Ultraconserved elements in insect genomes: a highly conserved intronic sequence implicated in the control of homothorax mRNA splicing

The animal in the genome: comparative genomics and evolution

Comparisons between completely sequenced metazoan genomes have generally emphasized how similar their encoded protein content is, even when the comparison is between phyla. Given the manifest differences between phyla and, in particular, intuitive notions that some animals are more complex than others, this creates something of a paradox. Simplistic explanations have included arguments such as increased numbers of genes; greater numbers of protein products produced through alternative splicing; increased numbers of regulatory non-coding RNAs and increased complexity of the cis-regulatory code. An obvious value of complete genome sequences lies in their ability to provide us with inventories of such components. I examine progress being made in linking genome content to the pattern of animal evolution, and argue that the gap between genomic and phenotypic complexity can only be understood through the totality of interacting components.

NcDNAlign: plausible multiple alignments of non-protein-coding genomic sequences

Genome-wide multiple sequence alignments (MSAs) are a necessary prerequisite for an increasingly diverse collection of comparative genomic approaches. Here we present a versatile method that generates high-quality MSAs for non-protein-coding sequences. The NcDNAlign pipeline combines pairwise BLAST alignments to create initial MSAs, which are then locally improved and trimmed. The program is optimized for speed and hence is particulary well-suited to pilot studies. We demonstrate the practical use of NcDNAlign in three case studies: the search for ncRNAs in gammaproteobacteria and the analysis of conserved noncoding DNA in nematodes and teleost fish, in the latter case focusing on the fate of duplicated ultra-conserved regions. Compared to the currently widely used genome-wide alignment program TBA, our program results in a 20- to 30-fold reduction of CPU time necessary to generate gammaproteobacterial alignments. A showcase application of bacterial ncRNA prediction based on alignments of both algorithms results in similar sensitivity, false discovery rates, and up to 100 putatively novel ncRNA structures. Similar findings hold for our application of NcDNAlign to the identification of ultra-conserved regions in nematodes and teleosts. Both approaches yield conserved sequences of unknown function, result in novel evolutionary insights into conservation patterns among these genomes, and manifest the benefits of an efficient and reliable genome-wide alignment package. The software is available under the GNU Public License at http://www.bioinf.uni-leipzig.de/Software/NcDNAlign/.

Large-scale analysis of transcriptional cis-regulatory modules reveals both common features and distinct subclasses

Analysis of 280 experimentally-verified cis-regulatory modules from Drosophila reveal features both common to all and unique to distinct subclasses of modules.

Background

Transcriptional cis-regulatory modules (for example, enhancers) play a critical role in regulating gene expression. While many individual regulatory elements have been characterized, they have never been analyzed as a class.

Results

We have performed the first such large-scale study of cis-regulatory modules in order to determine whether they have common properties that might aid in their identification and contribute to our understanding of the mechanisms by which they function. A total of 280 individual, experimentally verified cis-regulatory modules from Drosophila were analyzed for a range of sequence-level and functional properties. We report here that regulatory modules do indeed share common properties, among them an elevated GC content, an increased level of interspecific sequence conservation, and a tendency to be transcribed into RNA. However, we find that dense clustering of transcription factor binding sites, especially homotypic clustering, which is commonly believed to be a general characteristic of regulatory modules, is rather a feature that belongs chiefly to a specific subclass. This has important implications for current computational approaches, many of which are biased toward this subset. We explore two new strategies to assess binding site clustering and gauge their performances with respect to their ability to detect all 280 modules and various functionally coherent subsets.

Conclusion

Our findings demonstrate that cis-regulatory modules share common features that help to define them as a class and that may lead to new insights into mechanisms of gene regulation. However, these properties alone may not be sufficient to reliably distinguish regulatory from non-regulatory sequences. We also demonstrate that there are distinct subclasses of cis-regulatory modules that are more amenable to in silico detection than others and that these differences must be taken into account when attempting genome-wide regulatory element discovery.

Ancora: a web resource for exploring highly conserved noncoding elements and their association with developmental regulatory genes

Ancora is a web resource that provides data and tools for exploring genomic organization of highly conserved noncoding elements for multiple genomes.

Metazoan genomes contain arrays of highly conserved noncoding elements (HCNEs) that span developmental regulatory genes and define regulatory domains. We describe Ancora , a web resource that provides data and tools for exploring genomic organization of HCNEs for multiple genomes. Ancora includes a genome browser that shows HCNE locations and features novel HCNE density plots as a powerful tool to discover developmental regulatory genes and distinguish their regulatory elements and domains.

Multiple organism algorithm for finding ultraconserved elements

Background

Ultraconserved elements are nucleotide or protein sequences with 100% identity (no mismatches, insertions, or deletions) in the same organism or between two or more organisms. Studies indicate that these conserved regions are associated with micro RNAs, mRNA processing, development and transcription regulation. The identification and characterization of these elements among genomes is necessary for the further understanding of their functionality.

Results

We describe an algorithm and provide freely available software which can find all of the ultraconserved sequences between genomes of multiple organisms. Our algorithm takes a combinatorial approach that finds all sequences without requiring the genomes to be aligned. The algorithm is significantly faster than BLAST and is designed to handle very large genomes efficiently. We ran our algorithm on several large comparative analyses to evaluate its effectiveness; one compared 17 vertebrate genomes where we find 123 ultraconserved elements longer than 40 bps shared by all of the organisms, and another compared the human body louse, Pediculus humanus humanus, against itself and select insects to find thousands of non-coding, potentially functional sequences.

Conclusion

Whole genome comparative analysis for multiple organisms is both feasible and desirable in our search for biological knowledge. We argue that bioinformatic programs should be forward thinking by assuming analysis on multiple (and possibly large) genomes in the design and implementation of algorithms. Our algorithm shows how a compromise design with a trade-off of disk space versus memory space allows for efficient computation while only requiring modest computer resources, and at the same time providing benefits not available with other software.

REDfly 2.0: an integrated database of cis-regulatory modules and transcription factor binding sites in Drosophila

The identification and study of the cis-regulatory elements that control gene expression are important areas of biological research, but few resources exist to facilitate large-scale bioinformatics studies of cis-regulation in metazoan species. Drosophila melanogaster, with its well-annotated genome, exceptional resources for comparative genomics and long history of experimental studies of transcriptional regulation, represents the ideal system for regulatory bioinformatics. We have merged two existing Drosophila resources, the REDfly database of cis-regulatory modules and the FlyReg database of transcription factor binding sites (TFBSs), into a single integrated database containing extensive annotation of empirically validated cis-regulatory modules and their constituent binding sites. With the enhanced functionality made possible through this integration of TFBS data into REDfly, together with additional improvements to the REDfly infrastructure, we have constructed a one-stop portal for Drosophila cis-regulatory data that will serve as a powerful resource for both computational and experimental studies of transcriptional regulation. REDfly is freely accessible at http://redfly.ccr.buffalo.edu.

The nuclear OXPHOS genes in insecta: a common evolutionary origin, a common cis-regulatory motif, a common destiny for gene duplicates

Background

When orthologous sequences from species distributed throughout an optimal range of divergence times are available, comparative genomics is a powerful tool to address problems such as the identification of the forces that shape gene structure during evolution, although the functional constraints involved may vary in different genes and lineages.

Results

We identified and annotated in the MitoComp2 dataset the orthologs of 68 nuclear genes controlling oxidative phosphorylation in 11 Drosophilidae species and in five non-Drosophilidae insects, and compared them with each other and with their counterparts in three vertebrates (Fugu rubripes, Danio rerio and Homo sapiens) and in the cnidarian Nematostella vectensis, taking into account conservation of gene structure and regulatory motifs, and preservation of gene paralogs in the genome. Comparative analysis indicates that the ancestral insect OXPHOS genes were intron rich and that extensive intron loss and lineage-specific intron gain occurred during evolution. Comparison with vertebrates and cnidarians also shows that many OXPHOS gene introns predate the cnidarian/Bilateria evolutionary split. The nuclear respiratory gene element (NRG) has played a key role in the evolution of the insect OXPHOS genes; it is constantly conserved in the OXPHOS orthologs of all the insect species examined, while their duplicates either completely lack the element or possess only relics of the motif.

Conclusion

Our observations reinforce the notion that the common ancestor of most animal phyla had intron-rich gene, and suggest that changes in the pattern of expression of the gene facilitate the fixation of duplications in the genome and the development of novel genetic functions.

Genomic regulatory blocks underlie extensive microsynteny conservation in insects

Insect genomes contain larger blocks of conserved gene order (microsynteny) than would be expected under a random breakage model of chromosome evolution. We present evidence that microsynteny has been retained to keep large arrays of highly conserved noncoding elements (HCNEs) intact. These arrays span key developmental regulatory genes, forming genomic regulatory blocks (GRBs). We recently described GRBs in vertebrates, where most HCNEs function as enhancers and HCNE arrays specify complex expression programs of their target genes. Here we present a comparison of five Drosophila genomes showing that HCNE density peaks centrally in large synteny blocks containing multiple genes. Besides developmental regulators that are likely targets of HCNE enhancers, HCNE arrays often span unrelated neighboring genes. We describe differences in core promoters between the target genes and the unrelated genes that offer an explanation for the differences in their responsiveness to enhancers. We show examples of a striking correspondence between boundaries of synteny blocks, HCNE arrays, and Polycomb binding regions, confirming that the synteny blocks correspond to regulatory domains. Although few noncoding elements are highly conserved between Drosophila and the malaria mosquito Anopheles gambiae, we find that A. gambiae regions orthologous to Drosophila GRBs contain an equivalent distribution of noncoding elements highly conserved in the yellow fever mosquito Aëdes aegypti and coincide with regions of ancient microsynteny between Drosophila and mosquitoes. The structural and functional equivalence between insect and vertebrate GRBs marks them as an ancient feature of metazoan genomes and as a key to future studies of development and gene regulation.

Vertebrate conserved non coding DNA regions have a high persistence length and a short persistence time

Background

The comparison of complete genomes has revealed surprisingly large numbers of conserved non-protein-coding (CNC) DNA regions. However, the biological function of CNC remains elusive. CNC differ in two aspects from conserved protein-coding regions. They are not conserved across phylum boundaries, and they do not contain readily detectable sub-domains. Here we characterize the persistence length and time of CNC and conserved protein-coding regions in the vertebrate and insect lineages.

Results

The persistence length is the length of a genome region over which a certain level of sequence identity is consistently maintained. The persistence time is the evolutionary period during which a conserved region evolves under the same selective constraints.

Our main findings are: (i) Insect genomes contain 1.60 times less conserved information than vertebrates; (ii) Vertebrate CNC have a higher persistence length than conserved coding regions or insect CNC; (iii) CNC have shorter persistence times as compared to conserved coding regions in both lineages.

Conclusion

Higher persistence length of vertebrate CNC indicates that the conserved information in vertebrates and insects is organized in functional elements of different lengths. These findings might be related to the higher morphological complexity of vertebrates and give clues about the structure of active CNC elements.

Shorter persistence time might explain the previously puzzling observations of highly conserved CNC within each phylum, and of a lack of conservation between phyla. It suggests that CNC divergence might be a key factor in vertebrate evolution. Further evolutionary studies will help to relate individual CNC to specific developmental processes.

Analysis of invariant sequences in 266 complete genomes

To date, the complete genome sequences of more than 250 organisms have been determined. This information can now be used to determine whether there exist any invariant sequences that are conserved among all organisms, from bacteria to plants, animals, and humans. The existence of invariant sequences would strongly suggest that these sequences have been inherited unchanged from the last common ancestor of all life, and that they have essential functions. We have developed a new software program to identify invariant sequences conserved among the currently sequenced genomes and applied this analysis to the complete genome sequences of 266 organisms. We have identified 3 invariant DNA sequences longer than or equal to 11 bp and 6 invariant amino acid sequences longer than or equal to 6 aa. The longest invariant DNA sequence, AAGTCGTACAAGGT (15 bp), was found in the 16S/18S rRNA gene. Two 8 aa sequences, GHVDHGKT in IF2 and EF-Tu and DTPGHVDF in EF-G, were the longest invariant amino acid sequences detected. These sequences could be essential elements from the genome of the last common ancestor and may have remained unchanged throughout evolution.

RNA structure is a key regulatory element in pathological ATM and CFTR pseudoexon inclusion events

Genomic variations deep in the intronic regions of pre-mRNA molecules are increasingly reported to affect splicing events. However, there is no general explanation why apparently similar variations may have either no effect on splicing or cause significant splicing alterations. In this work we have examined the structural architecture of pseudoexons previously described in ATM and CFTR patients. The ATM case derives from the deletion of a repressor element and is characterized by an aberrant 5′ss selection despite the presence of better alternatives. The CFTR pseudoexon instead derives from the creation of a new 5′ss that is used while a nearby pre-existing donor-like sequence is never selected. Our results indicate that RNA structure is a major splicing regulatory factor in both cases. Furthermore, manipulation of the original RNA structures can lead to pseudoexon inclusion following the exposure of unused 5′ss already present in their wild-type intronic sequences and prevented to be recognized because of their location in RNA stem structures. Our data show that intrinsic structural features of introns must be taken into account to understand the mechanism of pseudoexon activation in genetic diseases. Our observations may help to improve diagnostics prediction programmes and eventual therapeutic targeting.

A balance between two nuclear localization sequences and a nuclear export sequence governs extradenticle subcellular localization

During animal development, transcription factor activities are modulated by several means, including subcellular localization. The Hox cofactor Extradenticle (Exd) has a dynamic subcellular localization, such that Exd is cytoplasmic by default, but is nuclear when complexed with another homeodomain protein, Homothorax (Hth). These observations raise the question of whether dimerization with Hth simply induces Exd's nuclear localization or, alternatively, if Hth is also necessary for Exd activity. To address this question, we analyzed the nuclear transport signals in Exd, including a divergent nuclear export signal (NES) and two nuclear localization signals (NLSs). We show that, although these signals are weak compared to canonical signals, they balance each other in Exd. We also provide evidence that Exd contains an NLS mask that contributes to its cytoplasmic localization. With these signals characterized, we generated forms of Exd that are nuclear localized in the absence of Hth. Surprisingly, although these Exd forms are functional, they do not phenocopy Hth overexpression. These findings suggest that Hth is required for Exd activity, not simply for inducing its nuclear localization.

Parallel evolution of conserved non-coding elements that target a common set of developmental regulatory genes from worms to humans

BACKGROUND

The human genome contains thousands of non-coding sequences that are often more conserved between vertebrate species than protein-coding exons. These highly conserved non-coding elements (CNEs) are associated with genes that coordinate development, and have been proposed to act as transcriptional enhancers. Despite their extreme sequence conservation in vertebrates, sequences homologous to CNEs have not been identified in invertebrates.

RESULTS

Here we report that nematode genomes contain an alternative set of CNEs that share sequence characteristics, but not identity, with their vertebrate counterparts. CNEs thus represent a very unusual class of sequences that are extremely conserved within specific animal lineages yet are highly divergent between lineages. Nematode CNEs are also associated with developmental regulatory genes, and include well-characterized enhancers and transcription factor binding sites, supporting the proposed function of CNEs as cis-regulatory elements. Most remarkably, 40 of 156 human CNE-associated genes with invertebrate orthologs are also associated with CNEs in both worms and flies.

CONCLUSION

A core set of genes that regulate development is associated with CNEs across three animal groups (worms, flies and vertebrates). We propose that these CNEs reflect the parallel evolution of alternative enhancers for a common set of developmental regulatory genes in different animal groups. This 're-wiring' of gene regulatory networks containing key developmental coordinators was probably a driving force during the evolution of animal body plans. CNEs may, therefore, represent the genomic traces of these 'hard-wired' core gene regulatory networks that specify the development of each alternative animal body plan.

Comparative genomic analysis using the UCSC genome browser

Comparative analysis of DNA sequence from multiple species can provide insights into the function and evolutionary processes that shape genomes. The University of California Santa Cruz (UCSC) Genome Bioinformatics group has developed several tools and methodologies in its study of comparative genomics, many of which have been incorporated into the UCSC Genome Browser (http://genome.ucsc.edu), an easy-to-use online tool for browsing genomic data and aligned annotation "tracks" in a single window. The comparative genomics annotations in the browser include pairwise alignments, which aid in the identification of orthologous regions between species, and conservation tracks that show measures of evolutionary conservation among sets of multiply aligned species, highlighting regions of the genome that may be functionally important. A related tool, the UCSC Table Browser, provides a simple interface for querying, analyzing, and downloading the data underlying the Genome Browser annotation tracks. Here, we describe a procedure for examining a genomic region of interest in the Genome Browser, analyzing characteristics of the region, filtering the data, and downloading data sets for further study.

Unraveling transcription regulatory networks by protein-DNA and protein-protein interaction mapping

Metazoan genomes contain thousands of protein-coding and noncoding RNA genes, most of which are differentially expressed, i.e., at different locations or at different times during development, function, or pathology of the organism. Differential gene expression is achieved in part by the action of regulatory transcription factors (TFs) that bind to cis-regulatory elements that are often located in or near their target genes. Each TF likely regulates many targets in the context of intricate transcription regulatory networks. Up to 10% of a genome may encode TFs, but only a handful of these have been studied in detail. Here, I will discuss the different steps involved in the mapping and analysis of transcription regulatory networks, including the identification of network nodes (TFs and their target sequences) and edges (TF-TF dimers and TF-DNA target interactions), integration with other data types, and network properties and emerging principles that provide insights into differential gene expression.

snoSeeker: an advanced computational package for screening of guide and orphan snoRNA genes in the human genome

Small nucleolar RNAs (snoRNAs) represent an abundant group of non-coding RNAs in eukaryotes. They can be divided into guide and orphan snoRNAs according to the presence or absence of antisense sequence to rRNAs or snRNAs. Current snoRNA-searching programs, which are essentially based on sequence complementarity to rRNAs or snRNAs, exist only for the screening of guide snoRNAs. In this study, we have developed an advanced computational package, snoSeeker, which includes CDseeker and ACAseeker programs, for the highly efficient and specific screening of both guide and orphan snoRNA genes in mammalian genomes. By using these programs, we have systematically scanned four human-mammal whole-genome alignment (WGA) sequences and identified 54 novel candidates including 26 orphan candidates as well as 266 known snoRNA genes. Eighteen novel snoRNAs were further experimentally confirmed with four snoRNAs exhibiting a tissue-specific or restricted expression pattern. The results of this study provide the most comprehensive listing of two families of snoRNA genes in the human genome till date.

Scale-invariant structure of strongly conserved sequence in genomic intersections and alignments

A power-law distribution of the length of perfectly conserved sequence from mouse/human whole-genome intersection and alignment is exhibited. Spatial correlations of these elements within the mouse genome are studied. It is argued that these power-law distributions and correlations are comprised in part by functional noncoding sequence and ought to be accounted for in estimating the statistical significance of apparent sequence conservation. These inter-genomic correlations of conservation are placed in the context of previously observed intra-genomic correlations, and their possible origins and consequences are discussed.

Conserved distances between vertebrate highly conserved elements

High numbers of sequence element with very high (>95%) sequence conservation between the human and other vertebrate genomes have been reported and ascribed putative cis-regulatory functions. We have investigated the structural relationships between such elements in mammalian genomes and find that not only their sequences, but also the distances between them are significantly (P<2.2x10(-16)) more conserved than corresponding distances between orthologous protein-coding genes or between exons within these genes. Regions of largely conserved distance between consecutive highly conserved elements (HCE) generally overlap previously identified HCE clusters, but may be far longer (up to 20 Mb) and possibly cover close to 25% of the human genome sequence. Similar conservation of distance is found between bird (chicken) and mammalian genomes and is also discernible in comparisons between fish and mammals. The data suggest either that a substantial amount of essential (functionally active) elements with lower sequence conservation occupy the space between the HCEs or that distance itself is an important factor in transcriptional regulation or chromatin modelling.

Distinct functions of homeodomain-containing and homeodomain-less isoforms encoded by homothorax

The homothorax (hth) gene of Drosophila melanogaster is required for executing Hox functions, for head development, and for forming the proximodistal (PD) axis of the appendages. We show that alternative splicing of hth generates two types of protein isoforms, one that contains a DNA-binding homeodomain (HthFL) and one that does not contain a homeodomain (HDless). Both types of Hth isoforms include the evolutionarily conserved HM domain, which mediates a direct interaction with Extradenticle (Exd), another homeodomain protein. We show that although both HthFL and HDless isoforms of Hth can induce the nuclear localization of Exd, they carry out distinct sets of functions during development. Surprisingly, we find that many of hth's functions, including PD patterning and most Hox-related activities, can be executed by the HDless isoforms. In contrast, antennal development shows an absolute dependency on the HthFL isoform. Thus, alternative splicing of hth results in the generation of multiple transcription factors that execute unique functions in vivo. We further demonstrate that the mouse ortholog of hth, Meis1, also encodes a HDless isoform, suggesting that homeodomain-less variants of this gene family are evolutionarily ancient.

MicroRNA enrichment among short 'ultraconserved' sequences in insects

MicroRNAs are short (∼22 nt) regulatory RNA molecules that play key roles in metazoan development and have been implicated in human disease. First discovered in Caenorhabditis elegans, over 2500 microRNAs have been isolated in metazoans and plants; it has been estimated that there may be more than a thousand microRNA genes in the human genome alone. Motivated by the experimental observation of strong conservation of the microRNA let-7 among nearly all metazoans, we developed a novel methodology to characterize the class of such strongly conserved sequences: we identified a non-redundant set of all sequences 20 to 29 bases in length that are shared among three insects: fly, bee and mosquito. Among the few hundred sequences greater than 20 bases in length are close to 40% of the 78 confirmed fly microRNAs, along with other non-coding RNAs and coding sequence.

Conservation patterns in different functional sequence categories of divergent Drosophila species

We have explored the distributions of fully conserved ungapped blocks in genome-wide pair-wise alignments of recently completed species of Drosophila: D. melanogaster, D. yakuba, D. ananassae, D. pseudoobscura, D. virilis, and D. mojavensis. Based on these distributions we have found that nearly every functional sequence category possesses its own distinctive conservation pattern, sometimes independent of the overall sequence conservation level. In the coding and regulatory regions, the ungapped blocks were longer than in introns, UTRs, and nonfunctional sequences. At the same time, the blocks in the coding regions carried a 3N + 2 signature characteristic of synonymous substitutions in the third-codon position. Larger block sizes in transcription regulatory regions can be explained by the presence of conserved arrays of binding sites for transcription factors. We also have shown that the longest ungapped blocks, or "ultraconserved" sequences, are associated with specific gene groups, including those encoding ion channels and components of the cytoskeleton. We discuss how restraining conservation patterns may help in mapping functional sequence categories and improve genome annotation.

Non-coding RNA

The term non-coding RNA (ncRNA) is commonly employed for RNA that does not encode a protein, but this does not mean that such RNAs do not contain information nor have function. Although it has been generally assumed that most genetic information is transacted by proteins, recent evidence suggests that the majority of the genomes of mammals and other complex organisms is in fact transcribed into ncRNAs, many of which are alternatively spliced and/or processed into smaller products. These ncRNAs include microRNAs and snoRNAs (many if not most of which remain to be identified), as well as likely other classes of yet-to-be-discovered small regulatory RNAs, and tens of thousands of longer transcripts (including complex patterns of interlacing and overlapping sense and antisense transcripts), most of whose functions are unknown. These RNAs (including those derived from introns) appear to comprise a hidden layer of internal signals that control various levels of gene expression in physiology and development, including chromatin architecture/epigenetic memory, transcription, RNA splicing, editing, translation and turnover. RNA regulatory networks may determine most of our complex characteristics, play a significant role in disease and constitute an unexplored world of genetic variation both within and between species.

Mapping of conserved RNA secondary structures predicts thousands of functional noncoding RNAs in the human genome

In contrast to the fairly reliable and complete annotation of the protein coding genes in the human genome, comparable information is lacking for noncoding RNAs (ncRNAs). We present a comparative screen of vertebrate genomes for structural noncoding RNAs, which evaluates conserved genomic DNA sequences for signatures of structural conservation of base-pairing patterns and exceptional thermodynamic stability. We predict more than 30,000 structured RNA elements in the human genome, almost 1,000 of which are conserved across all vertebrates. Roughly a third are found in introns of known genes, a sixth are potential regulatory elements in untranslated regions of protein-coding mRNAs and about half are located far away from any known gene. Only a small fraction of these sequences has been described previously. A comparison with recent tiling array data shows that more than 40% of the predicted structured RNAs overlap with experimentally detected sites of transcription. The widespread conservation of secondary structure points to a large number of functional ncRNAs and cis-acting mRNA structures in the human genome.

Genome-wide identification of human functional DNA using a neutral indel model

It has become clear that a large proportion of functional DNA in the human genome does not code for protein. Identification of this non-coding functional sequence using comparative approaches is proving difficult and has previously been thought to require deep sequencing of multiple vertebrates. Here we introduce a new model and comparative method that, instead of nucleotide substitutions, uses the evolutionary imprint of insertions and deletions (indels) to infer the past consequences of selection. The model predicts the distribution of indels under neutrality, and shows an excellent fit to human-mouse ancestral repeat data. Across the genome, many unusually long ungapped regions are detected that are unaccounted for by the neutral model, and which we predict to be highly enriched in functional DNA that has been subject to purifying selection with respect to indels. We use the model to determine the proportion under indel-purifying selection to be between 2.56% and 3.25% of human euchromatin. Since annotated protein-coding genes comprise only 1.2% of euchromatin, these results lend further weight to the proposition that more than half the functional complement of the human genome is non-protein-coding. The method is surprisingly powerful at identifying selected sequence using only two or three mammalian genomes. Applying the method to the human, mouse, and dog genomes, we identify 90 Mb of human sequence under indel-purifying selection, at a predicted 10% false-discovery rate and 75% sensitivity. As expected, most of the identified sequence represents unannotated material, while the recovered proportions of known protein-coding and microRNA genes closely match the predicted sensitivity of the method. The method's high sensitivity to functional sequence such as microRNAs suggest that as yet unannotated microRNA genes are enriched among the sequences identified. Furthermore, its independence of substitutions allowed us to identify sequence that has been subject to heterogeneous selection, that is, sequence subject to both positive selection with respect to substitutions and purifying selection with respect to indels. The ability to identify elements under heterogeneous selection enables, for the first time, the genome-wide investigation of positive selection on functional elements other than protein-coding genes.

Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes

We have conducted a comprehensive search for conserved elements in vertebrate genomes, using genome-wide multiple alignments of five vertebrate species (human, mouse, rat, chicken, and Fugu rubripes). Parallel searches have been performed with multiple alignments of four insect species (three species of Drosophila and Anopheles gambiae), two species of Caenorhabditis, and seven species of Saccharomyces. Conserved elements were identified with a computer program called phastCons, which is based on a two-state phylogenetic hidden Markov model (phylo-HMM). PhastCons works by fitting a phylo-HMM to the data by maximum likelihood, subject to constraints designed to calibrate the model across species groups, and then predicting conserved elements based on this model. The predicted elements cover roughly 3%-8% of the human genome (depending on the details of the calibration procedure) and substantially higher fractions of the more compact Drosophila melanogaster (37%-53%), Caenorhabditis elegans (18%-37%), and Saccharaomyces cerevisiae (47%-68%) genomes. From yeasts to vertebrates, in order of increasing genome size and general biological complexity, increasing fractions of conserved bases are found to lie outside of the exons of known protein-coding genes. In all groups, the most highly conserved elements (HCEs), by log-odds score, are hundreds or thousands of bases long. These elements share certain properties with ultraconserved elements, but they tend to be longer and less perfectly conserved, and they overlap genes of somewhat different functional categories. In vertebrates, HCEs are associated with the 3' UTRs of regulatory genes, stable gene deserts, and megabase-sized regions rich in moderately conserved noncoding sequences. Noncoding HCEs also show strong statistical evidence of an enrichment for RNA secondary structure.