2271-6.
Sharp, P.M. and Li, W.H., 1987. The rate of synonymous substitution in enterobacterial genes is
inversely related to codon usage bias. Mol Biol Evol. 4, 222-30.
Small, K.S., Brudno, M., Hill, M.M. and Sidow, A., 2007. Extreme genomic variation in a natural
population. Proc Natl Acad Sci U S A. 104, 5698-703.
Smith, N.G. and Eyre-Walker, A., 2002. Adaptive protein evolution in Drosophila. Nature. 415, 1022-4.
Solé-Cava, A.M. and Thorpe, J.P., 1991. High levels of genetic variation in natural populations of
marine lower invertebrates. Biological Journal of the Linnean Society. 44, 65-80.
Stoletzki, N. and Eyre-Walker, A., 2007. Synonymous Codon Usage in Escherichia coli: Selection for
Translational Accuracy. Mol Biol Evol. 24, 374-81.
Stoletzki, N., Welch, J., Hermisson, J. and Eyre-Walker, A., 2005. A Dissection of Volatility in Yeast.
Mol Biol Evol. 22, 2022-2026.
Tenaillon, M.I., Sawkins, M.C., Long, A.D., Gaut, R.L., Doebley, J.F. and Gaut, B.S., 2001. Patterns of
DNA sequence polymorphism along chromosome 1 of maize (Zea mays ssp. mays L.). Proc
Natl Acad Sci U S A. 98, 9161-6.
Urrutia, A.O. and Hurst, L.D., 2001. Codon usage bias covaries with expression breadth and the rate
of synonymous evolution in humans, but this is not evidence for selection. Genetics. 159,
1191-9.
Vignal, A., Milan, D., SanCristobal, M. and Eggen, A., 2002. A review on SNP and other types of
molecular markers and their use in animal genetics. Genetics Selection Evolution. 34, 275-
305.
Ward, R.D., Skibinski, D.O.F. and Woodward, M., 1992. Protein heterozygosity, protein structure, and
taxonomic differentiation. Evol Biol. 26, 73-159.
11

Page 12

12


Figure 1: Relationship between the oysters used in our analysis of polymorphism and codon usage bias

Page 13

0
2
4
6
8
10
12
14
Apolipophorin
Na/K
Inhibitor protein kappa B
Actophorin
Beta Tubuline
Gigasin 2 (exons 1-2)
Notch
Amylase Gene B
GlycoProtein Hormone Receptor
Plastin
Phospholipase C
Arha2
Scavenger
Flavin-containing monooxygenase 2
Myosin Light Chain
BQ426586
HA114
Super Oxyde Dismutase
Macrophage express
Calcium dependant Protein Kinase
Laccase
Vasa-like
Amylase Gene A
Drac 3
Glutathion S Transferase
Cyclophiline
ATPase H+
Ik Cytokine - down regulator of HLA
Tubulin
Glucose 6 Phosphatase
Adrenal gland protein
Elongation Factor 1alpha
kazal-type proteinase inhibitor
Paired type homeodomain protein
Cytochrome P450
S-adenosylhomocysteine hydrolase
Glutaryl-CoA dehydrogenase-like protein
Ciao1 like Protein
BQ427367
Astacin
Integrin
ESTs
Average number of SNP/100bp
Non coding
Coding

Figure 2: Average level of polymorphism detected in coding and non-coding parts of ESTs.
13

Page 14

14


0,000,010,02
0,2
0,3
0,4
0,5
fop
?
?
??
?
?
?
?
?
?
??
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?




Figure 3: Significant correlation between non-synonymous diversity and codon bias
Πns
ρ = -0.47; p=0.005

Page 15

15

Table 1. Nucleotide polymorphism in Crassostrea gigas nuclear genes

Genes
Accession
Number
Pn Ps Pnc
Type of non-
coding
sequence
Ln (bp) Ls (bp) L nc (bp) L cds
Πs
Πn
Πnc
Πsi
Πn/Πsi
Fop GC3
Drac 3
Elongation Factor 1alpha
Adrenal gland protein
Astacin
Cytochrome P450
Tubulin
GlycoProtein Hormone Receptor
Glutaryl-CoA dehydrogenase-like protein
Glutathion S Transferase
Amylase Gene B
BQ427023
AB122066
CK172341
AF075683
AF075692
AB185494
AJ549813
AJ563484
AJ577235
AJ496603
1 1
0 1
0 4
2 1
0 1
2 0
1 1
0 0
0 0
1 1
2
5
5
16
7
4
0
18
11
0
Intron
Intron
Intron
Intron
Intron
Intron
UTR
UTR
Intron
Intron
122
73
114
91
63
78
207
10
70
224
37
20
30
26
21
27
56
32
23
73
78
99
220
175
115
142
67
202
277
88
325
1386
714
752
151
533
3279
216
537
1557
0,027
0,050
0,133
0,038
0,048
0
0,018
0
0
0,014
0,008
0
0,000
0,022
0
0,026
0,005
0
0
0,004
0,026
0,051
0,023
0,091
0,061
0,028
0,000
0,089
0,040
0,000
0,026
0,050
0,036
0,085
0,059
0,024
0,008
0,077
0,037
0,006
0,314
0
0,000
0,260
0
1,083
0,594
0
0
0,719

0,39
0,42
0,33
0,26
0,35
0,28
0,3
0,3
0,37
0,34
0,71
0,48
0,52
0,39
0,6
0,46
0,49
0,41
0,53
0,49
Apolipophorin
Ciao1 like Protein
Vasa-like
ATPase H+
BQ427367
S-adenosylhomocysteine hydrolase
Super Oxyde Dismutase
Glucose 6 Phosphatase
Ik Cytokine - down regulator of HLA
CF369184
AY339886
AY423380
AY551099
BQ427367
BQ427368
AY551094
AM076951

0 0
2 6
0 1
0 0
0 3
0 0
0 1
0 1
0 0
0
6
9
6
15
12
2
5
11
Intron
Intron
Intron
Intron
Intron
Intron
Intron
Intron
Intron
265
154
138
61
79
97
97
100
44
74
40
48
17
23
32
32
29
7
88
107
303
141
187
145
145
117
250
562
564
2274
183
489
470
576
130
312
0 0 0,000
0,056
0,030
0,043
0,080
0,083
0,014
0,043
0,044
0
0,31
0,27
0,35
0,37
0,36
0,28
0,41
0,51
0,31
0,39
0,38
0,43
0,64
0,36
0,52
0,52
0,58
0,59
0,150
0,021
0
0,130
0
0,031
0,034
0
0,013
0
0
0
0
0
0
0
0,082
0,028
0,038
0,086
0,068
0,017
0,041
0,043
0,159
0
0
0
0
0
0
0

Na/K
Phospholipase C
Calcium dependant Protein Kinase
AJ563804

AY713401
0 0
0 0
0 0
0
2
5
Intron
Intron
Intron
79
76
131
20
23
34
98
201
177
100
343
744
0
0
0
0
0
0
0,000
0,010
0,028
0
0,41
0,32
0,31
0,52
0,55
0,51
0,009
0,024
0
0




Inhibitor protein kappa B DQ250326 0 0 0 Intron 42 15 166






10860 0 0,000






0
0,27 0,41
Plastin AF075690 1 0 96 27
177 0 0,010 0 0,19 0,36
Actophorin AF075694 0 0 81 21
201 0 0 0 0,25 0,35
Myosin Light Chain AJ563458 2 0 95 28
477 0 0,021 0 0,31 0,54
Amylase Gene A AJ496597 4 6 206 67
1560 0,090 0,019 0,090 0,217

0,32 0,51
Beta Tubuline AY713400 0 0 118 38
1128 0 0 0 0,46 0,54
BQ426586 BQ426586 2 1 132 39
279 0,026 0,015 0,026 0,591 0,29 0,49