Integrated gene and species phylogenies from unaligned whole genome protein sequences

MOTIVATION: Most molecular phylogenies are based on sequence alignments. Consequently, they fail to account for modes of sequence evolution that involve frequent insertions or deletions. Here we present a method for generating accurate gene and species phylogenies from whole genome sequence that makes use of short character string matches not placed within explicit alignments. In this work, the singular value decomposition of a sparse tetrapeptide frequency matrix is used to represent the proteins of organisms uniquely and precisely as vectors in a high-dimensional space. Vectors of this kind can be used to calculate pairwise distance values based on the angle separating the vectors, and the resulting distance values can be used to generate phylogenetic trees. Protein trees so derived can be examined directly for homologous sequences. Alternatively, vectors defining each of the proteins within an organism can be summed to provide a vector representation of the organism, which is then used to generate species trees. RESULTS: Using a large mitochondrial genome dataset, we have produced species trees that are largely in agreement with previously published trees based on the analysis of identical datasets using different methods. These trees also agree well with currently accepted phylogenetic theory. In principle, our method could be used to compare much larger bacterial or nuclear genomes in full molecular detail, ultimately allowing accurate gene and species relationships to be derived from a comprehensive comparison of complete genomes. In contrast to phylogenetic methods based on alignments, sequences that evolve by relative insertion or deletion would tend to remain recognizably similar.     

Alignment and topological accuracy of the direct optimization approach via POY and traditional phylogenetics via ClustalW + PAUP*

Direct optimization frameworks for simultaneously estimating alignments and phylogenies have recently been developed. One such method, implemented in the program POY, is becoming more common for analyses of variable length sequences (e.g., analyses using ribosomal genes) and for combined evidence analyses (morphology + multiple genes). Simulation of sequences containing insertion and deletion events was performed in order to directly compare a widely used method of multiple sequence alignment (ClustalW) and subsequent parsimony analysis in PAUP* with direct optimization via POY. Data sets were simulated for pectinate, balanced, and random tree shapes under different conditions (clocklike, non-clocklike, and ultrametric). Alignment accuracy scores for the implied alignments from POY and the multiple sequence alignments from ClustalW were calculated and compared. In almost all cases (99.95%), ClustalW produced more accurate alignments than POY-implied alignments, judged by the proportion of correctly identified homologous sites. Topological accuracy (distance to the true tree) for POY topologies and topologies generated under parsimony in PAUP* from the ClustalW alignments were also compared. In 44.94% of the cases, Clustal alignment tree reconstructions via PAUP* were more accurate than POY, whereas in 16.71% of the cases POY reconstructions were more topologically accurate (38.38% of the time they were equally accurate). Comparisons between POY hypothesized alignments and the true alignments indicated that, on average, as alignment error increased, topological accuracy decreased.     

Distribution of Indel lengths

Protein sequence alignment has become a widely used method in the study of newly sequenced proteins. Most sequence alignment methods use an affine gap penalty to assign scores to insertions and deletions. Although affine gap penalties represent the relative ease of extending a gap compared with initializing a gap, it is still an obvious oversimplification of the real processes that occur during sequence evolution. To improve the efficiency of sequence alignment methods and to obtain a better understanding of the process of sequence evolution, we wanted to find a more accurate model of insertions and deletions in homologous proteins. In this work, we extract the probability of a gap occurrence and the resulting gap length distribution in distantly related proteins (sequence identity < 25%) using alignments based on their common structures. We observe a distribution of gaps that can be fitted with a multiexponential with four distinct components. The results suggest new approaches to modeling insertions and deletions in sequence alignments.     

Mind the gaps: evidence of bias in estimates of multiple sequence alignments

Multiple sequence alignment (MSA) is a crucial first step in the analysis of genomic and proteomic data. Commonly occurring sequence features, such as deletions and insertions, are known to affect the accuracy of MSA programs, but the extent to which alignment accuracy is affected by the positions of insertions and deletions has not been examined independently of other sources of sequence variation. We assessed the performance of 6 popular MSA programs (ClustalW, DIALIGN-T, MAFFT, MUSCLE, PROBCONS, and T-COFFEE) and one experimental program, PRANK, on amino acid sequences that differed only by short regions of deleted residues. The analysis showed that the absence of residues often led to an incorrect placement of gaps in the alignments, even though the sequences were otherwise identical. In data sets containing sequences with partially overlapping deletions, most MSA programs preferentially aligned the gaps vertically at the expense of incorrectly aligning residues in the flanking regions. Of the programs assessed, only DIALIGN-T was able to place overlapping gaps correctly relative to one another, but this was usually context dependent and was observed only in some of the data sets. In data sets containing sequences with non-overlapping deletions, both DIALIGN-T and MAFFT (G-INS-I) were able to align gaps with near-perfect accuracy, but only MAFFT produced the correct alignment consistently. The same was true for data sets that comprised isoforms of alternatively spliced gene products: both DIALIGN-T and MAFFT produced highly accurate alignments, with MAFFT being the more consistent of the 2 programs. Other programs, notably T-COFFEE and ClustalW, were less accurate. For all data sets, alignments produced by different MSA programs differed markedly, indicating that reliance on a single MSA program may give misleading results. It is therefore advisable to use more than one MSA program when dealing with sequences that may contain deletions or insertions, particularly for high-throughput and pipeline applications where manual refinement of each alignment is not practicable.     

Sensitivity of Ribosomal RNA Character Sampling in the Phylogeny of Rhabditida

Near-full-length 18S and 28S rRNA gene sequences were obtained for 33 nematode species. Datasets were constructed based on secondary structure and progressive multiple alignments, and clades were compared for phylogenies inferred by Bayesian and maximum likelihood methods. Clade comparisons were also made following removal of ambiguously aligned sites as determined using the program ProAlign. Different alignments of these data produced tree topologies that differed, sometimes markedly, when analyzed by the same inference method. With one exception, the same alignment produced an identical tree topology when analyzed by different methods. Removal of ambiguously aligned sites altered the tree topology and also reduced resolution. Nematode clades were sensitive to differences in multiple alignments, and more than doubling the amount of sequence data by addition of 28S rRNA did not fully mitigate this result. Although some individual clades showed substantially higher support when 28S data were combined with 18S data, the combined analysis yielded no statistically significant increases in the number of clades receiving higher support when compared to the 18S data alone. Secondary structure alignment increased accuracy in positional homology assignment and, when used in combination with paired-site substitution models, these structural hypotheses of characters and improved models of character state change yielded high levels of phylogenetic resolution. Phylogenetic results included strong support for inclusion of Daubaylia potomaca within Cephalobidae, whereas the position of Fescia grossa within Tylenchina varied depending on the alignment, and the relationships among Rhabditidae, Diplogastridae, and Bunonematidae were not resolved.     

Evolution of Protein Binding Modes in Homooligomers

The evolution of protein interactions cannot be deciphered without a detailed analysis of interaction interfaces and binding modes. We performed a large-scale study of protein homooligomers in terms of their symmetry, interface sizes, and conservation of binding modes. We also focused specifically on the evolution of protein binding modes from nine families of homooligomers and mapped 60 different binding modes and oligomerization states onto the phylogenetic trees of these families. We observed a significant tendency for the same binding modes to be clustered together and conserved within clades on phylogenetic trees; this trend is especially pronounced for close homologs with 70% sequence identity or higher. Some binding modes are conserved among very distant homologs, pointing to their ancient evolutionary origin, while others are very specific for a certain phylogenetic group. Moreover, we found that the most ancient binding modes have a tendency to involve symmetrical (isologous) homodimer binding arrangements with larger interfaces, while recently evolved binding modes more often exhibit asymmetrical arrangements and smaller interfaces.     

Insertions and deletions as phylogenetic signal in an alignment-free context

Most methods for phylogenetic tree reconstruction are based on sequence alignments; they infer phylogenies from substitutions that may have occurred at the aligned sequence positions. Gaps in alignments are usually not employed as phylogenetic signal. In this paper, we explore an alignment-free approach that uses insertions and deletions (indels) as an additional source of information for phylogeny inference. For a set of four or more input sequences, we generate so-called quartet blocks of four putative homologous segments each. For pairs of such quartet blocks involving the same four sequences, we compare the distances between the two blocks in these sequences, to obtain hints about indels that may have happened between the blocks since the respective four sequences have evolved from their last common ancestor. A prototype implementation that we call Gap-SpaM is presented to infer phylogenetic trees from these data, using a quartet-tree approach or, alternatively, under the maximum-parsimony paradigm. This approach should not be regarded as an alternative to established methods, but rather as a complementary source of phylogenetic information. Interestingly, however, our software is able to produce phylogenetic trees from putative indels alone that are comparable to trees obtained with existing alignment-free methods.     

Quality of alignment comparison by COMPASS improves with inclusion of diverse confident homologs

MOTIVATION: Adding more distant homologs to a multiple alignment and thus increasing its diversity may eventually deteriorate the numerical profile constructed from this alignment. Here, we addressed the question whether such a diversity limit can be reached in the alignments of confident homologs found by PSI-BLAST, and we analyzed the dependence of the quality of the profile-profile comparison made by COMPASS on the sequence diversity within these alignments. RESULTS: Protein families that have a greater number of diverse confident homologs in the current sequence databases provide an increased quality of similarity detection in profile databases, but produce on average less accurate profile-profile alignments with their remote relatives. This lower alignment accuracy cannot be improved when the most distant members of these families are excluded from their profiles. On the contrary, the presence of more diverse members results in more accurate alignments. For families with a high diversity of confident homologs, the lower quality of profile alignments with their remote relatives seems to be an attribute of these families or their alignments, rather than to be caused by the large number of diverse sequences itself. Our results suggest that at any level of profile diversity, one should include in the multiple alignment as many confident sequence homologs as possible in order to produce the most accurate results.     

Scaling statistical multiple sequence alignment to large datasets

Background

Multiple sequence alignment is an important task in bioinformatics, and alignments of large datasets containing hundreds or thousands of sequences are increasingly of interest. While many alignment methods exist, the most accurate alignments are likely to be based on stochastic models where sequences evolve down a tree with substitutions, insertions, and deletions. While some methods have been developed to estimate alignments under these stochastic models, only the Bayesian method BAli-Phy has been able to run on even moderately large datasets, containing 100 or so sequences. A technique to extend BAli-Phy to enable alignments of thousands of sequences could potentially improve alignment and phylogenetic tree accuracy on large-scale data beyond the best-known methods today.

Results

We use simulated data with up to 10,000 sequences representing a variety of model conditions, including some that are significantly divergent from the statistical models used in BAli-Phy and elsewhere. We give a method for incorporating BAli-Phy into PASTA and UPP, two strategies for enabling alignment methods to scale to large datasets, and give alignment and tree accuracy results measured against the ground truth from simulations. Comparable results are also given for other methods capable of aligning this many sequences.

Conclusions

Extensions of BAli-Phy using PASTA and UPP produce significantly more accurate alignments and phylogenetic trees than the current leading methods.

     

MIR: Family of repeats common to vertebrate genomes

We studied the occurrence of mammalian interspersed repeats (MIRs) in DNA and RNA of vertebrates, invertebrates, and bacteria using the data from GenBank. A special algorithm based on a weight position matrix with optimal alignment using dynamic programming was developed to search for the traces of MIR dissemination. This allowed us to search for highly divergent MIRs carrying deletions and insertions. MIRs were detected in genomes of various fishes, includingLatimeria. This suggests that the origin of MIRs dates back more than 400 million years. The method to search for similarity between highly divergent sequences may be used to find the genome fragments from various ancient repeat families and from various gene families.     

RAxML and FastTree: comparing two methods for large-scale maximum likelihood phylogeny estimation

Statistical methods for phylogeny estimation, especially maximum likelihood (ML), offer high accuracy with excellent theoretical properties. However, RAxML, the current leading method for large-scale ML estimation, can require weeks or longer when used on datasets with thousands of molecular sequences. Faster methods for ML estimation, among them FastTree, have also been developed, but their relative performance to RAxML is not yet fully understood. In this study, we explore the performance with respect to ML score, running time, and topological accuracy, of FastTree and RAxML on thousands of alignments (based on both simulated and biological nucleotide datasets) with up to 27,634 sequences. We find that when RAxML and FastTree are constrained to the same running time, FastTree produces topologically much more accurate trees in almost all cases. We also find that when RAxML is allowed to run to completion, it provides an advantage over FastTree in terms of the ML score, but does not produce substantially more accurate tree topologies. Interestingly, the relative accuracy of trees computed using FastTree and RAxML depends in part on the accuracy of the sequence alignment and dataset size, so that FastTree can be more accurate than RAxML on large datasets with relatively inaccurate alignments. Finally, the running times of RAxML and FastTree are dramatically different, so that when run to completion, RAxML can take several orders of magnitude longer than FastTree to complete. Thus, our study shows that very large phylogenies can be estimated very quickly using FastTree, with little (and in some cases no) degradation in tree accuracy, as compared to RAxML.     

Inferring phylogenies from RAD sequence data

Reduced-representation genome sequencing represents a new source of data for systematics, and its potential utility in interspecific phylogeny reconstruction has not yet been explored. One approach that seems especially promising is the use of inexpensive short-read technologies (e.g., Illumina, SOLiD) to sequence restriction-site associated DNA (RAD)--the regions of the genome that flank the recognition sites of restriction enzymes. In this study, we simulated the collection of RAD sequences from sequenced genomes of different taxa (Drosophila, mammals, and yeasts) and developed a proof-of-concept workflow to test whether informative data could be extracted and used to accurately reconstruct "known" phylogenies of species within each group. The workflow consists of three basic steps: first, sequences are clustered by similarity to estimate orthology; second, clusters are filtered by taxonomic coverage; and third, they are aligned and concatenated for "total evidence" phylogenetic analysis. We evaluated the performance of clustering and filtering parameters by comparing the resulting topologies with well-supported reference trees and we were able to identify conditions under which the reference tree was inferred with high support. For Drosophila, whole genome alignments allowed us to directly evaluate which parameters most consistently recovered orthologous sequences. For the parameter ranges explored, we recovered the best results at the low ends of sequence similarity and taxonomic representation of loci; these generated the largest supermatrices with the highest proportion of missing data. Applications of the method to mammals and yeasts were less successful, which we suggest may be due partly to their much deeper evolutionary divergence times compared to Drosophila (crown ages of approximately 100 and 300 versus 60 Mya, respectively). RAD sequences thus appear to hold promise for reconstructing phylogenetic relationships in younger clades in which sufficient numbers of orthologous restriction sites are retained across species.     

Improved sensitivity of profile searches through the use of sequence weights and gap excision

Position-specific substitution matrices, known as profiles,derived from multiple sequence alignments are currently usedto search sequence databases for distantly related members ofprotein families. The performance of the database searches isenhanced by using (i) a sequence weighting scheme which assignshigher weights to more distantly related sequences based onbranch lengths derived from phylogenetic trees, (ii) exclusionof positions with mainly padding characters at sites of insertionsor deletions and (iii) the BLOSUM62 residue comparison matrix.A natural consequence of these modifications is an improvementin the alignment of new sequences to the profiles. However,the accuracy of the alignments can be further increased by employinga similarity residue comparison matrix. These developments areimplemented in a program called PROFILEWEIGHT which runs onUnix and Vax computers. The only input required by the programis the multiple sequence alignment. The output from PROFILEWEIGHTis a profile designed to be used by existing searching and alignmentprograms. Test results from database searches with four differentfamilies of proteins show the improved sensitivity of the weightedprofiles.     

Combining evolutionary information and neural networks to predict protein secondary structure

Using evolutionary information contained in multiple sequence alignments as input to neural networks, secondary structure can be predicted at significantly increased accuracy. Here, we extend our previous three-level system of neural networks by using additional input information derived from multiple alignments. Using a position-specific conservation weight as part of the input increases performance. Using the number of insertions and deletions reduces the tendency for overprediction and increases overall accuracy. Addition of the global amino acid content yields a further improvement, mainly in predicting structural class. The final network system has a sustained overall accuracy of 71.6% in a multiple cross-validation test on 126 unique protein chains. A test on a new set of 124 recently solved protein structures that have no significant sequence similarity to the learning set confirms the high level of accuracy. The average cross-validated accuracy for all 250 sequence-unique chains is above 72%. Using various data sets, the method is compared to alternative prediction methods, some of which also use multiple alignments: the performance advantage of the network system is at least 6 percentage points in three-state accuracy. In addition, the network estimates secondary structure content from multiple sequence alignments about as well as circular dichroism spectroscopy on a single protein and classifies 75% of the 250 proteins correctly into one of four protein structural classes. Of particular practical importance is the definition of a position-specific reliability index. For 40% of all residues the method has a sustained three-state accuracy of 88%, as high as the overall average for homology modelling. A further strength of the method is greatly increased accuracy in predicting the placement of secondary structure segments. © 1994 Wiley-Liss, Inc.     

Evolutionary distances in the twilight zone--a rational kernel approach

Phylogenetic tree reconstruction is traditionally based on multiple sequence alignments (MSAs) and heavily depends on the validity of this information bottleneck. With increasing sequence divergence, the quality of MSAs decays quickly. Alignment-free methods, on the other hand, are based on abstract string comparisons and avoid potential alignment problems. However, in general they are not biologically motivated and ignore our knowledge about the evolution of sequences. Thus, it is still a major open question how to define an evolutionary distance metric between divergent sequences that makes use of indel information and known substitution models without the need for a multiple alignment. Here we propose a new evolutionary distance metric to close this gap. It uses finite-state transducers to create a biologically motivated similarity score which models substitutions and indels, and does not depend on a multiple sequence alignment. The sequence similarity score is defined in analogy to pairwise alignments and additionally has the positive semi-definite property. We describe its derivation and show in simulation studies and real-world examples that it is more accurate in reconstructing phylogenies than competing methods. The result is a new and accurate way of determining evolutionary distances in and beyond the twilight zone of sequence alignments that is suitable for large datasets.     

The performance of several multiple-sequence alignment programs in relation to secondary-structure features for an rRNA sequence

The performances of five global multiple-sequence alignment programs (CLUSTAL W, Divide and Conquer, Malign, PileUp, and TreeAlign) were evaluated using part of the animal mitochondrial small subunit (12S) rRNA molecule. Conserved sequence motifs derived from an alignment based on secondary structural information were used to score how well each program aligned a data set of five vertebrate and five invertebrate taxa over a range of parameter values. All of the programs could align the motifs with reasonable accuracy for at least one set of parameter conditions, although if the whole sequence was considered, similarity to the structural alignment was only 25%-34%. Use of small gap costs generally gave more accurate results, although Malign and TreeAlign generated longer alignments when gap costs were low. The programs differed in the consistency of the alignments when gap cost was varied; CLUSTAL W, Divide and Conquer, and TreeAlign were the most accurate and robust, while PileUp performed poorly as gap cost values increased, and the accuracy of Malign fluctuated. Default settings for the programs did not give the best results, and attempting to select similar parameter values in different programs did not always result in more similar alignments. Poor alignment of even well-conserved motifs can occur if these are near sites with insertions or deletions. Since there is no a priori way to determine gap costs and because such costs can vary over the gene, alignment of rRNA sequences, particularly the less well conserved regions, should be treated carefully and aided by secondary structure and conserved motifs. Some motifs are single bases and so are often invisible to alignment programs. Our tests involved the most conserved regions of the 12S rRNA gene, and alignment of less well conserved regions will be more problematical. None of the alignments we examined produced a fully resolved phylogeny for the data set, indicating that this portion of 12S rRNA is insufficient for resolution of distant evolutionary relationships.     

The evolutionary history of the coral genus Acropora (Scleractinia, Cnidaria) based on a mitochondrial and a nuclear marker: reticulation, incomplete lineage sorting, or morphological convergence?

This study examines molecular relationships across a wide range of species in the mass spawning scleractinian coral genus Acropora. Molecular phylogenies were obtained for 28 species using DNA sequence analyses of two independent markers, a nuclear intron and the mtDNA putative control region. Although the compositions of the major clades in the phylogenies based on these two markers were similar, there were several important differences. This, in combination with the fact that many species were not monophyletic, suggests either that introgressive hybridization is occurring or that lineage sorting is incomplete. The molecular tree topologies bear little similarity to the results of a recent cladistic analysis based on skeletal morphology and are at odds with the fossil record. We hypothesize that these conflicting results may be due to the same morphology having evolved independently more than once in Acropora and/or the occurrence of extensive interspecific hybridization and introgression in combination with morphology being determined by a small number of genes. Our results indicate that many Acropora species belong to a species complex or syngameon and that morphology has little predictive value with regard to syngameon composition. Morphological species in the genus often do not correspond to genetically distinct evolutionary units. Instead, species that differ in timing of gamete release tend to constitute genetically distinct clades.     

A galaxy of folds

Many protein classification systems capture homologous relationships by grouping domains into families and superfamilies on the basis of sequence similarity. Superfamilies with similar 3D structures are further grouped into folds. In the absence of discernable sequence similarity, these structural similarities were long thought to have originated independently, by convergent evolution. However, the growth of databases and advances in sequence comparison methods have led to the discovery of many distant evolutionary relationships that transcend the boundaries of superfamilies and folds. To investigate the contributions of convergent versus divergent evolution in the origin of protein folds, we clustered representative domains of known structure by their sequence similarity, treating them as point masses in a virtual 2D space which attract or repel each other depending on their pairwise sequence similarities. As expected, families in the same superfamily form tight clusters. But often, superfamilies of the same fold are linked with each other, suggesting that the entire fold evolved from an ancient prototype. Strikingly, some links connect superfamilies with different folds. They arise from modular peptide fragments of between 20 and 40 residues that co‐occur in the connected folds in disparate structural contexts. These may be descendants of an ancestral pool of peptide modules that evolved as cofactors in the RNA world and from which the first folded proteins arose by amplification and recombination. Our galaxy of folds summarizes, in a single image, most known and many yet undescribed homologous relationships between protein superfamilies, providing new insights into the evolution of protein domains.