Bioinformatic approaches were utilized to investigate the pattern of codon usage in 109 epilepsy-related genes. The genes were found to be rich in G and C nucleotides, with GC usage 7% higher than AT. Preferred codon analysis confirmed the dominance of G and C at the wobble position of codons. CAG is the most frequently used codon. The ENc analysis, on the other hand, did not reveal any exceptional codon usage bias. Correlation of GC content at the first and second codon positions with the third codon position (neutrality plot) suggested the impact of selection pressure in shaping the GC enriched compositional pattern observed in epilepsy-related genes. It was further confirmed by correspondence analysis. Altogether, our findings imply that the pattern of evolutionary processes (especially selection pressure) operating on epilepsy-related genes, might aid in the development of novel therapies.
Research Article
Selection pressure dominates the use of synonymous codons in genes linked to epilepsy
https://doi.org/10.21203/rs.3.rs-1601828/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this older preprint version
Codon usage bias
Gene expression
Selection
Epilepsy.
Epilepsy is the fourth most common neurological disorder with more than a million cases per year in India and around 50 million worldwide (Organization 2006). Epilepsy is characterized by improper brain activity that results in abnormal behavior (Engel 2013). For a few neuropsychiatric disorders, variations in gene expression have been linked to brain activity (Darmanis et al. 2015). Wang et al. (2017) identified 84 epilepsy-related genes and discovered that any mutation in these genes results in exhibited syndromes (Wang et al. 2017). A few detailed locus-specific studies reported that eukaryotes have a wide range of mutation rates (Drake 1999). Mutations in the PRRT2 gene cause the manifestation of paroxysmal dyskinesia (Liu et al. 2013). Any alteration due to mutation in the coding portion results in largely synonymous modifications, which is a key predictor of gene expression levels (Zhou et al. 2016). However, there is no information regarding the codon usage bias and compositional properties of the genes linked with epilepsy which will further help in uncovering novel insights on the molecular basis of epilepsy.
Due to the degeneracy of the genetic code, except for methionine and tryptophan, all other amino acids are encoded by more than one codon. However, all the codons encoding the same amino acid (synonymous codons) are not used in equal proportion. This bias in synonymous codon usage is generally termed as codon usage bias (CUB). Previous research revealed that the observed CUB of a gene and genome has a direct link with their expression pattern and stability, respectively (Mishima and Tomari 2016; Novoa and de Pouplana 2012). Extensive research on the CUB pattern of a good number of genomes discovered that codons not only specify the amino acid sequences of a polypeptide chain but also show complex relation with transcript level, protein synthesis, host-pathogen interaction, heterologous gene expression etc. (Gupta and Singh 2021; Roth et al. 2012). Founded by these observations, several genes allied with the human disease were brought under the CUB study. Uddin et al reported that the genes linked with the central nervous system are under strong mutation pressure (Uddin and Chakraborty 2019). Similarly, cancer-related genes are reported to be GC biased, having a higher abundance of GC-ending codons (Uddin et al. 2019). Founded by these observations, we conducted a detailed analysis of the codon usage bias and compositional pattern analysis of epilepsy-related genes. Our findings are useful in understanding the underlying evolutionary mechanisms in epilepsy-related genes, as well as their influence on CUB.
Sequence data
Following a thorough review of the literature and DisGeNET database search, we identified and retrieved information on 109 genes associated with epilepsy (Piñero et al. 2020). The coding sequences were obtained from the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/). All of the retrieved sequences were examined for sequencing errors, as well as initiator and termination codons. Finally, the perfect coding sequences were considered for downstream CUB analysis.
Relative synonymous codon usage (RSCU)
To quantify the difference in codon usage between the observed and expected value without the confounding effect of the amino acid composition of different genes, we calculated the RSCU score of each codon using the formula given by Sharp and Li 1987 (Sharp and Li 1987). Over-represented and under-represented codons were defined as those with an RSCU score of higher than 1.6 and less than 0.6, respectively.
Effective number of codons (ENc)
Gene length brings variation in observed CUB (Moriyama and Powell 1998). To overcome this Wright 1990 proposed a measure of CUB named ENc. ENc is not affected by the gene length and amino acid composition of a gene. ENc was calculated for each of the epilepsy-related genes as per the formula given by Wright 1990 (Wright 1990).
Codon adaptation index (CAI)
The CAI index evaluates the relative merits of each codon using a reference set of highly expressed genes from a species, and a gene's score is generated using the frequency of use of all codons in that gene. The index measures how successful selection has been in shaping the pattern of codon usage and can be used to forecast a gene's level of expression (Sharp and Li 1987). Hence, CAI was used as a measure of gene expression.
GC content analysis
GC content usage frequency varies at each codon position. Therefore, to find their compositional differences at each codon position for the epilepsy-related genes GC1s, GC2s & GC3s, i.e. frequency of the GC content at first, second and third positions, respectively was calculated using the formula given below:
Where GCN represents the frequency of occurrence of G/C at the nth codon position, N is the total number of codons in the coding sequence and NGC is the sum of codons with G or C on the nth codon position.
Neutrality Plot
Most of the mutations at the synonymous third position do not change the final amino acid in the growing polypeptide chain, however, any changes in the first or second codon position bring significant change (amino acids) in the final protein chain. Mutation theory states that it brings a change in the gene sequence randomly, whereas, selection does the reverse. Sueoka 1988 proposed that a neutrality plot, a graphic plot of GC12 versus GC3, reveals the functions of directional mutational pressure against natural selection (Sueoka 1988). A neutrality plot was used to identify the influence of evolutionary forces on CUB.
PR2 analysis
It was proposed by Sueoka in 1995. It measures the impact of evolutionary forces on CUB (Sueoka 1995). Here, the composition of [A3/(A3 + T3)] was plotted against [G3/(G3 + C3)]. In the absence of mutation/selection bias according to PR2, the intra-strand composition will obey the rule A = T and G = C. Therefore, the plot center, in which both coordinates are equal to 0.5, is the position where A is equal to T and G is equal to C. Deviation from the center point suggests uneven pressure of mutation/selection between the two DNA strands (Lobry and Sueoka 2002).
Correspondence analysis (COA)
It is a multivariate statistical analysis widely used to find the trends in CUB. Based on RSCU, each gene was plotted in 59-dimensional vectors. The key trends in the use of codons can be assessed with relative inertia, according to which the coding sequences are analyzed to examine the major factors influencing the codon usage pattern.
Correlation analysis and software used
We measured the relationship between the compositional properties of epilepsy-related genes with the CUB using Karl Pearson correlation analysis. Anaconda 2 software was used to assess the codon-pair context of the epilepsy-related genes (Moura et al. 2005). SPSS software (version 16.0) was used to estimate the correlation values. Past software was used for COA analysis (Frey 2017). All the CUB indices, as well as compositional indices for the epilepsy-related genes were estimated using ACUA and CodonW software (Vetrivel et al. 2007)( http://codonw.sourceforge.net/).
Compositional properties
Our compositional analysis revealed that the genes accountable for inflicting epilepsy are GC biased (53%) (Fig. 1(a)). G showed the highest usage i.e. 26.34%, followed by C (26.14%), A (25.60%), and T (25.60%). The highest GC was found at the 2nd codon position among all codon positions (Fig. 1(b)). According to previous research, GC biased genomes use more GC ending codons, owing to selection pressure (Sun et al. 2017). Individual nucleotide composition at three codon sites indicated that A is utilized mostly at the 1st codon position, C at the 2nd codon position, and G at the 3rd codon position. At the 1st codon position, the four nucleotides were used in the order A > T > C > G. C > G > T > A was the sequence of four nucleotides used at the 2nd codon position. G > A > C > T was the sequence of four nucleotides used at the 3rd codon position. (Figure.1 (c)). Furthermore, we observed a strong correlation between genomic GC and GC3s (r = 0.746, p < 0.01). Correlation analysis between the overall usage frequency of nucleotides and their usage at the synonymous position confirmed the combined impact of both the evolutionary forces (Table 1). Overall, our findings highlight the role of genomic GC in shaping codon nucleotide compositions.
| A3 | T3 | G3 | C3 | GC3 | |
|---|---|---|---|---|---|
| A | 0.987** | 0.952** | 0.927** | 0.841** | -0.192* |
| T | 0.974** | 0.985** | 0.903** | 0.802** | -0.256** |
| G | 0.908** | 0.884** | 0.988** | 0.947** | 0.087 |
| C | 0.859** | 0.856** | 0.944** | 0.974** | 0.127 |
| GC | -0.284** | -0.262** | -0.083 | -0.089 | 0.746** |
RSCU and Synonymous Codon Usage Order Analysis
Synonymous codon usage order analysis was carried out to establish a relationship between SCUO and GC contents (GC1, GC2 and GC3). The SCUO values were plotted against overall GC, GC1, GC2 and GC3 respectively. Interestingly, overall GC and GC3 showed two horns (“U” shaped distribution). In the SCUO plot, the U shape distribution of genes revealed a direct influence of overall GC and GC3 in the codon usage pattern found in epilepsy-related genes (Figure. 2). The RSCU analysis confirmed that the codon CAG had the highest RSCU score while the codon TCG was the most under-represented codon amongst all the investigated codons (Figure. 3). CTG, GAA, GCC, TAT, AAT, TGT, GAT, TTT, GTG, ATC, TTC, GAC, TGC, AAC, TAC, GAG, AAG, CAC, CAG codons were found to be over-represented with RSCU > 1.6. Figure 3 further demonstrates that 14 codons (TCG, TTA, CTA, CGT, GCG, GTA, CGA, CCG, CTT, ACG, TTG, ATA, TCA and AGT) in epilepsy-related genes were under-represented (RSCU < 0.6).
Relationship between RSCU and GC3 composition
Figure. 4 represents the correlation coefficient between codon usage bias and GC3. Codons ending with A and T showed a negative correlation, but codons ending in G and C had a positive correlation. A small number of codons did, however, show some variation. For example, G-ending codons TTG, TGG, TAG, AGG, and the C-ending codon ATC showed negative correlation with GC composition.
Relationship between CAI and ENc
ENc value ranged from 31.28 to 61 with a mean value of 48.25. Similarly, CAI values were found within the range of 0.66 to 0.88 (Avg CAI = 0.77). After analyzing the relationship between CAI and ENc we found that the CAI values were increasing with the decreasing ENc values (r= -0.749**, p < 0.01). It's also worth noting that the majority of the genes have high ENc values, implying that they are less biased and have strong expression patterns.
Parity Rule 2 (PR2) plot analysis
PR2 analysis was used to find the nucleotide bias present at the third codon position. If a gene's codon use bias is influenced only by mutation pressure, all the four nucleotides should be used equally at the wobble position, however, natural selection may not always cause equal use of nucleotides (Sueoka 1996). Interestingly, here we observed scattered distribution of genes away from the center suggesting bias and in particular, we observed GA bias at the synonymous third codon position (Figure. 5).
Neutrality Plot Analysis
Neutrality plot is a method used to measure the relationship between GC3 and GC12 (average of GC1 and GC2). The epileptic genes showed a narrow distribution of GC3s, i.e. 14.13–23.50%. From Fig. 6, it is evident that the slope of the regression is close to zero. These observations suggest that selection pressure played a major role in modulating the codon usage bias of epilepsy-related genes.
Correspondence Analysis
Correspondence analysis was performed based on the RSCU score. This analysis helps in understanding the codon usage differences present in the coding sequences. Axis 1 contributed 35.85% and axis 2 contributed 15.04% of the total variation for CUB of the epilepsy-related genes (Figure. 7). Interestingly, we observed that there was a minute variation in gene position scattering with respect to axis 2 whereas major variation was detected on axis 1. Table 2 shows the relationship between CAI and ENc with axis 1 to axis 10. ENc showed a correlation value of 0.807 with axis 1 at p < 0.05. Similarly, CAI showed a negative correlation value of 0.893 with axis 1 at p < 0.05. These findings imply that the codon usage bias of epilepsy-related genes was regulated by GC composition under the effect of positive selection pressure.
| Axis1 | Axis2 | Axis3 | Axis4 | Axis5 | Axis6 | Axis7 | Axis8 | Axis9 | Axis10 | CAI | ENc | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Axis1 | Pearson Correlation | 1 | .001 | .008 | .010 | − .009 | .003 | .000 | .002 | .002 | .000 | − .893** | .807** |
| Sig. (2-tailed) | .991 | .932 | .920 | .930 | .978 | .997 | .987 | .981 | .996 | .000 | .000 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis2 | Pearson Correlation | .001 | 1 | .003 | .004 | − .006 | − .002 | − .002 | .001 | .005 | .002 | − .066 | − .110 |
| Sig. (2-tailed) | .991 | .973 | .965 | .948 | .987 | .986 | .994 | .961 | .980 | .498 | .253 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis3 | Pearson Correlation | .008 | .003 | 1 | .025 | − .023 | − .005 | − .009 | .001 | .009 | − .004 | − .165 | .158 |
| Sig. (2-tailed) | .932 | .973 | .794 | .814 | .963 | .928 | .990 | .924 | .966 | .086 | .100 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis4 | Pearson Correlation | .010 | .004 | .025 | 1 | − .032 | − .015 | − .011 | − .004 | .017 | − .003 | .045 | − .187 |
| Sig. (2-tailed) | .920 | .965 | .794 | .739 | .874 | .913 | .968 | .864 | .979 | .640 | .051 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis5 | Pearson Correlation | − .009 | − .006 | − .023 | − .032 | 1 | .003 | .007 | .000 | − .011 | .003 | .209* | .078 |
| Sig. (2-tailed) | .930 | .948 | .814 | .739 | .977 | .945 | 1.000 | .908 | .971 | .029 | .418 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis6 | Pearson Correlation | .003 | − .002 | − .005 | − .015 | .003 | 1 | .010 | .003 | .000 | .006 | − .042 | − .020 |
| Sig. (2-tailed) | .978 | .987 | .963 | .874 | .977 | .921 | .978 | .996 | .949 | .663 | .837 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis7 | Pearson Correlation | .000 | − .002 | − .009 | − .011 | .007 | .010 | 1 | − .006 | − .007 | .006 | .063 | .100 |
| Sig. (2-tailed) | .997 | .986 | .928 | .913 | .945 | .921 | .952 | .946 | .951 | .512 | .299 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis8 | Pearson Correlation | .002 | .001 | .001 | − .004 | .000 | .003 | − .006 | 1 | .005 | .000 | .081 | − .078 |
| Sig. (2-tailed) | .987 | .994 | .990 | .968 | 1.000 | .978 | .952 | .955 | .992 | .405 | .419 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis9 | Pearson Correlation | .002 | .005 | .009 | .017 | − .011 | .000 | − .007 | .005 | 1 | .020 | .111 | .026 |
| Sig. (2-tailed) | .981 | .961 | .924 | .864 | .908 | .996 | .946 | .955 | .837 | .252 | .791 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| Axis10 | Pearson Correlation | .000 | .002 | − .004 | − .003 | .003 | .006 | .006 | .000 | .020 | 1 | .020 | .070 |
| Sig. (2-tailed) | .996 | .980 | .966 | .979 | .971 | .949 | .951 | .992 | .837 | .839 | .471 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| CAI | Pearson Correlation | − .893** | − .066 | − .165 | .045 | .209* | − .042 | .063 | .081 | .111 | .020 | 1 | − .760** |
| Sig. (2-tailed) | .000 | .498 | .086 | .640 | .029 | .663 | .512 | .405 | .252 | .839 | .000 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| ENc | Pearson Correlation | .807** | − .110 | .158 | − .187 | .078 | − .020 | .100 | − .078 | .026 | .070 | − .760** | 1 |
| Sig. (2-tailed) | .000 | .253 | .100 | .051 | .418 | .837 | .299 | .419 | .791 | .471 | .000 | ||
| N | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | 109 | |
| **. Correlation is significant at the 0.01 level (2-tailed). | |||||||||||||
| *. Correlation is significant at the 0.05 level (2-tailed). | |||||||||||||
In humans, neurodegenerative illnesses are now fairly common. Several factors like hereditary and environmental influences, and so forth contribute to the development of these disorders. A few studies have demonstrated the impact of mutation in gene expression change and disease development (Armstrong et al. 2019; Gerstung et al. 2015). Recent research has primarily focused on neurodegenerative disorders such as Alzheimer's and Parkinson's, but no work has been published on the pattern of CUB in epilepsy-related genes. Codon usage bias is a common phenomenon that can be found in a wide range of species, from prokaryotes to eukaryotes. It is a multifaceted phenomenon impacted mostly by nucleotide composition (Zhang et al. 2011) and natural selection (Yannai et al. 2018). Apart from these, several other factors influence CUB, for example, environmental factors, tRNA availability etc. (Behura and Severson 2013; Salim and Cavalcanti 2008). The findings of our study would shed light on the molecular intricacies of critical genes linked to epilepsy.
Genomic composition modulates the overall codon usage and amino acid usage pattern recently confirmed in coronavirus genomes (Tort et al. 2020). In mammals, GC content and CpG islands actively regulate the transcription of genes via chromatin conformation (Fenouil et al. 2012). The epilepsy-related genes had an overall GC of 53%, which might have influenced gene expression as seen in our CAI analysis (avg. CAI = 0.77). Individual nucleotide composition analysis at three codon positions showed the highest use of G at the synonymous position. Which is prone to methylation. Existing reports suggest that the nucleotide composition of a gene at a synonymous position follows overall genomic composition due to positive section pressure (Kryazhimskiy et al. 2008). Our findings on nucleotide composition corroborated well with these reports which were further validated by SCUO vs GC plot (Fig. 2) and the correlation observed between genomic GC and GC3s.
RSCU analysis showed the top most used 11 codons (RSCU > 1.6) are either G or C ending. These over-represented codons are GTG (1.936), ATC (2.004), TTC (2.196), GAC (2.228), TGC (2.252), AAC (2.288), TAC (2.324), GAG (2.38), AAG (2.428), CAC (2.472), and CAG (2.984). Similarly, Uddin et al. showed that codons CTG, CAG, TCC, AGC, CGC, ATC, GCC, GGC, ACC, GTG, and CGG are overrepresented in genes related to the central nervous system (Uddin and Chakraborty 2019). These findings are comparable to those of Jie Yang (Yang et al. 2010), who studied codon usage bias in genes related to Alzheimer's disease and discovered that G/C-ending codons were substantially favored, as well as the fact that most optimal codons had either G or C at the 3rd codon position. This suggests that genes linked to neurological disorders have a distinct codon usage pattern. Furthermore, the relationship between AT/GC at synonymous positions and codon usage (Fig. 4) revealed that G/C-ending codons are positively connected to compositional patterns and vice versa by A/T-ending codons. As a result, our findings show that GC composition has a considerable impact on the CUB found in the epilepsy-related genes investigated. Here, the mean ENC observed was 48.25 indicating a moderate CUB. Proopiomelanocortin (ENc = 31.28), ADRA2A (ENc = 32), and OPRD1 (ENc = 32.08) gene showed the highest CUB. A previous study on genes related to central nervous systems showed an average ENc of 40.53 (Uddin and Chakraborty 2019). The ENC values for the albumin superfamily were observed to vary from 51.56 to 56.62 indicating low codon usage bias (Mirsafian et al. 2014). Here, except for 14 genes, all other genes showed an ENc value greater than 40. The high ENc value observed in epilepsy-related genes might be related to the availability of tRNAs maintained by cells expressing these genes for error-free replication and final protein synthesis (Jenkins and Holmes 2003; Kanaya et al. 2001; Novoa and de Pouplana 2012). Furthermore, mRNA stability may be connected to the ENc pattern adopted by epilepsy-associated genes (Burow et al. 2018). The negative relationship between ENc and CAI (r= -0.749**, p < 0.01) suggests that codon usage has a significant influence on epilepsy linked genes expression.
PR2 plot analysis confirmed the G/A nucleotide dominance at the wobble position of epilepsy-related genes. Further, it can be observed in Fig. 5 that genes are clustered in a certain area of the plot, implying a significant degree of selection pressure on the genes under study. However, a few genes were found in the PR2 plot's central area, confirming the role of mutation pressure as previously reported (Rahman et al. 2022). The neutrality plot revealed a small range of GC3s and a regression slope that was close to zero (Fig. 6). This demonstrates that selection pressure has a 62% role in influencing the CUB. According to our PR2 plot and neutrality plot analyses, epileptic genes were subjected to significant selection pressure and modest mutation pressure. Axis 1 from correspondence analysis showed the highest significant correlations with gene expression and codon usage bias of genes. Moreover, the distribution of synonymous codons and genes was found to be in line with axis 2. Taken together, it's conceivable that genes connected to the central nervous system, notably epilepsy, are under great selection pressure.
The nucleotide composition and codon use pattern of the 109 epilepsy genes were analyzed. The results showed that the genes had low codon use bias but high gene expression levels. Over-representations (RSCU > 1.6) were found for GC3-rich codons. While our PR2 plot and neutrality plot analyses suggested that GC selection biases may play a substantial role in codon usage patterns in epilepsy-related genes, we cannot rule out the possibility that mutations played a part in determining the GC/AT ratio. According to our findings, translational selection plays a significant role in optimizing codon usage at high transcription levels. To establish that the CUB pattern found here is a unique property of genes associated with the central nervous system, more study with additional neurological disorders is needed.
Ethical statement:
Funding: NA
Conflict of Interest: The authors declare that they have no conflict of interest.
Ethical approval: NA
Informed consent: NA
Author contribution:
K.P.C, Y.S, and A.N wrote the manuscript. R.T, H.B, S.C, J.S.H, and AS analyzed the datasets. A.K.M, and P.P designed the study and reviewed the manuscript. All authors read, and approved the submitted manuscript.
Acknowledgments
The authors are grateful to Negenome Bio Solutions Pvt. Ltd for providing the necessary facilities to carry out the research work.
- Armstrong N, Ryder S, Forbes C, Ross J, Quek RG (2019) A systematic review of the international prevalence of BRCA mutation in breast cancer Clinical epidemiology 11:543
- Behura SK, Severson DW (2013) Codon usage bias: causative factors, quantification methods and genome-wide patterns: with emphasis on insect genomes Biological Reviews 88:49–61
- Burow DA, Martin S, Quail JF, Alhusaini N, Coller J, Cleary MD (2018) Attenuated codon optimality contributes to neural-specific mRNA decay in Drosophila Cell reports 24:1704–1712
- Darmanis S et al. (2015) A survey of human brain transcriptome diversity at the single cell level Proceedings of the National Academy of Sciences 112:7285–7290
- Drake JW (1999) The distribution of rates of spontaneous mutation over viruses, prokaryotes, and eukaryotes Annals of the New York Academy of Sciences 870:100–107
- Engel J (2013) Seizures and epilepsy vol 83. Oxford University Press,
- Fenouil R et al. (2012) CpG islands and GC content dictate nucleosome depletion in a transcription-independent manner at mammalian promoters Genome research 22:2399–2408
- Frey F (2017) SPSS (software) The International Encyclopedia of Communication Research Methods:1–2
- Gerstung M et al. (2015) Combining gene mutation with gene expression data improves outcome prediction in myelodysplastic syndromes Nature communications 6:1–11
- Gupta S, Singh R (2021) Comparative study of codon usage profiles of Zingiber officinale and its associated fungal pathogens Molecular Genetics and Genomics:1–14
- Jenkins GM, Holmes EC (2003) The extent of codon usage bias in human RNA viruses and its evolutionary origin Virus research 92:1–7
- Kanaya S, Yamada Y, Kinouchi M, Kudo Y, Ikemura T (2001) Codon usage and tRNA genes in eukaryotes: correlation of codon usage diversity with translation efficiency and with CG-dinucleotide usage as assessed by multivariate analysis Journal of molecular evolution 53:290–298
- Kryazhimskiy S, Bazykin GA, Dushoff J (2008) Natural selection for nucleotide usage at synonymous and nonsynonymous sites in influenza A virus genes Journal of virology 82:4938–4945
- Liu XR et al. (2013) Novel PRRT2 mutations in paroxysmal dyskinesia patients with variant inheritance and phenotypes Genes, Brain and Behavior 12:234–240
- Lobry JR, Sueoka N (2002) Asymmetric directional mutation pressures in bacteria Genome biology 3:1–14
- Mirsafian H, Mat Ripen A, Singh A, Teo PH, Merican AF, Mohamad SB (2014) A comparative analysis of synonymous codon usage bias pattern in human albumin superfamily The Scientific World Journal 2014
- Mishima Y, Tomari Y (2016) Codon usage and 3′ UTR length determine maternal mRNA stability in zebrafish Molecular cell 61:874–885
- Moriyama EN, Powell JR (1998) Gene length and codon usage bias in Drosophila melanogaster, Saccharomyces cerevisiae and Escherichia coli Nucleic acids research 26:3188–3193
- Moura G et al. (2005) Comparative context analysis of codon pairs on an ORFeome scale Genome biology 6:1–14
- Novoa EM, de Pouplana LR (2012) Speeding with control: codon usage, tRNAs, and ribosomes Trends in Genetics 28:574–581
- Organization WH (2006) Neurological disorders: public health challenges. World Health Organization,
- Piñero J, Ramírez-Anguita JM, Saüch-Pitarch J, Ronzano F, Centeno E, Sanz F, Furlong LI (2020) The DisGeNET knowledge platform for disease genomics: 2019 update Nucleic acids research 48:D845-D855
- Rahman SU, Abdullah M, Khan AW, Haq MIU, ul Haq N, Aziz A, Tao S (2022) A detailed comparative analysis of codon usage bias in Alongshan virus Virus Research 308:198646
- Roth A, Anisimova M, Cannarozzi GM (2012) Measuring codon usage bias Codon Evolution Mechanisms and Models:189–217
- Salim HM, Cavalcanti AR (2008) Factors influencing codon usage bias in genomes Journal of the Brazilian Chemical Society 19:257–262
- Sharp PM, Li W-H (1987) The codon adaptation index-a measure of directional synonymous codon usage bias, and its potential applications Nucleic acids research 15:1281–1295
- Sueoka N (1988) Directional mutation pressure and neutral molecular evolution Proceedings of the National Academy of Sciences 85:2653–2657
- Sueoka N (1995) Intrastrand parity rules of DNA base composition and usage biases of synonymous codons Journal of molecular evolution 40:318–325
- Sueoka N (1996) Intrastrand parity rules of DNA base composition and usage biases of synonymous codons Journal of Molecular Evolution 42:323–323
- Sun Y, Tamarit D, Andersson SG (2017) Switches in genomic GC content drive shifts of optimal codons under sustained selection on synonymous sites Genome biology and evolution 9:2560–2579
- Tort FL, Castells M, Cristina J (2020) A comprehensive analysis of genome composition and codon usage patterns of emerging coronaviruses Virus research 283:197976
- Uddin A, Chakraborty S (2019) Codon usage pattern of genes involved in central nervous system Molecular neurobiology 56:1737–1748
- Uddin A, Paul N, Chakraborty S (2019) The codon usage pattern of genes involved in ovarian cancer Annals of the New York Academy of Sciences:67–78
- Vetrivel U, Arunkumar V, Dorairaj S (2007) ACUA: a software tool for automated codon usage analysis Bioinformation 2:62
- Wang J et al. (2017) Epilepsy-associated genes Seizure 44:11–20
- Wright F (1990) The ‘effective number of codons’ used in a gene Gene 87:23–29
- Yang J et al. (2010) Codon usage biases in Alzheimer's disease and other neurodegenerative diseases Protein and peptide letters 17:630–645
- Yannai A, Katz S, Hershberg R (2018) The codon usage of lowly expressed genes is subject to natural selection Genome biology and evolution 10:1237–1246
- Zhang J et al. (2011) Analysis of codon usage and nucleotide composition bias in polioviruses Virology Journal 8:1–8
- Zhou Z et al. (2016) Codon usage is an important determinant of gene expression levels largely through its effects on transcription Proceedings of the National Academy of Sciences 113:E6117-E6125
No competing interests reported.