Main

As one might expect from a genus with species living in deserts, in the tropics, on chains of volcanic islands and, often, commensally with humans, Drosophila species vary considerably in their morphology, ecology and behaviour1. Species in this genus span a wide range of global distributions: the 12 sequenced species originate from Africa, Asia, the Americas and the Pacific Islands, and also include cosmopolitan species that have colonized the planet (D. melanogaster and D. simulans) as well as closely related species that live on single islands (D. sechellia)2. A variety of behavioural strategies is also encompassed by the sequenced species, ranging in feeding habit from generalist, such as D. ananassae, to specialist, such as D. sechellia, which feeds on the fruit of a single plant species.

Despite this wealth of phenotypic diversity, Drosophila species share a distinctive body plan and life cycle. Although only D. melanogaster has been extensively characterized, it seems that the most important aspects of the cellular, molecular and developmental biology of these species are well conserved. Thus, in addition to providing an extensive resource for the study of the relationship between sequence and phenotypic diversity, the genomes of these species provide an excellent model for studying how conserved functions are maintained in the face of sequence divergence. These genome sequences provide an unprecedented dataset to contrast genome structure, genome content, and evolutionary dynamics across the well-defined phylogeny of the sequenced species (Fig. 1).

Figure 1: Phylogram of the 12 sequenced species of Drosophila.
figure 1

Phylogram derived using pairwise genomic mutation distances and the neighbour-joining method152,153. Numbers below nodes indicate the per cent of genes supporting a given relationship, based on evolutionary distances estimated from fourfold-degenerate sites (left of solidus) and second codon positions (right of solidus). Coloured blocks indicate support from bayesian (posterior probability (PP), upper blocks) and maximum parsimony (MP; bootstrap values, lower blocks) analyses of data partitioned by chromosome arm. Branch lengths indicate the number of mutations per site (at fourfold-degenerate sites) using the ordinary least squares method. See ref. 154 for a discussion of the uncertainties in the D. yakuba/D. erecta clade.

Genome assembly, annotation and alignment

Genome sequencing and assembly

We used the previously published sequence and updated assemblies for two Drosophila species, D. melanogaster3,4 (release 4) and D. pseudoobscura5 (release 2), and generated DNA sequence data for 10 additional Drosophila genomes by whole-genome shotgun sequencing6,7. These species were chosen to span a wide variety of evolutionary distances, from closely related pairs such as D. sechellia/D. simulans and D. persimilis/D. pseudoobscura to the distantly related species of the Drosophila and Sophophora subgenera. Whereas the time to the most recent common ancestor of the sequenced species may seem small on an evolutionary timescale, the evolutionary divergence spanned by the genus Drosophila exceeds that of the entire mammalian radiation when generation time is taken into account, as discussed further in ref. 8. We sequenced seven of the new species (D. yakuba, D. erecta, D. ananassae, D. willistoni, D. virilis, D. mojavensis and D. grimshawi) to deep coverage (8.4× to 11.0×) to produce high quality draft sequences. We sequenced two species, D. sechellia and D. persimilis, to intermediate coverage (4.9× and 4.1×, respectively) under the assumption that the availability of a sister species sequenced to high coverage would obviate the need for deep sequencing without sacrificing draft genome quality. Finally, seven inbred strains of D. simulans were sequenced to low coverage (2.9× coverage from w501 and 1× coverage of six other strains) to provide population variation data9. Further details of the sequencing strategy can be found in Table 1, Supplementary Table 1 and section 1 in Supplementary Information.

Table 1 A summary of sequencing and assembly properties of each new genome

We generated an initial draft assembly for each species using one of three different whole-genome shotgun assembly programs (Table 1). For D. ananassae, D. erecta, D. grimshawi, D. mojavensis, D. virilis and D. willistoni, we also generated secondary assemblies; reconciliation of these with the primary assemblies resulted in a 7–30% decrease in the estimated number of misassembled regions and a 12–23% increase in the N50 contig size10 (Supplementary Table 2). For D. yakuba, we generated 52,000 targeted reads across low-quality regions and gaps to improve the assembly. This doubled the mean contig and scaffold sizes and increased the total fraction of high quality bases (quality score (Q) > 40) from 96.5% to 98.5%. We improved the initial 2.9× D. simulans w501 whole-genome shotgun assembly by filling assembly gaps with contigs and unplaced reads from the 1× assemblies of the six other D. simulans strains, generating a ‘mosaic’ assembly (Supplementary Table 3). This integration markedly improved the D. simulans assembly: the N50 contig size of the mosaic assembly, for instance, is more than twice that of the initial w501 assembly (17 kb versus 7 kb).

Finally, one advantage of sequencing genomes of multiple closely related species is that these evolutionary relationships can be exploited to dramatically improve assemblies. D. yakuba and D. simulans contigs and scaffolds were ordered and oriented using pairwise alignment to the well-validated D. melanogaster genome sequence (Supplementary Information section 2). Likewise, the 4–5× D. persimilis and D. sechellia assemblies were improved by assisted assembly using the sister species (D. pseudoobscura and D. simulans, respectively) to validate both alignments between reads and linkage information. For the remaining species, comparative syntenic information, and in some cases linkage information, were also used to pinpoint locations of probable genome mis-assembly, to assign assembly scaffolds to chromosome arms and to infer their order and orientation along euchromatic chromosome arms, supplementing experimental analysis based on known markers (A. Bhutkar, S. Russo, S. Schaeffer, T. F. Smith and W. M. Gelbart, personal communication) (Supplementary Information section 2).

The mitochondrial (mt)DNA of D. melanogaster, D. sechellia, D. simulans (siII), D. mauritiana (maII) and D. yakuba have been previously sequenced11,12. For the remaining species (except D. pseudoobscura, the DNA from which was prepared from embryonic nuclei), we were able to assemble full mitochondrial genomes, excluding the A+T-rich control region (Supplementary Information section 2)13. In addition, the genome sequences of three Wolbachia endosymbionts (Wolbachia wSim, Wolbachia wAna and Wolbachia wWil) were assembled from trace archives, in D. simulans, D. ananassae and D. willistoni, respectively14. All of the genome sequences described here are available in FlyBase (www.flybase.org) and GenBank (www.ncbi.nlm.nih.gov) (Supplementary Tables 4 and 5).

Repeat and transposable element annotation

Repetitive DNA sequences such as transposable elements pose challenges for whole-genome shotgun assembly and annotation. Because the best approach to transposable element discovery and identification is still an active and unresolved research question, we used several repeat libraries and computational strategies to estimate the transposable element/repeat content of the 12 Drosophila genome assemblies (Supplementary Information section 3). Previously curated transposable element libraries in D. melanogaster provided the starting point for our analysis; to limit the effects of ascertainment bias, we also developed de novo repeat libraries using PILER-DF15,16 and ReAS17. We used four transposable element/repeat detection methods (RepeatMasker, BLASTER-TX, RepeatRunner and CompTE) in conjunction with these transposable element libraries to identify repetitive elements in non-melanogaster species. We assessed the accuracy of each method by calibration with the estimated 5.5% transposable element content in the D. melanogaster genome, which is based on a high-resolution transposable element annotation18 (Supplementary Fig. 1). On the basis of our results, we suggest a hybrid strategy for new genome sequences, employing translated BLAST with general transposable element libraries and RepeatMasker with species-specific ReAS libraries to estimate the upper and lower bound on transposable element content.

Protein-coding gene annotation

We annotated protein-coding sequences in the 11 non-melanogaster genomes, using four different de novo gene predictors (GeneID19, SNAP20, N-SCAN21 and CONTRAST22); three homology-based predictors that transfer annotations from D. melanogaster (GeneWise23, Exonerate24, GeneMapper25); and one predictor that combined de novo and homology-based evidence (Gnomon26). These gene prediction sets were combined using GLEAN, a gene model combiner that chooses the most probable combination of start, stop, donor and acceptor sites from the input predictions27,28. All analyses reported here, unless otherwise noted, relied on a reconciled consensus set of predicted gene models—the GLEAN-R set (Table 2, and Supplementary Information section 4.1).

Table 2 A summary of annotated features across all 12 genomes

Quality of gene models

As the first step in assessing the quality of the GLEAN-R gene models, we used expression data from microarray experiments on adult flies, with arrays custom-designed for D. simulans, D. yakuba, D. ananassae, D. pseudoobscura, D. virilis and D. mojavensis29 (GEO series GSE6640; Supplementary Information section 4.2). We detected expression significantly above negative controls (false-discovery-rate-corrected Mann–Whitney U (MWU) P < 0.001) for 77–93% of assayed GLEAN-R models, representing 50–68% of the total GLEAN-R predictions in each species (Supplementary Table 6). Evolutionarily conserved gene models are much more likely to be expressed than lineage-specific ones (Fig. 2). Although these data cannot confirm the detailed structure of gene models, they do suggest that the majority of GLEAN-R models contain sequence that is part of a poly-adenylated transcript. Approximately 20% of transcription in D. melanogaster seems to be unassociated with protein-coding genes30, and our microarray experiments fail to detect conditionally expressed genes. Thus, transcript abundance cannot conclusively establish the presence or absence of a protein-coding gene. Nonetheless, we believe these expression data increase our confidence in the reliability of the GLEAN-R models, particularly those supported by homology evidence (Fig. 2).

Figure 2: Gene models in 12 Drosophila genomes.
figure 2

Number of gene models that fall into one of five homology classes: single-copy orthologues in all species (single-copy orthologues), conserved in all species as orthologues or paralogues (conserved homologues), a D. melanogaster homologue, but not found in all species (patchy homologues with mel.), conserved in at least two species but without a D. melanogaster homologue (patchy homologues, no mel.), and found only in a single lineage (lineage specific). For those species with expression data29, pie charts indicate the fraction of genes in each homology class that fall into one of four evidence classes (see text for details).

Because the GLEAN-R gene models were built using assemblies that were not repeat masked, it is likely that some proportion of gene models are false positives corresponding to coding sequences of transposable elements. We used RepeatMasker with de novo ReAS libraries and PFAM structural annotations of the GLEAN-R gene set to flag potentially transposable element-contaminated gene models (Supplementary Information section 4.2). These procedures suggest that 5.6–32.3% of gene models in non-melanogaster species correspond to protein-coding content derived from transposable elements (Supplementary Table 7); these transposable element-contaminated gene models are almost exclusively confined to gene predictions without strong homology support (Fig. 2). Transposable element-contaminated gene models are excluded from the final gene prediction set used for subsequent analysis, unless otherwise noted.

Homology assignment

Two independent approaches were used to assign orthology and paralogy relationships among euchromatic D. melanogaster gene models and GLEAN-R predictions. The first approach was a fuzzy reciprocal BLAST (FRB) algorithm, which is an extension of the reciprocal BLAST method31 applicable to multiple species simultaneously (Supplementary Information section 5.1). Because the FRB algorithm does not integrate syntenic information, we also used a second approach based on Synpipe (Supplementary Information section 5.2), a tool for synteny-aided orthology assignment32. To generate a reconciled set of homology calls, pairwise Synpipe calls (between each species and D. melanogaster) were mapped to GLEAN-R models, filtered to retain only 1:1 relationships, and added to the FRB calls when they did not conflict and were non-redundant. This reconciled FRB + Synpipe set of homology calls forms the basis of our subsequent analyses. There were 8,563 genes with single-copy orthologues in the melanogaster group and 6,698 genes with single-copy orthologues in all 12 species; similar numbers of genes were also obtained with an independent approach33. Most single-copy orthologues are expressed and are free from potential transposable element contamination, suggesting that the reconciled orthologue set contains robust and high-quality gene models (Fig. 2).

Validation of homology calls

Because both the FRB algorithm and Synpipe rely on BLAST-based methods to infer similarities, rapidly evolving genes may be overlooked. Moreover, assembly gaps and poor-quality sequence may lead to erroneous inferences of gene loss. To validate putative gene absences, we used a synteny-based GeneWise pipeline to find potentially missed homologues of D. melanogaster proteins (Supplementary Information section 5.4). Of the 21,928 cases in which a D. melanogaster gene was absent from another species in the initial homology call set, we identified plausible homologues for 13,265 (60.5%), confirmed 4,546 (20.7%) as genuine absences, and were unable to resolve 4,117 (18.8%). Because this approach is conservative and only confirms strongly supported absences, we are probably underestimating the number of genuine absences.

Coding gene alignment and filtering

Investigating the molecular evolution of orthologous and paralogous genes requires accurate multi-species alignments. Initial amino acid alignments were generated using TCOFFEE34 and converted to nucleotide alignments (Supplementary Table 8). To reduce biases in downstream analyses, a simple computational screen was developed to identify and mask problematic regions of each alignment (Supplementary Information section 6). Overall, 2.8% of bases were masked in the melanogaster group alignments, and 3.0% of bases were masked in the full 12 species alignments, representing 8.5% and 13.8% of alignment columns, respectively. The vast majority of masked bases are masked in no more than one species (Supplementary Fig. 3), suggesting that the masking procedure is not simply eliminating rapidly evolving regions of the genome. We find an appreciably higher frequency of masked bases in lower-quality D. simulans and D. sechellia assemblies, compared to the more divergent (from D. melanogaster) but higher-quality D. erecta and D. yakuba assemblies, suggesting a higher error rate in accurately predicting and aligning gene models in lower-quality assemblies (Supplementary Information section 6 and Supplementary Fig. 3). We used masked versions of the alignments, including only the longest D. melanogaster transcripts for all subsequent analysis unless otherwise noted.

Annotation of non-coding (nc)RNA genes

Using de novo and homology-based approaches we annotated over 9,000 ncRNA genes from recognized ncRNA classes (Table 2, and Supplementary Information section 7). In contrast to the large number of predictions observed for many ncRNA families in vertebrates (due in part to large numbers of ncRNA pseudogenes35,36), the number of ncRNA genes per family predicted by RFAM and tRNAscan in Drosophila is relatively low (Table 2). This suggests that ncRNA pseudogenes are largely absent from Drosophila genomes, which is consistent with the low number of protein-coding pseudogenes in Drosophila37. The relatively low numbers of some classes of ncRNA genes (for example, small nucleolar (sno)RNAs) in the Drosophila subgenus are likely to be an artefact of rapid rates of evolution in these types of genes and the limitation of the homology-based methods used to annotate distantly related species.

Evolution of genome structure

Coarse-level similarities among Drosophilids

At a coarse level, genome structure is well conserved across the 12 sequenced species. Total genome size estimated by flow cytometry varies less than threefold across the phylogeny, ranging from 130 Mb (D. mojavensis) to 364 Mb (D. virilis)38 (Table 2), in contrast to the order of magnitude difference between Drosophila and mammals. Total protein-coding sequence ranges from 38.9 Mb in D. melanogaster to 65.4 Mb in D. willistoni. Intronic DNA content is also largely conserved, ranging from 19.6 Mb in D. simulans to 24.0 Mb in D. pseudoobscura (Table 2). This contrasts dramatically with transposable element-derived genomic DNA content, which varies considerably across genomes (Table 2) and correlates significantly with euchromatic genome size (estimated as the summed length of contigs > 200 kb) (Kendall's τ = 0.70, P = 0.0016).

To investigate overall conservation of genome architecture at an intermediate scale, we analysed synteny relationships across species using Synpipe32 (Supplementary Information section 9.1). Synteny block size and average number of genes per block varies across the phylogeny as expected, with the number of blocks increasing and the average size of blocks decreasing with increasing evolutionary distance from D. melanogaster (A. Bhutkar, S. Russo, T. F. Smith and W. M. Gelbart, personal communication) (Supplementary Fig. 4). We inferred 112 syntenic blocks between D. melanogaster and D. sechellia (with an average of 122 genes per block), compared to 1,406 syntenic blocks between D. melanogaster and D. grimshawi (with an average of 8 genes per block). On average, 66% of each genome assembly was covered by syntenic blocks, ranging from 68% in D. sechellia to 58% in D. grimshawi.

Similarity across genomes is largely recapitulated at the level of individual genes, with roughly comparable numbers of predicted protein-coding genes across the 12 species (Table 2). The majority of predicted genes in each species have homologues in D. melanogaster (Table 2, Supplementary Table 9). Moreover, most of the 13,733 protein-coding genes in D. melanogaster are conserved across the entire phylogeny: 77% have identifiable homologues in all 12 genomes, 62% can be identified as single-copy orthologues in the six genomes of the melanogaster group and 49% can be identified as single-copy orthologues in all 12 genomes. The number of functional non-coding RNA genes predicted in each Drosophila genome is also largely conserved, ranging from 584 in D. mojavensis to 908 in D. ananassae (Table 2).

There are several possible explanations for the observed interspecific variation in gene content. First, approximately 700 D. melanogaster gene models have been newly annotated since the FlyBase Release 4.3 annotations used in the current study, reducing the discrepancy between D. melanogaster and the other sequenced genomes in this study. Second, because low-coverage genomes tend to have more predicted gene models, we suspect that artefactual duplication of genomic segments due to assembly errors inflates the number of predicted genes in some species. Finally, the non-melanogaster species have many more predicted lineage-specific genes than D. melanogaster, and it is possible that some of these are artefactual. In the absence of experimental evidence, it is difficult to distinguish genuine lineage-specific genes from putative artefacts. Future experimental work will be required to fully disentangle the causes of interspecific variation in gene number.

Abundant genome rearrangements during Drosophila evolution

To study the structural relationships among genomes on a finer scale, we analysed gene-level synteny between species pairs. These synteny maps allowed us to infer the history and locations of fixed genomic rearrangements between species. Although Drosophila species vary in their number of chromosomes, there are six fundamental chromosome arms common to all species. For ease of denoting chromosomal homology, these six arms are referred to as ‘Muller elements’ after Hermann J. Muller, and are denoted A–F. Although most pairs of orthologous genes are found on the same Muller element, there is extensive gene shuffling within Muller elements between even moderately diverged genomes (Fig. 3, and Supplementary Information section 9.1).

Figure 3: Synteny plots for Muller elements B and C with respect to D. melanogaster gene order.
figure 3

The horizontal axis shows D. melanogaster gene order for Muller elements B and C, and the vertical axis maps homologous locations32,155 in individual species (af in increasing evolutionary distance from D. melanogaster). Left to right on the x axis is from telomere to centromere for Muller element B, followed by Muller element C from centromere to telomere. Red and green lines represent syntenic segments in the same or reverse orientation along the chromosome relative to D. melanogaster, respectively. Blue segments show gene transposition of genes from one element to the other.

Previous analysis has revealed heterogeneity in rearrangement rates among close relatives: careful inspection of 29 inversions that differentiate the chromosomes of D. melanogaster and D. yakuba revealed that 28 were fixed in the lineage leading to D. yakuba, and only one was fixed on the lineage leading to D. melanogaster39. Rearrangement rates are also heterogeneous across the genome among the 12 species: simulations reject a random-breakage model, which assumes that all sites are free to break in inversion events, but fail to reject a model of coldspots and hotspots for breakpoints (S. Schaeffer, personal communication). Furthermore, inversions seem to have played important roles in the process of speciation in at least some of these taxa40.

One particularly striking example of the dynamic nature of genome micro-structure in Drosophila is the homeotic homeobox (Hox) gene cluster(s)41. Hox genes typically occur in genomic clusters, and this clustering is conserved across many vertebrate and invertebrate taxa, suggesting a functional role for the precise and collinear arrangement of these genes. However, several cluster splits have been previously identified in Drosophila42,43, and the 12 Drosophila genome sequences provide additional evidence against the functional importance of Hox gene clustering in Drosophila. There are seven different gene arrangements found across 13 Drosophila species (the 12 sequenced genomes and D. buzzatii), with no species retaining the inferred ancestral gene order44. It thus seems that, in Drosophila, Hox genes do not require clustering to maintain proper function, and are a powerful illustration of the dynamism of genome structure across the sequenced genomes.

Transposable element evolution

Mobile, repetitive transposable element sequences are a particularly dynamic component of eukaryotic genomes. Transposable element/repeat content (in scaffolds >200 kb) varies by over an order of magnitude across the genus, ranging from 2.7% in D. simulans and D. grimshawi to 25% in D. ananassae (Table 2, and Supplementary Fig. 1). These data support the lower euchromatic transposable element content in D. simulans relative to D. melanogaster45, and reveal that euchromatic transposable element/repeat content is generally similar within the melanogaster subgroup. Within the Drosophila subgenus, D. grimshawi has the lowest transposable element/repeat content, possibly relating to its ecological status as an island endemic, which may minimize the chance for horizontal transfer of transposable element families. Finally, the highest levels of transposable element/repeat content are found in D. ananassae and D. willistoni. These species also have the highest numbers of pseudo-transfer (t)RNA genes (Table 2), indicating a potential relationship between pseudo-tRNA genesis and repetitive DNA, as has been established in the mouse genome36.

Different classes of transposable elements can vary in abundance owing to a variety of host factors, motivating an analysis of the intragenomic ecology of transposable elements in the 12 genomes. In D. melanogaster, long terminal repeat (LTR) retrotransposons have the highest abundance, followed by LINE (long interspersed nuclear element)-like retrotransposons and terminal inverted repeat (TIR) DNA-based transposons18. An unbiased, conservative approach (Supplementary Information section 3) for estimating the rank order abundance of major transposable element classes suggests that these abundance trends are conserved across the entire genus (Supplementary Fig. 5). Two exceptions are an increased abundance of TIR elements in D. erecta and a decreased abundance of LTR elements in D. pseudoobscura; the latter observation may represent an assembly artefact because the sister species D. persimilis shows typical LTR abundance. Given that individual instances of transposable element repeats and transposable element families themselves are not conserved across the genus, the stability of abundance trends for different classes of transposable elements is striking and suggests common mechanisms for host–transposable element co-evolution in Drosophila.

Although comprehensive analysis of the structural and evolutionary relationships among families of transposable elements in the 12 genomes remains a major challenge for Drosophila genomics, some initial insights can be gleaned from analysis of particularly well-characterized transposable element families. Previous analysis has shown variable dynamics for the most abundant transposable element family (DINE-1)46 in the D. melanogaster genome18,47: although inactive in D. melanogaster48, DINE-1 has experienced a recent transpositional burst in D. yakuba49. Our analysis confirms that this element is highly abundant in all of the other sequenced genomes of Drosophila, but is not found outside of Diptera50,51. Moreover, the inferred phylogenetic relationship of DINE-1 paralogues from several Drosophila species suggests vertical transmission as the major mechanism for DINE-1 propagation. Likewise, analysis of the Galileo and 1360 transposons reveals a widespread but discontinuous phylogenetic distribution for both families, notably with both families absent in the geographically isolated Hawaiian species, D. grimshawi52. These results are consistent with an ancient origin of the Galileo and 1360 families in the genus and subsequent horizontal transfer and/or loss in some lineages.

The use of these 12 genomes also facilitated the discovery of transposable element lineages not yet documented in Drosophila, specifically the P instability factor (PIF) superfamily of DNA transposons. Our analysis indicates that there are four distinct lineages of this transposon in Drosophila, and that this element has indeed colonized many of the sequenced genomes53. This superfamily is particularly intriguing given that PIF-transposase-like genes have been implicated in the origin of at least seven different genes during the Drosophila radiation53, suggesting that not only do transposable elements affect the evolution of genome structure, but that their domestication can play a part in the emergence of novel genes.

D. melanogaster maintains its telomeres by occasional targeted transposition of three telomere-specific non-LTR retrotransposons (HeT-A, TART and TAHRE) to chromosome ends54,55 and not by the more common mechanism of telomerase-generated G-rich repeats56. Multiple telomeric retrotransposons have originated within the genus, where they now maintain telomeres, and recurrent loss of most of the ORF2 from telomeric retrotransposons (for example, TAHRE) has given rise to half-telomeric-retrotransposons (for example, HeT-A) during Drosophila evolution57. The phylogenetic relationship among these telomeric elements is congruent with the species phylogeny, suggesting that they have been vertically transmitted from a common ancestor57.

ncRNA gene family evolution

Using ncRNA gene annotations across the 12-species phylogeny, we inferred patterns of gene copy number evolution in several ncRNA families. Transfer RNA genes are the most abundant family of ncRNA genes in all 12 genomes, with 297 tRNAs in D. melanogaster and 261–484 tRNA genes in the other species (Table 2). Each genome encodes a single selenocysteine tRNA, with the exception of D. willistoni, which seems to lack this gene (R. Guigo, personal communication). Elevated tRNA gene counts in D. ananassae and D. willistoni are explained almost entirely by pseudo-tRNA gene predictions. We infer from the lack of pseudo-tRNAs in most Drosophila species, and from similar numbers of tRNAs obtained from an analysis of the chicken genome (n = 280)58, that the minimal metazoan tRNA set is encoded by 300 genes, in contrast to previous estimates of 497 in human and 659 in Caenorhabditis elegans59,60. Similar numbers of snoRNAs are predicted in the D. melanogaster subgroup (n = 242–255), in which sequence similarity is high enough for annotation by homology, with fewer snoRNAs (n = 194–216) annotated in more distant members of the Sophophora subgenus, and even fewer snoRNAs (n = 139–165) predicted in the Drosophila subgenus, in which annotation by homology becomes much more difficult.

Of 78 previously reported micro (mi)RNA genes, 71 (91%) are highly conserved across the entire genus, with the remaining seven genes (mir-2b-1, -289, -303, -310, -311, -312 and -313) restricted to the subgenus Sophophora (Supplementary Information section 7.2). All the species contain similar numbers of spliceosomal snRNA genes (Table 2), including at least one copy each of the four U12-dependent (minor) spliceosomal RNAs, despite evidence for birth and death of these genes and the absence of stable subtypes61. The unusual, lineage-specific expansion in size of U11 snRNA, previously described in Drosophila61,62, is even more extreme in D. willistoni. We annotated 99 copies of the 5S ribosomal (r)RNA gene in a cluster in D. melanogaster, and between 13 and 73 partial 5S rRNA genes in clusters in the other genomes. Finally, we identified members of several other classes of ncRNA genes, including the RNA components of the RNase P (1 per genome) and the signal recognition particle (SRP) RNA complexes (1–3 per genome), suggesting that these functional RNAs are involved in similar biological processes throughout the genus. We were only able to locate the roX (RNA on X)63,64 genes involved in dosage compensation using nucleotide homology in the melanogaster subgroup, although analyses incorporating structural information have identified roX genes in other members of the genus65.

We investigated the evolution of rRNA genes in the 12 sequenced genomes, using trace archives to locate sequence variants within the transcribed portions of these genes. This analysis revealed moderate levels of variation that are not distributed evenly across the rRNA genes, with fewest variants in conserved core coding regions, more variants in coding expansion regions, and higher still variant abundances in non-coding regions. The level and distribution of sequence variation in rRNA genes are suggestive of concerted evolution, in which recombination events uniformly distribute variants throughout the rDNA loci, and selection dictates the frequency to which variants can expand66.

Protein-coding gene family evolution

For a general perspective on how the protein-coding composition of these 12 genomes has changed, we examined gene family expansions and contractions in the 11,434 gene families (including those of size one in each species) predicted to be present in the most recent common ancestor of the two subgenera. We applied a maximum likelihood model of gene gain and loss67 to estimate rates of gene turnover. This analysis suggests that gene families expand or contract at a rate of 0.0012 gains and losses per gene per million years, or roughly one fixed gene gain/loss across the genome every 60,000 yr68. Many gene families (4,692 or 41.0%) changed in size in at least one species, and 342 families showed significantly elevated (P < 0.0001) rates of gene gain and loss compared to the genomic average, indicating that non-neutral processes may play a part in gene family evolution. Twenty-two families exhibit rapid copy number evolution along the branch leading to D. melanogaster (eighteen contractions and four expansions; Supplementary Table 10). The most common Gene Ontology (GO) terms among families with elevated rates of gain/loss include ‘defence response’, ‘protein binding’, ‘zinc ion binding’, ‘proteolysis’, and ‘trypsin activity’. Interestingly, genes involved in ‘defence response’ and ‘proteolysis’ also show high rates of protein evolution (see below). We also found heterogeneity in overall rates of gene gain and loss across lineages, although much of this variation could result from interspecific differences in assembly quality68.

Lineage-specific genes

The vast majority of D. melanogaster proteins that can be unambiguously assigned a homology pattern (Supplementary Information section 5) are inferred to be ancestrally present at the genus root (11,348/11,644, or 97.5%). Of the 296 non-ancestrally present genes, 252 are either Sophophora-specific, or have a complicated pattern of homology requiring more than one gain and/or loss on the phylogeny, and are not discussed further. The remaining 44 proteins include 14 present in the melanogaster group, 23 present only in the melanogaster subgroup, 3 unique to the melanogaster species complex, and 4 found in D. melanogaster only. Because we restricted this analysis to unambiguous homologues of high-confidence protein-coding genes in D. melanogaster8, we are probably undercounting the number of genes that have arisen de novo in any particular lineage. However, ancestrally heterochromatic genes that are currently euchromatic in D. melanogaster may spuriously seem to be lineage-specific.

The 44 lineage-specific genes (Supplementary Table 11) differ from ancestrally present genes in several ways. They have a shorter median predicted protein length (lineage-specific median 177 amino acids, other median 421 amino acids, MWU, P = 3.6 × 10-13), are more likely to be intronless (Fisher's exact test (FET), P = 6.2 × 10-6), and are more likely to be located in the intron of another gene on the opposite strand (FET, P = 3.5 × 10-4). In addition, 18 of these 44 genes are testis- or accessory-gland-specific in D. melanogaster, a significantly greater fraction than is found in the ancestral set (FET, P = 1.25 × 10-4). This is consistent with previous observations that novel genes are often testis-specific in Drosophila69,70,71,72,73 and expression studies on seven of the species show that species-restricted genes are more likely to exhibit male-biased expression29. Further, these genes are significantly more tissue-specific in expression (as measured by τ; ref. 74) (MWU, P = 9.6 × 10-6), and this pattern is not solely driven by genes with testis-specific expression patterns.

Protein-coding gene evolution

Positive selection and selective constraints in Drosophila genomes

To study the molecular evolution of protein-coding genes, we estimated rates of synonymous and non-synonymous substitution in 8,510 single-copy orthologues within the six melanogaster group species using PAML75 (Supplementary Information section 11.1); synonymous site saturation prevents analysis of more divergent comparisons. We investigate only single-copy orthologues because when paralogues are included, alignments become increasingly problematic. Rates of amino acid divergence for single-copy orthologues in all 12 species were also calculated; these results are largely consistent with the analysis of non-synonymous divergence in the melanogaster group, and are not discussed further.

To understand global patterns of divergence and constraint across functional classes of genes, we examined the distributions of ω ( = dN/dS, the ratio of non-synonymous to synonymous divergence) across Gene Ontology categories (GO)76, excluding GO annotations based solely on electronic support (Supplementary Information section 11.2). Most functional categories of genes are strongly constrained, with median estimates of ω much less than one. In general, functionally similar genes are similarly constrained: 31.8% of GO categories have significantly lower variance in ω than expected (q-value true-positive test77). Only 11% of GO categories had statistically significantly elevated ω (relative to the median of all genes with GO annotations) at a 5% false-discovery rate (FDR), suggesting either positive selection or a reduction in selective constraint. The GO categories with elevated ω include the biological process terms ‘defence response’, ‘proteolysis’, ‘DNA metabolic process’ and ‘response to biotic stimulus’; the molecular function terms ‘transcription factor activity’, ‘peptidase activity’, ‘receptor binding’, ‘odorant binding’, ‘DNA binding’, ‘receptor activity’ and ‘G-protein-coupled receptor activity’; and the cellular location term ‘extracellular’ (Fig. 4, and Supplementary Table 12). Similar results are obtained when dN is compared across GO categories, suggesting that in most cases differences in ω among GO categories is driven by amino acid rather than synonymous site substitutions. The two exceptions are the molecular function terms ‘transcription factor activity’ and ‘DNA binding activity’, for which we observe significantly decelerated dS (FDR = 7.2 × 10-4 for both; Supplementary Information section 11.2) and no significant differences in dN.

Figure 4: Patterns of constraint and positive selection among GO terms.
figure 4

Distribution of average ω per gene and the negative log10 of the probability of positive selection (Supplementary Information section 11.2) for genes annotated with: a, biological process GO terms; b, cellular component GO terms; and c, molecular function GO terms. Only GO terms with 200 or more genes annotated are plotted. See Supplementary Table 12 for median values and significance. Note that most genes evolve under evolutionary constraint at most of their sites, leading to low values of ω; even genes that experience positive selection do not typically have an average ω across all codons that exceeds one.

To distinguish possible positive selection from relaxed constraint, we tested explicitly for genes that have a subset of codons with signatures of positive selection, using codon-based likelihood models of molecular evolution, implemented in PAML78,79 (Supplementary Information section 11.1). Although this test is typically regarded as a conservative test for positive selection, it may be confounded by selection at synonymous sites. However, selection at synonymous sites (that is, codon bias, see below) is quite weak. Moreover, variability in ω presented here tends to reflect variability in dN. We therefore believe that it is appropriate to treat synonymous sites as nearly neutral and sites with ω > 1 as consistent with positive selection. Despite a number of functional categories with evidence for elevated ω, ‘helicase activity’ is the only functional category significantly more likely to be positively selected (permutation test, P = 2 × 10-4, FDR = 0.007; Supplementary Table 12); the biological significance of this finding merits further investigation. Furthermore, within each GO class, there is greater dispersion among genes in their probability of positive selection than in their estimate of ω (MWU one-tailed, P = 0.011; Supplementary Information section 11.1), suggesting that although functionally similar genes share patterns of constraint, they do not necessarily show similar patterns of positive selection (Fig. 4).

Interestingly, protein-coding genes with no annotated (‘unknown’) function in the GO database seem to be less constrained (permutation test, P < 1 × 10-4, FDR = 0.006)80 and to have on average lower P-values for the test of positive selection than genes with annotated functions (permutation test, P = 0.001, FDR = 0.058). It is unlikely that this observation results entirely from an over-representation of mis-annotated or non-protein-coding genes in the ‘unknown’ functional class, because this finding is robust to the removal of all D. melanogaster genes predicted to be non-protein-coding in ref. 8. The bias in the way biological function is ascribed to genes (to laboratory-induced, easily scorable functions) leaves open the possibility that unannotated biological functions may have an important role in evolution. Indeed, genes with characterized mutant alleles in FlyBase evolve significantly more slowly than other genes (median ωwith alleles = 0.0525 and ωwithout alleles = 0.0701; MWU, P < 1 × 10-16).

Previous work has suggested that a substantial fraction of non-synonymous substitutions in Drosophila were fixed through positive selection81,82,83,84,85. We estimate that 33.1% of single-copy orthologues in the melanogaster group have experienced positive selection on at least a subset of codons (q-value true-positive tests77) (Supplementary Information section 11.1). This may be an underestimate, because we have only examined single-copy orthologues, owing to difficulties in producing accurate alignments of paralogues by automated methods. On the basis of the 878 genes inferred to have experienced positive selection with high confidence (FDR < 10%), we estimated that an average of 2% of codons in positively selected genes have ω > 1. Thus, several lines of evidence, based on different methodologies, suggest that patterns of amino acid fixation in Drosophila genomes have been shaped extensively by positive selection.

The presence of functional domains within a protein may lead to heterogeneity in patterns of constraint and adaptation along its length. Among genes inferred to be evolving by positive selection at a 10% FDR, 63.7% (q-value true-positive tests77) show evidence for spatial clustering of positively selected codons (Supplementary Information section 11.2). Spatial heterogeneity in constraint is further supported by contrasting ω for codons inside versus outside defined InterPro domains (genes lacking InterPro domains are treated as ‘outside’ a defined InterPro domain). Codons within InterPro domains were significantly more conserved than codons outside InterPro domains (median ω: 0.062 InterPro domains, 0.084 outside InterPro domains; MWU, P < 2.2 × 10-16; Supplementary Information section 11.2). Similarly, there were significantly more positively selected codons outside of InterPro domains than inside domains (FET P < 2.2 × 10-16), suggesting that in addition to being more constrained, codons in protein domains are less likely to be targets of positive selection (Supplementary Fig. 6).

Factors affecting the rate of protein evolution in Drosophila

The sequenced genomes of the melanogaster group provide unprecedented statistical power to identify factors affecting rates of protein evolution. Previous analyses have suggested that although the level of gene expression consistently seems to be a major determinant of variation in rates of evolution among proteins86,87, other factors probably play a significant, if perhaps minor, part88,89,90,91. In Drosophila, although highly expressed genes do evolve more slowly, breadth of expression across tissues, gene essentiality and intron number all also independently correlate with rates of protein evolution, suggesting that the additional complexities of multicellular organisms are important factors in modulating rates of protein evolution78. The presence of repetitive amino acid sequences has a role as well: non-repeat regions in proteins containing repeats evolve faster and show more evidence for positive selection than genes lacking repeats92.

These data also provide a unique opportunity to examine the impact of chromosomal location on evolutionary rates. Population genetic theory predicts that for new recessive mutations, both purifying and positive selection will be more efficient on the X chromosome given its hemizygosity in males93. In contrast, the lack of recombination on the small, mainly heterochromatic dot chromosome94,95 is expected to reduce the efficacy of selection96. Because codon bias, or the unequal usage of synonymous codons in protein-coding sequences, reflects weak but pervasive selection, it is a sensitive metric for evaluating the efficacy of purifying selection. Consistent with expectation, in all 12 species, we find significantly elevated levels of codon bias on the X chromosome and significantly reduced levels of codon bias on the dot chromosome97. Furthermore, X-chromosome-linked genes are marginally over-represented within the set of positively selected genes in the melanogaster group (FET, P = 0.055), which is consistent with increased rates of adaptive substitution on this chromosome. This analysis suggests that chromosomal context also serves to modulate rates of molecular evolution in protein-coding genes.

To examine further the impact of genomic location on protein evolution, we examined the subset of genes that have moved within or between chromosome arms32,98. Genes inferred to have moved between Muller elements have a significantly higher rate of protein evolution than genes inferred to have moved within a Muller element (MWU, P = 1.32 × 10-14) and genes that have maintained their genomic position (MWU, P = 0.008) (Supplementary Fig. 7). Interestingly, genes that move within Muller elements have a significantly lower rate of protein evolution than those for which genomic locations have been maintained (MWU, P = 3.85 × 10-14). It remains unclear whether these differences reflect underlying biases in the types of genes that move inter- versus intra-chromosomally, or whether they are due to in situ patterns of evolution in novel genomic contexts.

Codon bias

Codon bias is thought to enhance the efficiency and/or accuracy of translation99,100,101 and seems to be maintained by mutation–selection–drift balance101,102,103,104. Across the 12 Drosophila genomes, there is more codon bias in the Sophophora subgenus than in the Drosophila subgenus, and a previously noted105,106,107,108,109 striking reduction in codon bias in D. willistoni110,111 (Fig. 5). However, with only minor exceptions, codon preferences for each amino acid seem to be conserved across 11 of the 12 species. The striking exception is D. willistoni, in which codon usage for 6 of 18 redundant amino acids has diverged (Fig. 5). Mutation alone is not sufficient to explain codon-usage bias in D. willistoni, which is suggestive of a lineage-specific shift in codon preferences111,112. We found evidence for a lineage-specific genomic reduction in codon bias in D. melanogaster (Fig. 5), as has been suggested previously113,114,115,116,117,118,119. In addition, maximum-likelihood estimation of the strength of selection on synonymous sites in 8,510 melanogaster group single-copy orthologues revealed a marked reduction in the number of genes under selection for increased codon bias in D. melanogaster relative to its sister species D. sechellia120.

Figure 5: Deviations in codon bias from D. melanogaster in 11 Drosophila species.
figure 5

The upper panel depicts differences in ENC (effective number of codons) between D. melanogaster and the 11 non-melanogaster species, calculated on a gene-by-gene basis. Note that increasing levels of ENC indicates a decrease in codon bias. The Sophophora subgenus in general has higher levels of codon bias than the Drosophila subgenus with the exception of D. willistoni, which shows a dramatic reduction in codon bias. The lower panel shows the 7 codons for which preference changes across the 12 Drosophila species. A dot indicates identical codon preference to D. melanogaster; otherwise the preferred codon is indicated.

Evolution of genes associated with ecology and reproduction

Given the ecological and environmental diversity encompassed by the 12 Drosophila species, we examined the evolution of genes and gene families associated with ecology and reproduction. Specifically, we selected genes with roles in chemoreception, detoxification/metabolism, immunity/defence, and sex/reproduction for more detailed study.

Chemoreception

Drosophila species have complex olfactory and gustatory systems used to identify food sources, hazards and mates, which depend on odorant-binding proteins, and olfactory/odorant and gustatory receptors (Ors and Grs). The D. melanogaster genome has approximately 60 Ors, 60 Grs and 50 odorant-binding protein genes. Despite overall conservation of gene number across the 12 species and widespread evidence for purifying selection within the melanogaster group, there is evidence that a subset of Or and Gr genes experiences positive selection121,122,123. Furthermore, clear lineage-specific differences are detectable between generalist and specialist species within the melanogaster subgroup. First, the two independently evolved specialists (D. sechellia and D. erecta) are losing Gr genes approximately five times more rapidly than the generalist species121,124. We believe this result is robust to sequence quality, because all pseudogenes and deletions were verified by direct re-sequencing and synteny-based orthologue searches, respectively. Generalists are expected to encounter the most diverse set of tastants and seem to have maintained the greatest diversity of gustatory receptors. Second, Or and Gr genes that remain intact in D. sechellia and D. erecta evolve significantly more rapidly along these two lineages (ω = 0.1556 for Ors and 0.1874 for Grs) than along the generalist lineages (ω = 0.1049 for Ors and 0.1658 for Grs; paired Wilcoxon, P = 0.0003 and 0.003, respectively124). There is some evidence that odorant-binding protein genes also evolve significantly faster in specialists compared to generalists122. This elevated ω reflects a trend observed throughout the genomes of the two specialists and is likely to result, at least in part, from demographic phenomena. However, the difference between specialist and generalist ω for Or/Gr genes (0.0292) is significantly greater than the difference for genes across the genome (0.0091; MWU, P = 0.0052)121, suggesting a change in selective regime. Moreover, the observation that elevated ω as well as accelerated gene loss disproportionately affect groups of Or and Gr genes that respond to specific chemical ligands and/or are expressed during specific life stages suggests that rapid evolution at Or/Gr loci in specialists is related to the ecological shifts these species have sustained121.

Detoxification/metabolism

The larval food sources for many Drosophila species contain a cocktail of toxic compounds, and consequently Drosophila genomes encode a wide variety of detoxification proteins. These include members of the cytochrome P450 (P450), carboxyl/choline-esterase (CCE) and glutathione S-transferase (GST) multigene families, all of which also have critical roles in resistance to insecticides125,126,127. Among the P450s, the five enzymes associated with insecticide resistance are highly dynamic across the phylogeny, with 24 duplication events and 4 loss events since the last common ancestor of the genus, which is in striking contrast to genes with known developmental roles, eight of which are present as a single copy in all 12 species (C. Robin, personal communication). As with chemoreceptors, specialists seem to lose detoxification genes at a faster rate than generalists. For instance, D. sechellia has lost the most P450 genes; these 14 losses comprise almost one-third of all P450 loss events (Supplementary Table 13) (C. Robin, personal communication). Positive selection has been implicated in detoxification-gene evolution as well, because a search for positive selection among GSTs identified the parallel evolution of a radical glycine to lysine amino acid change in GSTD1, an enzyme known to degrade DDT128. Finally, although metabolic enzymes in general are highly constrained (median ω = 0.045 for enzymes, 0.066 for non-enzymes; MWU, P = 5.7 × 10-24), enzymes involved in xenobiotic metabolism evolve significantly faster than other enzymes (median ω = 0.05 for the xenobiotic group versus ω = 0.045 overall, two-tailed permutation test, P = 0.0110; A. J. Greenberg, personal communication).

Metazoans deal with excess selenium in the diet by sequestration in selenoproteins, which incorporate the rare amino acid selenocysteine (Sec) at sites specified by the TGA codon. The recoding of the normally terminating signal TGA as a Sec codon is mediated by the selenocystein insertion sequence (SECIS), a secondary structure in the 3′ UTR of selenoprotein messenger RNAs. All animals examined so far have selenoproteins; three have been identified in D. melanogaster (SELG, SELM and SPS2129,130). Interestingly, although the three known melanogaster selenoproteins are all present in the genomes of the other Drosophila species, in D. willistoni the TGA Sec codons have been substituted by cysteine codons (TGT/TGC). Consistent with this finding, analysis of the seven genes implicated to date in selenoprotein synthesis including the Sec-specific tRNA suggests that most of these genes are absent in D. willistoni (R. Guigo, personal communication). D. willistoni thus seems to be the first animal known to lack selenoproteins. If correct, this observation is all the more remarkable given the ubiquity of selenoproteins and the selenoprotein biosynthesis machinery in metazoans, the toxicity of excess selenium, and the protection from oxidative stress mediated by selenoproteins. However, it remains possible that this species encodes selenoproteins in a different way, and this represents an exciting avenue of future research.

Immunity/defence

Drosophila, like all insects, possesses an innate immune system with many components analogous to the innate immune pathways of mammals, although it lacks an antibody-mediated adaptive immune system131. Immune system genes often evolve rapidly and adaptively, driven by selection pressures from pathogens and parasites132,133,134. The genus Drosophila is no exception: immune system genes evolve more rapidly than non-immune genes, showing both high total divergence rates and specific signs of positive selection135. In particular, 29% of receptor genes involved in phagocytosis seem to evolve under positive selection, suggesting that molecular co-evolution between Drosophila pattern recognition receptors and pathogen antigens is driving adaptation in the immune system135. Somewhat surprisingly, genes encoding effector proteins such as antimicrobial peptides are far less likely to exhibit adaptive sequence evolution. Only 5% of effector genes (and no antimicrobial peptides) show evidence of adaptive evolution, compared to 10% of genes genome-wide. Instead, effector genes seem to evolve by rapid duplication and deletion. Whereas 49% of genes genome-wide, 63% of genes involved in pathogen recognition and 81% of genes implicated in immune-related signal transduction can be found as single-copy orthologues in all 12 species, only 40% of effector genes exist as single-copy orthologues across the genus (χ2 = 41.13, P = 2.53 × 10-8), suggesting rapid radiation of effector protein classes along particular lineages135. Thus, much of the Drosophila immune system seems to evolve rapidly, although the mode of evolution varies across immune-gene functional classes.

Sex/reproduction

Genes encoding sex- and reproduction-related proteins are subject to a wide array of selective forces, including sexual conflict, sperm competition and cryptic female choice, and to the extent that these selective forces are of evolutionary consequence, this should lead to rapid evolution in these genes136 (for an overview see refs 137, 138). The analysis of 2,505 sex- and reproduction-related genes within the melanogaster group indicated that male sex- and reproduction-related genes evolve more rapidly at the protein level than genes not involved in sex or reproduction or than female sex- and reproduction-related genes (Supplementary Fig. 8). Positive selection seems to be at least partially responsible for these patterns, because genes involved in spermatogenesis have significantly stronger evidence for positive selection than do non-spermatogenesis genes (permutation test, P = 0.0053). Similarly, genes that encode components of seminal fluid have significantly stronger evidence for positive selection than ‘non-sex’ genes139. Moreover, protein-coding genes involved in male reproduction, especially seminal fluid and testis genes, are particularly likely to be lost or gained across Drosophila species29,139.

Evolutionary forces in the mitochondrial genome

Functional elements in mtDNA are strongly conserved, as expected: tRNAs are relatively more conserved than the mtDNA overall (average pairwise nucleotide distance = 0.055 substitutions per site for tRNAs versus 0.125 substitutions per site overall). We observe a deficit of substitutions occurring in the stem regions of the stem-loop structure in tRNAs, consistent with strong selective pressure to maintain RNA secondary structure, and there is a strong signature of purifying selection in protein-coding genes13. However, despite their shared role in aerobic respiration, there is marked heterogeneity in the rates of amino acid divergence between the oxidative phosphorylation enzyme complexes across the 12 species (NADH dehydrogenase, 0.059 > ATPase, 0.042 > CytB, 0.037 > cytochrome oxidase, 0.020; mean pairwise dN ), which contrasts with the relative homogeneity in synonymous substitution rates. A model with distinct substitution rates for each enzyme complex rather than a single rate provides a significantly better fit to the data (P < 0.0001), suggesting complex-specific selective effects of mitochondrial mutations13.

Non-coding sequence evolution

ncRNA sequence evolution

The availability of complete sequence from 12 Drosophila genomes, combined with the tractability of RNA structure predictions, offers the exciting opportunity to connect patterns of sequence evolution directly with structural and functional constraints at the molecular level. We tested models of RNA evolution focusing on specific ncRNA gene classes in addition to inferring patterns of sequence evolution using more general datasets that are based on predicted intronic RNA structures.

The exquisite simplicity of miRNAs and their shared stem-loop structure makes these ncRNAs particularly amenable to evolutionary analysis. Most miRNAs are highly conserved within the Drosophila genus: for the 71 previously described miRNA genes inferred to be present in the common ancestor of these 12 species, mature miRNA sequences are nearly invariant. However, we do find a small number of substitutions and a single deletion in mature miRNA sequences (Supplementary Table 14), which may have functional consequences for miRNA–target interactions and may ultimately help identify targets through sequence covariation. Pre-miRNA sequences are also highly conserved, evolving at about 10% of the rate of synonymous sites.

To link patterns of evolution with structural constraints, we inferred ancestral pre-miRNA sequences and deduced secondary structures at each ancestral node on the phylogeny (Supplementary Information section 12.1). Although conserved miRNA genes show little structural change (little change in free energy), the five melanogaster group-specific miRNA genes (miR-303 and the mir-310/311/312/313 cluster) have undergone numerous changes across the entire pre-miRNA sequence, including the ordinarily invariant mature miRNA. Patterns of polymorphism and divergence in these lineage-specific miRNA genes, including a high frequency of derived mutations, are suggestive of positive selection140. Although lineage-specific miRNAs may evolve under less constraint because they have fewer target transcripts in the genome, it is also possible that recent integration into regulatory networks causes accelerated rates of miRNA evolution.

We further investigated patterns of sequence evolution for the subset of 38 conserved pre-miRNAs with mature miRNA sequences at their 3′ end by calculating evolutionary rates in distinct site classes (Fig. 6, and Supplementary Information section 12.2). Outside the mature miRNA and its complementary sequence, loops had the highest rate of evolution, followed by unpaired sites, with paired sites having the lowest rate of evolution. Inside the mature miRNA, unpaired sites evolve more slowly than paired sites, whereas the opposite is true for the sequence complementary to the mature miRNA. Surprisingly, a large fraction of unpaired bulges or internal loops in the mature miRNA seem to be conserved—a pattern which may have implications for models of miRNA biogenesis and the degree of mismatch allowed in miRNA–target prediction methods. Overall these results support the qualitative model proposed in ref. 141 for the canonical progression of miRNA evolution, and show that functional constraints on the miRNA itself supersede structural constraints imposed by maintenance of the hairpin-loop.

Figure 6: Substitution rate of site classes within miRNAs.
figure 6

Bootstrap distributions of miRNA substitution rates. Structural alignments of miRNA precursor hairpins were partitioned into six site-classes (inset): (1) hairpin loops; unpaired sites (2) outside, (3) in the complementary region of, and (4) inside the miRNA; and base pairs (5) adjacent to and (6) involving the miRNA. Whiskers show approximate 95% confidence intervals for median differences, boxes show interquartile range.

To assess constraint on stem regions of RNA structures more generally, we compared substitution rates in stems (S) to those in nominally unconstrained loop regions (L) in a wide variety of ncRNAs (Supplementary Information section 12.3). We estimated substitution rates using a maximum likelihood framework, and compared the observed L/S ratio with the average L/S ratio estimated from published secondary structures in RFAM, which we normalized to 1.0. L/S ratios for Drosophila ncRNA families range from a highly constrained 2.57 for the nuclear RNase P family to 0.56 for the 5S ribosomal RNA (Supplementary Table 15).

Finally, we predicted a set of conserved intronic RNA structures and analysed patterns of compensatory nucleotide substitution in D. melanogaster, D. yakuba, D. ananassae, D. pseudoobscura, D. virilis and D. mojavensis (Supplementary Information section 13). Signatures of compensatory evolution in RNA helices are detected as covarying nucleotide sites or ‘covariations’ (that is, two Watson–Crick bases that interact in species A replaced by a different Watson–Crick pair in species B). The number of covariations (per base pair of a helix) depends on the physical distance between the interacting nucleotides (Supplementary Fig. 9), as has been observed for the RNA helices in the Drosophila bicoid 3′ UTR region142. Short-range pairings exhibit a higher average number of covariations with a larger variance among helices than longer-range pairings. The decrease in rate of covariation with increasing distance may be explained by physical properties of a helix, which may impose selective constraints on the evolution of covarying nucleotides within a helix. Alternatively, if individual mutations at each locus are deleterious but compensated by mutations at a second locus, given sufficiently strong selection against the first deleterious mutation these epistatic fitness interactions could generate the observed distance effect143.

Evolution of cis-regulatory DNAs

Comparative analyses of cis-regulatory sequences may provide insights into the evolutionary forces acting on regulatory components of genes, shed light on the constraints of the cis-regulatory code and aid in annotation of new regulatory sequences. Here we rely on two recently compiled databases, and present results comparing cis-regulatory modules144 and transcription factor binding sites (derived from DNase I footprints)145 between D. melanogaster and D. simulans (Supplementary Information section 8). We estimated mean selective constraint (C, the fraction of mutations removed by natural selection) relative to the ‘fastest evolving intron’ sites at the 5′ end of short introns, which represent putatively unconstrained neutral standards (Supplementary Information section 8.2)146. Note that this approach ignores the contribution of positively selected sites, potentially underestimating the fraction of functionally relevant sites147.

Consistent with previous findings, Drosophila cis-regulatory sequences are highly constrained148,149. Mean constraint within cis-regulatory modules is 0.643 (95% bootstrap confidence interval = 0.621–0.662) and within footprints is 0.692 (0.655–0.723), both of which are significantly higher than mean constraint in non-coding DNA overall (0.555 (0.546–0.563)) and significantly lower than constraint at non-degenerate coding sites (0.862 (0.856–0.868)) and ncRNA genes (0.864 (0.846–0.880)) (Supplementary Fig. 10). The high level of constraint in cis-regulatory sequences also extends into flanking sequences, only declining to constraint levels typical of non-coding DNA 40 bp away. This is consistent with previous findings that transcription factor binding sites tend to be found in larger blocks of constraint that cluster to form cis-regulatory modules150. To understand selective constraints on nucleotides within cis-regulatory sequences that have direct contact with transcription factors, we estimated the selective constraint for the best match to position weight matrices within each footprint151; core motifs in transcription-factor-binding sites have a mean constraint of 0.773 (0.729–0.814), significantly greater than the mean for the footprints as a whole, and approaching the level of constraint found at non-degenerate coding sites and in ncRNA genes (Supplementary Fig. 10).

We next examined the variation in selective constraint across cis-regulatory sequences. Surprisingly, we find no evidence that selective constraint is correlated with predicted transcription-factor-binding strength (estimated as the position weight matrix score P-value) (Spearman’s r = 0.0681, P = 0.0609). We observe significant variation in constraint both among target genes (Kruskal–Wallis tests, footprints, P < 0.0001; and position weight matrix matches within footprints, P = 0.0023) and among chromosomes (cis-regulatory modules, P = 0.0186; footprints, P = 0.0388; and position weight matrix matches within footprints, P = 0.0108; Supplementary Table 16).

Discussion and conclusion

Each new genome sequence affords novel opportunities for comparative genomic inference. What makes the analysis of these 12 Drosophila genomes special is the ability to place every one of these genomic comparisons on a phylogeny with a taxon separation that is ideal for asking a wealth of questions about evolutionary patterns and processes. It is without question that this phylogenomic approach places additional burdens on bioinformatics efforts, multiplying the amount of data many-fold, requiring extra care in generating multi-species alignments, and accommodating the reality that not all genome sequences have the same degree of sequencing or assembly accuracy. These difficulties notwithstanding, phylogenomics has extraordinary advantages not only for the analyses that are possible, but also for the ability to produce high-quality assemblies and accurate annotations of functional features in a genome by using closely related genomes as guides. The use of multi-species orthology provides especially convincing evidence in support of particular gene models, not only for protein-coding genes, but also for miRNA and other ncRNA genes.

Many attributes of the genomes of Drosophila are remarkably conserved across species. Overall genome size, number of genes, distribution of transposable element classes, and patterns of codon usage are all very similar across these 12 genomes, although D. willistoni is an exceptional outlier by several criteria, including its unusually skewed codon usage, increased transposable element content and potential lack of selenoproteins. At a finer scale, the number of structural changes and rearrangements is much larger; for example, there are several different rearrangements of genes in the Hox cluster found in these Drosophila species.

The vast majority of multigene families are found in all 12 genomes, although gene family size seems to be highly dynamic: almost half of all gene families change in size on at least one lineage, and a noticeable fraction shows rapid and lineage-specific expansions and contractions. Particularly notable are cases consistent with adaptive hypotheses, such as the loss of Gr genes in ecological specialists and the lineage-specific expansions of antimicrobial peptides and other immune effectors. All species were found to have novel genes not seen in other species. Although lineage-specific genes are challenging to verify computationally, we can confirm at least 44 protein-coding genes unique to the melanogaster group, and these proteins have very different properties from ancestral proteins. Similarly, although the relative abundance of transposable element subclasses across these genomes does not differ dramatically, total genomic transposable element content varies substantially among species, and several instances of lineage-specific transposable elements were discovered.

There is considerable variation among protein-coding genes in rates of evolution and patterns of positive selection. Functionally similar proteins tend to evolve at similar rates, although variation in genomic features such as gene expression level, as well as chromosomal location, are also associated with variation in evolutionary rate among proteins. Whereas broad functional classes do not seem to share patterns of positive selection, and although very few GO categories show excesses of positive selection, a number of genes involved in interactions with the environment and in sex and reproduction do show signatures of adaptive evolution. It thus seems likely that adaptation to changing environments, as well as sexual selection, shape the evolution of protein-coding genes.

Annotation of ncRNA genes across all 12 species allows comprehensive analysis of the evolutionary divergence of these genes. MicroRNA genes in particular are more conserved than protein-coding genes with respect to their primary DNA sequence, and the substitutions that do occur often have compensatory changes such that the average estimated free energy of the folding structures remains remarkably constant across the phylogeny. Surprisingly, mismatches in miRNAs seem to be highly conserved, which may impact models of miRNA biogenesis and target recognition. Lineage-restricted miRNAs, however, have considerably elevated rates of change, suggesting either reduced constraint due to novel miRNAs having fewer targets, or adaptive evolution of evolutionarily young miRNAs.

Virtually any question about the function of genome features in Drosophila is now empowered by being embedded in the context of this 12 species phylogeny, allowing an analysis of the ways by which evolution has tuned myriad biological processes across the hundreds of millions of years spanned in total by this phylogeny. The analyses presented herein have generated more questions than they have answered, and these results represent a small fraction of that which is possible. Because much of this rich and extraordinary comparative genomic dataset remains to be explored, we believe that these 12 Drosophila genome sequences will serve as a powerful tool for gleaning further insight into genetic, developmental, regulatory and evolutionary processes.

Methods

The full methods for this paper are described in Supplementary Information. Here, we describe the datasets generated by this project and their availability.

Genomic sequence

Scaffolds and assemblies for all genomic sequence generated by this project are available from GenBank (Supplementary Tables 4 and 5), and FlyBase (ftp://ftp.flybase.net/12_species_analysis/). Genome browsers are available from UCSC (http://genome.ucsc.edu/cgi-bin/hgGateway?hgsid =98180333&clade = insect&org = 0&db = 0) and Flybase (http://flybase.org/cgi-bin/gbrowse/dmel/). BLAST search of these genomes is available at FlyBase (http://flybase.org/blast).

Predicted gene models

Consensus gene predictions for the 11 non-melanogaster species, produced by combining several different GLEAN runs that weight homology evidence more or less strongly, are available from FlyBase as GFF files for each species (ftp://ftp.flybase.net/12_species_analysis/). These gene models can also be accessed from the Genome Browser in FlyBase (Gbrowse; http://flybase.org/cgi-bin/gbrowse/dmel/). Predictions of non-protein-coding genes are also available in GFF format for each species, from FlyBase (ftp://ftp.flybase.net/12_species_analysis/).

Homology

Multiway homology assignments are available from FlyBase (ftp://ftp.flybase.net/12_species_analysis/), and also in the Genome Browser (Gbrowse).

Alignments

All alignment sets produced are available in FASTA format from FlyBase (ftp://ftp.flybase.net/12_species_analysis/).

PAML parameters

Output from PAML models for the alignments of single copy orthologues in the melanogaster group, including the q-value for the test for positive selection, are available from FlyBase (ftp://ftp.flybase.net/12_species_analysis/).