The Genealogical Sorting Index (gsi) has been widely used in species-delimitation studies, where it is usually interpreted as a measure of the degree to which each of several predefined groups of specimens display a pattern of divergent evolution in a phylogenetic tree. Here we show that the gsi value obtained for a given group is highly dependent on the structure of the tree outside of the group of interest. By calculating the gsi from simulated datasets we demonstrate this dependence undermines some of desirable properties of the statistic. We also review the use of the gsi delimitation studies, and show that the gsi has typically been used under scenarios in which it is expected to produce large and statistically significant results for samples that are not divergent from all other populations and thus should not be considered species. Our proposed solution to this problem performs better than the gsi in under these conditions. Nevertheless, we show that our modified approach can produce positive results for populations that are connected by substantial levels of gene flow, and are thus unlikely to represent distinct species. We stress that the properties of gsi made clear in this manuscript must be taken into account if the statistic is used in species-delimitation studies. More generally, we argue that the results of genetic species-delimitation methods need to be interpreted in the light the biological and ecological setting of a study, and not treated as the final test applied to hypotheses generated by other data.

Inferring the ancestral dynamics of effective population size is a long-standing question in population genetics, which can now be tackled much more accurately thanks to the massive genomic data available in many species. Several promising methods that take advantage of whole-genome sequences have been recently developed in this context. However, they can only be applied to rather small samples, which limits their ability to estimate recent population size history. Besides, they can be very sensitive to sequencing or phasing errors. Here we introduce a new approximate Bayesian computation approach named PopSizeABC that allows estimating the evolution of the effective population size through time, using a large sample of complete genomes. This sample is summarized using the folded allele frequency spectrum and the average zygotic linkage disequilibrium at different bins of physical distance, two classes of statistics that are widely used in population genetics and can be easily computed from unphased and unpolarized SNP data. Our approach provides accurate estimations of past population sizes, from the very first generations before present back to the expected time to the most recent common ancestor of the sample, as shown by simulations under a wide range of demographic scenarios. When applied to samples of 15 or 25 complete genomes in four cattle breeds (Angus, Fleckvieh, Holstein and Jersey), PopSizeABC revealed a series of population declines, related to historical events such as domestication or modern breed creation. We further highlight that our approach is robust to sequencing errors, provided summary statistics are computed from SNPs with common alleles.

The phylogenetic likelihood function is the major computational bottleneck in maximum likelihood and Bayesian phylogenetic inference. Given the alignment, a tree and the evolutionary model parameters, the likelihood function computes the conditional likelihood vectors for every node of the tree. Vector entries for which all input data are identical result in repeated likelihood operations which, in turn, yield identical conditional values. Such operations can be omitted for improving run-time and, by using appropriate data structures, also reduce memory usage. Here, we present a novel, fast method for identifying and omitting such redundant operations in phylogenetic likelihood calculations, and assess the performance improvement that is attained by our method. Using empirical and real data sets, we show that a prototype implementation of our method yields up to 10-fold speedups and uses up to 78% less memory than one of the fastest and most highly tuned implementations of the phylogenetic likelihood function available. Our method is generic and can seamlessly be integrated into any phylogenetic likelihood implementation.

The historic developmental hourglass concept depicts the convergence of animal embryos to a common form during the phylotypic period. Recently, it has been shown that a transcriptomic hourglass is associated with this morphological pattern, consistent with the idea of underlying selective constraints due to intense molecular interactions during body plan establishment. Although plants do not exhibit a morphological hourglass during embryogenesis, a transcriptomic hourglass has nevertheless been identified in the model plant Arabidopsis thaliana. Here, we investigated whether plant hourglass patterns are also found post-embryonically. We found that the two main phase changes during the life cycle of Arabidopsis, from embryonic to vegetative and from vegetative to reproductive development, are associated with transcriptomic hourglass patterns. In contrast, flower development, a process dominated by organ formation, is not. This suggests that plant hourglass patterns are decoupled from organogenesis and body plan establishment. Instead, they may reflect general transitions through organizational checkpoints.

Transposable elements (TEs) are a major source of genome variation across the branches of life. Although TEs may occasionally play an adaptive role in their host’s genome, they are much more often deleterious, and purifying selection is thus an important factor controlling genomic TE loads. In contrast, life history and genomic characteristics such as mating system, parasitism, GC content, and RNAi pathways, have been suggested to account for the startling disparity of TE loads in different species. Previous studies of fungal, plant, and animal genomes have reported conflicting results regarding the direction in which these genomic features drive TE evolution. Many of these studies have had limited power because they studied taxonomically narrow systems, comparing only a limited number of phylogenetically independent contrasts, and did not address long term effects on TE evolution. Here we explicitly test the long term determinants of TE evolution by comparing 42 nematode genomes that span over 500 million years of diversification, and include numerous transitions between life history states and RNAi pathways. We have analysed the reconstructed TE loads of ancestors through the Nematoda phylogeny to account for correlation with GC content and transitions in TE evolutionary models. We also analysed the effect of transitions in life history characteristics and RNAi using ANOVA of phylogenetically independent contrasts. We show that purifying selection against TEs is the dominant force throughout the evolutionary history of Nematoda, as indicated by reconstructed ancestral TE loads, and that strong stochastic Ornstein-Uhlenbeck processes are the underlying models which best explain TE diversification among extant species. In contrast we found no evidence that life history or RNAi variations have a significant influence upon genomic TE load across extended periods of evolutionary history. We suggest that these are largely inconsequential to the large differences in TE content observed between genomes and only by these large-scale comparisons can we distinguish long term and persistent effects from transient effects or misleading random changes.