An algorithm for detecting directional and non-directional positive selection,neutrality and negative selection in protein coding DNA sequences

Positive selection or adaptive evolution is thought to be responsible, at least some of the time, for the rapid accumulation of advantageous changes in protein-coding genes. The origin of new enzymatic functions, erection of barriers to heterospecific fertilization, and evasion of host response by pathogens, among other things, are thought to be instances of adaptive evolution. Detecting positive selection in protein-coding genes is fraught with difficulties. Saturation for sequence change, codon usage bias, ephemeral selection events and differential selective pressures on amino acids all contribute to the problem. A number of solutions have been proposed with varying degrees of success, however they suffer from limitations of not being accurate enough or being prohibitively computationally intensive. We have developed a character-based method of identifying lineages that undergo positive selection. In our method we assess the possibility that for each internal branch of a phylogenetic tree an event occurred that subsequently gave rise to a greater number of replacement substitutions than might be expected. We classify these replacement substitutions into two categories – whether they subsequently became invariable or changed again in at least one descendent lineage. The former situation indicates that the new character state is under strong selection to preserve its new identity (directional selection), while the latter situation indicates that there is a persistent pressure to change identity (non-directional selection). The method is fast and accurate, easy to implement, sensitive to short-lived selection events and robust with respect to sampling density and proportion of sites under the influence of positive selection.     

Rate Variation as a Function of Gene Origin in Plastid-Derived Genes of Peridinin-Containing Dinoflagellates

Peridinin-pigmented dinoflagellates contain secondary plastids that seem to have undergone more nearly complete plastid genome reduction than other eukaryotes. Many typically plastid-encoded genes appear to have been transferred to the nucleus, with a few remaining genes found on minicircles. To understand better the evolution of the dinoflagellate plastid, four categories of plastid-associated genes in dinoflagellates were defined based on their history of transfer and evaluated for rate of sequence evolution, including minicircle genes (presumably plastid-encoded), genes probably transferred from the plastid to the nucleus (plastid-transferred), and genes that were likely acquired directly from the nucleus of the previous plastid host (nuclear-transferred). The fourth category, lateral-transferred genes, are plastid-associated genes that do not appear to have a cyanobacterial origin. The evolutionary rates of these gene categories were compared using relative rate tests and likelihood ratio tests. For comparison with other secondary plastid-containing organisms, rates were calculated for the homologous sequences from the haptophyte Emiliania huxleyi. The evolutionary rate of minicircle and plastid-transferred genes in the dinoflagellate was strikingly higher than that of nuclear-transferred and lateral-transferred genes and, also, substantially higher than that of all plastid-associated genes in the haptophyte. Plastid-transferred genes in the dinoflagellate had an accelerated rate of evolution that was variable but, in most cases, not as extreme as the minicircle genes. Furthermore, the nuclear-transferred and lateral-transferred genes showed rates of evolution that are similar to those of other taxa. Thus, nucleus-to-nucleus transferred genes have a more typical rate of sequence evolution, while those whose history was wholly or partially within the dinoflagellate plastid genome have a markedly accelerated rate of evolution. Electronic Supplementary Material Electronic Supplementary material is available for this article at and accessible for authorised users. [Reviewing Editor: Dr. Debashish Battacharya]     

Substantial Regional Variation in Substitution Rates in the Human Genome: Importance of GC Content, Gene Density, and Telomere-Specific Effects

This study presents the first global, 1-Mbp-level analysis of patterns of nucleotide substitutions along the human lineage. The study is based on the analysis of a large amount of repetitive elements deposited into the human genome since the mammalian radiation, yielding a number of results that would have been difficult to obtain using the more conventional comparative method of analysis. This analysis revealed substantial and consistent variability of rates of substitution, with the variability ranging up to twofold among different regions. The rates of substitutions of C or G nucleotides with A or T nucleotides vary much more sharply than the reverse rates, suggesting that much of that variation is due to differences in mutation rates rather than in the probabilities of fixation of C/G vs. A/T nucleotides across the genome. For all types of substitution we observe substantially more hotspots than coldspots, with hotspots showing substantial clustering over tens of Mbp’s. Our analysis revealed that GC-content of surrounding sequences is the best predictor of the rates of substitution. The pattern of substitution appears very different near telomeres compared to the rest of the genome and cannot be explained by the genome-wide correlations of the substitution rates with GC content or exon density. The telomere pattern of substitution is consistent with natural selection or biased gene conversion acting to increase the GC-content of the sequences that are within 10–15 Mbp away from the telomere.Reviewing Editor: Dr. Jerzy Jurka

---

This revised version was published online in July 2005 with corrected page numbers.     

Intron evolution in a phylogenetic perspective: Divergent trends in the two copies of the duplicated def gene in Impatiens L. (Balsaminaceae)

The history of MADS box genes is well-known in angiosperms. While duplication events and gene losses occur frequently, gene structure and intron positions are very conserved. We investigated all six introns in a duplicated MADS box gene (deficiens, def) in selected Impatiens taxa, thereby assessing intron features. For the first time, our study provides a comparison of molecular changes in all introns of a gene from a phylogenetic perspective. Interestingly, a uniform pattern of molecular evolution in the introns of each copy was not observed, but intron length increases, decreases, and size retention can be found in each copy. A tendency to accumulate long autapomorphic indels is also present, thus, a longer intron length does not reflect a higher number of parsimony-informative characters. Substitution rates vary between introns of each gene copy. While four of the six introns of def1 exhibit a change in their substitution rate, five of the six def2 introns maintain their rates throughout the genus albeit at different levels. In MADS box genes several regulatory sequences are found residing in introns. Thus, presence of putative regulatory motifs was investigated. Most of them are not conserved in position and usually present in only one of the gene copies. In addition, the potential for phylogenetic reconstruction of introns in both def copies is shortly discussed.     

The Origin of Trypsin: Evidence for Multiple Gene Duplications in Trypsins

The trypsin family of serine proteases is one of the most studied protein families, with a wealth of amino acid sequence information available in public databases. Since trypsin-like enzymes are widely distributed in living organisms in nature, likely evolutionary scenarios have been proposed. A novel methodology for Fourier transformation of biological sequences (FOTOBIS) is presented. The methodology is well suited for the identification of the size and extent of short repeats in protein sequences. In the present paper the trypsin family of enzymes is analyzed with FOTOBIS and strong evidence for tandem gene duplication is found. A likely evolutionary path for the development of present-day trypsins involved an intrinsic extensive tandem gene duplication of a small DNA fragment of 15–18 nucleotides, corresponding to five or six amino acids. This ancestral trypsin gene was subsequently duplicated, leading to the earliest version of a full-sized trypsin, from which the contemporary trypsins have developed. Received: 22 November 1997 / Accepted: 26 January 1998     

Sequence homologies amongE. coli ribosomal proteins: Evidence for evolutionarily related groupings and internal duplications

Summary The complete or partial sequences of 47E. coli ribosomal proteins described in the literature have been examined by computerized search and matching programs. In contrast to results previously reported by other investigators, sequence homologies were uncovered among some of these ribosomal proteins that are well beyond statistical expectations. Moreover, alignments of the most strongly homologous sequences suggested the existence of a network of family groupings. Several of these proteins also exhibit internal homologies, indicating that they have been elongated by a series of tandem duplications.     

Covariation of Mitochondrial Genome Size with Gene Lengths: Evidence for Gene Length Reduction During Mitochondrial Evolution

Reduction of genome size and gene shortening have been observed in a number of parasitic and mutualistic intracellular symbionts. Reduction of coding capacity is also a unifying principle in the evolutionary history of mitochondria, but little is known about the evolution of gene length in mitochondria. The genes for cytochrome c oxidase subunits I–III, cytochrome b, and the large and small subunit rRNAs are, with very few exceptions, always found on the mitochondrial genome. These resident mitochondrial genes can therefore be used to test whether the reduction in gene lengths observed in a number of intracellular symbionts is also seen in mitochondria. Here we show that resident mitochondrial gene products are shorter than their corresponding counterparts in -proteobacteria and, furthermore, that the reduction of mitochondrial genome size is correlated with a reduction in the length of the corresponding resident gene products. We show that relative genomic AT content, which has been identified as a factor influencing gene lengths in other systems, cannot explain gene length/genome size covariance observed in mitochondria. Our data are therefore in agreement with the idea that gene length evolves as a consequence of selection for smaller genomes, either to avoid accumulation of deleterious mutations or triggered by selection for a replication advantage.     

TheAdh genomic region ofDrosophila ambigua: Evolutionary trends in different species

Summary The study of individual genes is essential to a comprehensive understanding of genome evolution. The wealth of information on alcohol dehydrogenase (Adh) inDrosophila makes this gene particularly suitable for such analysis. We have characterized more than 4 kb of the genomicAdh region inDrosophila ambigua and compared this region toDrosophila mauritiana andDrosophila pseudoobscura. The presence of two genes,Adh and 3ORF (open reading frame), has been confirmed and some of their essential features have been inferred from primary structural analysis. Inter- and intraspecific comparisons have led us to support that both genes may have diverged from an ancient precursor. They appear to be evolving independently, and show a species-specific pattern. TheAdh in theobscura group species lacks amino acids three and four when compared to the species of themelanogaster group and has accumulated most of its amino acid replacements in the third exon. Neither characteristic is observed when any other group species are compared, which suggests that these may be particular features of the evolution of theobscura group. The 3ORF is highly conserved among the three species analyzed, although variability in the length of the third exon and the nucleotide substitution rate, which is much higher than inAdh, are worth noting. According to our data, both mutation/fixation rates and the distribution of mutations vary over time, which makes it difficult to predict the evolutionary dynamics of specific genome regions.     

Identification of the First Fungal Annexin: Analysis of Annexin Gene Duplications and Implications for Eukaryotic Evolution

Annexin homologues have been found in animals, plants, and distinct protist lineages. We report the identification of the first fungal annexin, encoded by the anx14 gene of the filamentous ascomycete Neurospora crassa. Annexins have a complex evolutionary history and exhibit a large number of gene duplications and gene losses in various taxa, including the complete loss of annexin sequences from another ascomycete, the budding yeast Saccharomyces cerevisiae. Surprisingly, the N. crassa annexin homologue is most closely related to the annexin homologue of the slime mold Dictyostelium discoideum, suggesting a phylogenetic link between cellular slime molds and true fungi. Both of these annexin homologues are closely related to the family of annexin homologues present in animals, an observation consistent with the existence of the animal–fungal clade. These data further suggest that the gene duplications that generated the family of annexin sequences present in animals, fungi, and slime molds began prior to the divergence of these taxa. Received: 10 December 1997 / Accepted: 17 April 1998     

Molecular Mechanisms for the Variation of Mitochondrial Gene Content and Gene Arrangement Among Chigger Mites of the Genus Leptotrombidium (Acari: Acariformes)

The gene content of a mitochondrial (mt) genome, i.e., 37 genes and a large noncoding region (LNR), is usually conserved in Metazoa. The arrangement of these genes and the LNR is generally conserved at low taxonomic levels but varies substantially at high levels. We report here a variation in mt gene content and gene arrangement among chigger mites of the genus Leptotrombidium. We found previously that the mt genome of Leptotrombidium pallidum has an extra gene for large-subunit rRNA (rrnL), a pseudo-gene for small-subunit rRNA (PrrnS), and three extra LNRs, additional to the 37 genes and an LNR typical of Metazoa. Further, the arrangement of mt genes of L. pallidum differs drastically from that of the hypothetical ancestor of the arthropods. To find to what extent the novel gene content and gene arrangement occurred in Leptotrombidium, we sequenced the entire or partial mt genomes of three other species, L. akamushi, L. deliense, and L. fletcheri. These three species share the arrangement of all genes with L. pallidum, except trnQ (for tRNA-glutamine). Unlike L. pallidum, however, these three species do not have extra rrnL or PrrnS and have only one extra LNR. By comparison between Leptotrombidium species and the ancestor of the arthropods, we propose that (1) the type of mt genome present in L. pallidum evolved from the type present in the other three Leptotrombidium species, and (2) three molecular mechanisms were involved in the evolution of mt gene content and gene arrangement in Leptotrombidium species. [Reviewing Editor: Dr. Martin Kreitman]     

The local-clock permutation test: a simple test to compare rates of molecular evolution on phylogenetic trees

Rates of molecular evolution vary substantially between lineages, and a growing effort is directed at uncovering the causes and consequences of this variation. Comparing local-clocks (rates of molecular evolution estimated from different sets of branches of a phylogenetic tree) is a common tool in this research effort. Here, I show that a commonly used test (the Likelihood Ratio Test, LRT) will not be statistically valid for comparing local-clocks in most cases. Instead, I propose the local-clock permutation test (LCPT), a simple test that can be used to test the significance of differences between local-clocks. The LCPT could also be used to test for differences between any parameter that can be assigned to individual branches on a phylogenetic tree. Using simulated data, I show that the LCPT has good power to detect differences between local-clocks.     

Choosing the right molecular genetic markers for studying biodiversity: from molecular evolution to practical aspects

The use of molecular genetic markers (MGMs) has become widespread among evolutionary biologists, and the methods of analysis of genetic data improve rapidly, yet an organized framework in which scientists can work is lacking. Elements of molecular evolution are summarized to explain the origin of variation at the DNA level, its measures, and the relationships linking genetic variability to the biological parameters of the studied organisms. MGM are defined by two components: the DNA region(s) screened, and the technique used to reveal its variation. Criteria of choice belong to three categories: (1) the level of variability, (2) the nature of the information (e.g. dominance vs. codominance, ploidy, ... ) which must be determined according to the biological question and (3) some practical criteria which mainly depend on the equipment of the laboratory and experience of the scientist. A three-step procedure is proposed for drawing up MGMs suitable to answer given biological questions, and compiled data are organized to guide the choice at each step: (1) choice, determined by the biological question, of the level of variability and of the criteria of the nature of information, (2) choice of the DNA region and (3) choice of the technique.     

Rapid Evolution by Positive Selection and Gene Gain and Loss: PLA<Subscript>2</Subscript> Venom Genes in Closely Related <Emphasis Type="Italic">Sistrurus</Emphasis> Rattlesnakes with Divergent Diets

Rapid evolution of snake venom genes by positive selection has been reported previously but key features of this process such as the targets of selection, rates of gene turnover, and functional diversity of toxins generated remain unclear. This is especially true for closely related species with divergent diets. We describe the evolution of PLA₂ gene sequences isolated from genomic DNA from four taxa of Sistrurus rattlesnakes which feed on different prey. We identified four to seven distinct PLA₂ sequences in each taxon and phylogenetic analyses suggest that these sequences represent a rapidly evolving gene family consisting of both paralogous and homologous loci with high rates of gene gain and loss. Strong positive selection was implicated as a driving force in the evolution of these protein coding sequences. Exons coding for amino acids that make up mature proteins have levels of variation two to three times greater than those of the surrounding noncoding intronic sequences. Maximum likelihood models of coding sequence evolution reveal that a high proportion (∼30%) of all codons in the mature protein fall into a class of codons with an estimated d _N /d _S (ω) ratio of at least 2.8. An analysis of selection on individual codons identified nine residues as being under strong (p < 0.01) positive selection, with a disproportionately high proportion of these residues found in two functional regions of the PLA₂ protein (surface residues and putative anticoagulant region). This is direct evidence that diversifying selection has led to high levels of functional diversity due to structural differences in proteins among these snakes. Overall, our results demonstrate that both gene gain and loss and protein sequence evolution via positive selection are important evolutionary forces driving adaptive divergence in venom proteins in closely related species of venomous snakes.     

Selection for Antimicrobial Peptide Diversity in Frogs Leads to Gene Duplication and Low Allelic Variation

Antimicrobial peptides are highly diverse pathogen-killing molecules. In many taxa, their evolution is characterized by positive selection and frequent gene duplication. It has been proposed that genes encoding antimicrobial peptides might be subject to balancing selection and/or an enhanced mutation rate, but these hypotheses have not been well evaluated because allelic variation has rarely been studied at antimicrobial peptide loci. We present an evolutionary analysis of novel antimicrobial peptide genes from leopard frogs, Rana. Our results demonstrate that a single genome contains multiple homologous copies, among which there is an excess of nonsynonymous nucleotide site divergence relative to that expected from synonymous site divergence. Thus, we confirm the trends of recurrent duplication and positive selection. Allelic variation is quite low relative to interspecies divergence, indicating a recent positive selective sweep with no evidence of balancing selection. Repeated gene duplication, rather than a balanced maintenance of divergent allelic variants at individual loci, appears to be how frogs have responded to selection for a diverse suite of antimicrobial peptides. Our data also support a pattern of enhanced synonymous site substitution in the mature peptide region of the gene, but we cannot conclude that this is due to an elevated mutation rate.