SPRIT: Identifying horizontal gene transfer in rooted phylogenetic trees

A Novel Glaesserella sp. Isolated from Pigs with Severe Respiratory Infections Has a Mosaic Genome with Virulence Factors Putatively Acquired by Horizontal Transfer

An unknown member of the family Pasteurellaceae was repeatedly isolated from 20- to 24-week-old pigs with severe pulmonary lesions reared on the same farm in Victoria, Australia. The etiological diagnosis of the disease was inconclusive. The complete genome sequence analysis of one strain, 15-184, revealed some phylogenic proximity to Glaesserella (Haemophilus) parasuis, the cause of Glasser's disease. However, the sequences of the 16S rRNA and housekeeping genes, as well as the average nucleotide identity scores, differed from those of all other known species in the family Pasteurellaceae The protein content of 15-184 was composite, with 60% of coding sequences matching known G. parasuis products, while more than 20% had a closer relative in the genera Actinobacillus, Mannheimia, Pasteurella, and Bibersteinia Several putative virulence genes absent from G. parasuis but present in other Pasteurellaceae were also found, including the apxIII RTX toxin gene from Actinobacillus pleuropneumoniae, ABC transporters from Actinobacillus minor, and iron transporters from various species. Three prophages and one integrative conjugative element were present in the isolate. Horizontal gene transfers might explain the mosaic genomic structure and atypical metabolic and virulence characteristics of 15-184. This organism has not been assigned a taxonomic position in the family, but this study underlines the need for a large-scale epidemiological and clinical characterization of this novel pathogen in swine populations, as a genomic analysis suggests it could have a severe impact on pig health.IMPORTANCE Several species of Pasteurellaceae cause a range of significant diseases in pigs. A novel member of this family was recently isolated from Australian pigs suffering from severe respiratory infections. Comparative whole-genome analyses suggest that this bacterium represents a new species, which possesses a number of virulence genes horizontally acquired from a diverse range of other Pasteurellaceae While the possible contribution of other coinfecting noncultivable agents to the disease has not been ruled out in this study, the repertoire of virulence genes found in this organism may nevertheless explain some aspects of the associated pathology observed on the farm. The prevalence of this novel pathogen within pig populations is currently unknown. This finding is of particular importance for the pig industry, as this organism can have a serious impact on the health of these animals.

Widespread of horizontal gene transfer in the human genome

Background

A fundamental concept in biology is that heritable material is passed from parents to offspring, a process called vertical gene transfer. An alternative mechanism of gene acquisition is through horizontal gene transfer (HGT), which involves movement of genetic materials between different species. Horizontal gene transfer has been found prevalent in prokaryotes but very rare in eukaryote. In this paper, we investigate horizontal gene transfer in the human genome.

Results

From the pair-wise alignments between human genome and 53 vertebrate genomes, 1,467 human genome regions (2.6 M bases) from all chromosomes were found to be more conserved with non-mammals than with most mammals. These human genome regions involve 642 known genes, which are enriched with ion binding. Compared to known horizontal gene transfer regions in the human genome, there were few overlapping regions, which indicated horizontal gene transfer is more common than we expected in the human genome.

Conclusions

Horizontal gene transfer impacts hundreds of human genes and this study provided insight into potential mechanisms of HGT in the human genome.

Reconciliation of gene and species trees

The first part of the paper briefly overviews the problem of gene and species trees reconciliation with the focus on defining and algorithmic construction of the evolutionary scenario. Basic ideas are discussed for the aspects of mapping definitions, costs of the mapping and evolutionary scenario, imposing time scales on a scenario, incorporating horizontal gene transfers, binarization and reconciliation of polytomous trees, and construction of species trees and scenarios. The review does not intend to cover the vast diversity of literature published on these subjects. Instead, the authors strived to overview the problem of the evolutionary scenario as a central concept in many areas of evolutionary research. The second part provides detailed mathematical proofs for the solutions of two problems: (i) inferring a gene evolution along a species tree accounting for various types of evolutionary events and (ii) trees reconciliation into a single species tree when only gene duplications and losses are allowed. All proposed algorithms have a cubic time complexity and are mathematically proved to find exact solutions. Solving algorithms for problem (ii) can be naturally extended to incorporate horizontal transfers, other evolutionary events, and time scales on the species tree.

Detection of homologous recombination events in bacterial genomes

We study the detection of mutations, sequencing errors, and homologous recombination events (HREs) in a set of closely related microbial genomes. We base the model on single nucleotide polymorphisms (SNPs) and break the genomes into blocks to handle the rearrangement problem. Then we apply a dynamic programming algorithm to model whether changes within each block are likely a result of mutations, sequencing errors, or HREs. Results from simulation experiments show that we can detect 31%–61% of HREs and the precision of our detection is about 48%–90% depending on the rates of mutation and missing data. The HREfinder software for predicting HREs in a set of whole genomes is available as open source (http://sourceforge.net/projects/hrefinder/).

Systematic inference of highways of horizontal gene transfer in prokaryotes

MOTIVATION

Horizontal gene transfer (HGT) plays a crucial role in the evolution of prokaryotic species. Typically, no more than a few genes are horizontally transferred between any two species. However, several studies identified pairs of species (or linages) between which many different genes were horizontally transferred. Such a pair is said to be linked by a highway of gene sharing. Inferring such highways is crucial to understanding the evolution of prokaryotes and for inferring past symbiotic and ecological associations among different species.

RESULTS

We present a new improved method for systematically detecting highways of gene sharing. As we demonstrate using a variety of simulated datasets, our method is highly accurate and efficient, and robust to noise and high rates of HGT. We further validate our method by applying it to a published dataset of >22 000 gene trees from 144 prokaryotic species. Our method makes it practical, for the first time, to perform accurate highway analysis quickly and easily even on large datasets with high rates of HGT.

AVAILABILITY AND IMPLEMENTATION

An implementation of the method can be freely downloaded from: http://acgt.cs.tau.ac.il/hide.

Widespread horizontal transfer of retrotransposons

In higher organisms such as vertebrates, it is generally believed that lateral transfer of genetic information does not readily occur, with the exception of retroviral infection. However, horizontal transfer (HT) of protein coding repetitive elements is the simplest way to explain the patchy distribution of BovB, a long interspersed element (LINE) about 3.2 kb long, that has been found in ruminants, marsupials, squamates, monotremes, and African mammals. BovB sequences are a major component of some of these genomes. Here we show that HT of BovB is significantly more widespread than believed, and we demonstrate the existence of two plausible arthropod vectors, specifically reptile ticks. A phylogenetic tree built from BovB sequences from species in all of these groups does not conform to expected evolutionary relationships of the species, and our analysis indicates that at least nine HT events are required to explain the observed topology. Our results provide compelling evidence for HT of genetic material that has transformed vertebrate genomes.

A fast tool for minimum hybridization networks

BACKGROUND

Due to hybridization events in evolution, studying two different genes of a set of species may yield two related but different phylogenetic trees for the set of species. In this case, we want to combine the two phylogenetic trees into a hybridization network with the fewest hybridization events. This leads to three computational problems, namely, the problem of computing the minimum size of a hybridization network, the problem of constructing one minimum hybridization network, and the problem of enumerating a representative set of minimum hybridization networks. The previously best software tools for these problems (namely, Chen and Wang's HybridNet and Albrecht et al.'s Dendroscope 3) run very slowly for large instances that cannot be reduced to relatively small instances. Indeed, when the minimum size of a hybridization network of two given trees is larger than 23 and the problem for the trees cannot be reduced to relatively smaller independent subproblems, then HybridNet almost always takes longer than 1 day and Dendroscope 3 often fails to complete. Thus, a faster software tool for the problems is in need.

RESULTS

We develop a software tool in ANSI C, named FastHN, for the following problems: Computing the minimum size of a hybridization network, constructing one minimum hybridization network, and enumerating a representative set of minimum hybridization networks. We obtain FastHN by refining HybridNet with three ideas. The first idea is to preprocess the input trees so that the trees become smaller or the problem becomes to solve two or more relatively smaller independent subproblems. The second idea is to use a fast algorithm for computing the rSPR distance of two given phylognetic trees to cut more branches of the search tree in the exhaustive-search stage of the algorithm. The third idea is that during the exhaustive-search stage of the algorithm, we find two sibling leaves in one of the two forests (obtained from the given trees by cutting some edges) such that they are as far as possible in the other forest. As the result, FastHN always runs much faster than HybridNet. Unlike Dendroscope 3, FastHN is a single-threaded program. Despite this disadvantage, our experimental data shows that FastHN runs substantially faster than the multi-threaded Dendroscope 3 on a PC with multiple cores. Indeed, FastHN can finish within 16 minutes (on average on a Windows-7 (x64) desktop PC with i7-2600 CPU) even if the minimum size of a hybridization network of two given trees is about 25, the trees each have 100 leaves, and the problem for the input trees cannot be reduced to two or more independent subproblems via cluster reductions. It is also worth mentioning that like HybridNet, FastHN does not use much memory (indeed, the amount of memory is at most quadratic in the input size). In contrast, Dendroscope 3 uses a huge amount of memory. Executables of FastHN for Windows XP (x86), Windows 7 (x64), Linux, and Mac OS are available (see the Results and discussion section for details).

CONCLUSIONS

For both biological datasets and simulated datasets, our experimental results show that FastHN runs substantially faster than HybridNet and Dendroscope 3. The superiority of FastHN in speed over the previous tools becomes more significant as the hybridization number becomes larger. In addition, FastHN uses much less memory than Dendroscope 3 and uses the same amount of memory as HybridNet.

Efficient algorithms for the reconciliation problem with gene duplication, horizontal transfer and loss

MOTIVATION

Gene family evolution is driven by evolutionary events such as speciation, gene duplication, horizontal gene transfer and gene loss, and inferring these events in the evolutionary history of a given gene family is a fundamental problem in comparative and evolutionary genomics with numerous important applications. Solving this problem requires the use of a reconciliation framework, where the input consists of a gene family phylogeny and the corresponding species phylogeny, and the goal is to reconcile the two by postulating speciation, gene duplication, horizontal gene transfer and gene loss events. This reconciliation problem is referred to as duplication-transfer-loss (DTL) reconciliation and has been extensively studied in the literature. Yet, even the fastest existing algorithms for DTL reconciliation are too slow for reconciling large gene families and for use in more sophisticated applications such as gene tree or species tree reconstruction.

RESULTS

We present two new algorithms for the DTL reconciliation problem that are dramatically faster than existing algorithms, both asymptotically and in practice. We also extend the standard DTL reconciliation model by considering distance-dependent transfer costs, which allow for more accurate reconciliation and give an efficient algorithm for DTL reconciliation under this extended model. We implemented our new algorithms and demonstrated up to 100 000-fold speed-up over existing methods, using both simulated and biological datasets. This dramatic improvement makes it possible to use DTL reconciliation for performing rigorous evolutionary analyses of large gene families and enables its use in advanced reconciliation-based gene and species tree reconstruction methods.

AVAILABILITY

Our programs can be freely downloaded from http://compbio.mit.edu/ranger-dtl/.

Detecting highways of horizontal gene transfer

In a horizontal gene transfer (HGT) event, a gene is transferred between two species that do not have an ancestor-descendant relationship. Typically, no more than a few genes are horizontally transferred between any two species. However, several studies identified pairs of species between which many different genes were horizontally transferred. Such a pair is said to be linked by a highway of gene sharing. We present a method for inferring such highways. Our method is based on the fact that the evolutionary histories of horizontally transferred genes disagree with the corresponding species phylogeny. Specifically, given a set of gene trees and a trusted rooted species tree, each gene tree is first decomposed into its constituent quartet trees and the quartets that are inconsistent with the species tree are identified. Our method finds a pair of species such that a highway between them explains the largest (normalized) fraction of inconsistent quartets. For a problem on n species and m input quartet trees, we give an efficient O(m + n(2))-time algorithm for detecting highways, which is optimal with respect to the quartets input size. An application of our method to a dataset of 1128 genes from 11 cyanobacterial species, as well as to simulated datasets, illustrates the efficacy of our method.

Algorithms for reticulate networks of multiple phylogenetic trees

A reticulate network N of multiple phylogenetic trees may have vertices with two or more parents (called reticulation vertices). There are two ways to define the reticulation number of N. One is to define it as the number of reticulation vertices in N; in this case, a reticulate network with the smallest reticulation number is called an optimal type-I reticulate network of the trees. The other is to define it as the total number of parents of reticulation vertices in N minus the number of reticulation vertices in N; in this case, a reticulate network with the smallest reticulation number is called an optimal type-II reticulate network of the trees. In this paper, we present a fast algorithm for constructing one or all optimal type-I reticulate networks of multiple phylogenetic trees. We then use the algorithm together with other ideas to obtain an algorithm for estimating a lower bound on the reticulation number of an optimal type-II reticulate network of the input trees. To our knowledge, these are the first fast algorithms for the problems. Our experimental data shows that our algorithms can construct optimal type-I reticulate networks rapidly and can compute better lower bounds for optimal type-II reticulate networks within much shorter time than the previously best program.

A polynomial-time algorithm computing lower and upper bounds of the rooted subtree prune and regraft distance

Rooted, leaf-labeled trees are used in biology to represent hierarchical relationships of various entities, most notably the evolutionary history of molecules and organisms. Rooted Subtree Prune and Regraft (rSPR) operation is a tree rearrangement operation that is used to transform a tree into another tree that has the same set of leaf labels. The minimum number of rSPR operations that transform one tree into another is denoted by d(rSPR) and gives a measure of dissimilarity between the trees, which can be used to compare trees obtained by different approaches, or, in the context of phylogenetic analysis, to detect horizontal gene transfer events by finding incongruences between trees of different evolving characters. The problem of computing the exact d(rSPR) measure is NP-hard, and most algorithms resort to finding sequences of rSPR operations that are sufficient for transforming one tree into another, thereby giving upper bound heuristics for the distance. In this article, we present an O(n⁴) recursive algorithm D-Clust that gives both lower bound and upper bound heuristics for the distance between trees with n shared leaves and also gives a sequence of operations that transforms one tree into another. Our experiments on simulated pairs of trees containing up to 100 leaves showed that the two bounds are almost equal for small distances, thereby giving the nearly-precise actual value, and that the upper bound tends to be close to the upper bounds given by other approaches for all pairs of trees.