Marginalizing Sample Consensus.

A new method for robust estimation, MAGSAC++, is proposed. It introduces a new model quality (scoring) function that does not make inlier-outlier decisions, and a novel marginalization procedure formulated as an M-estimation with a novel class of M-estimators (a robust kernel) solved by an iteratively re-weighted least squares procedure. Instead of the inlier-outlier threshold, it requires only its loose upper bound which can be chosen from a significantly wider range. Also, we propose a new termination criterion and a technique for selecting a set of inliers in a data-driven manner as a post-processing step after the robust estimation finishes. On a number of publicly available real-world datasets for homography, fundamental matrix fitting and relative pose, MAGSAC++ produces results superior to the state-of-the-art robust methods. It is more geometrically accurate, fails fewer times, and it is often faster. It is shown that MAGSAC++ is significantly less sensitive to the setting of the threshold upper bound than the other state-of-the-art algorithms to the inlier-outlier threshold. Therefore, it is easier to be applied to unseen problems and scenes without acquiring information by hand about the setting of the inlier-outlier threshold. The source code and examples both in C++ and Python are available at https://github.com/danini/magsac.

Parallelism in gene expression between foothill and alpine ecotypes in Arabidopsis arenosa.

Parallel adaptation results from the independent evolution of similar traits between closely related lineages and allows us to test to what extent evolution is repeatable. Similar gene expression changes are often detected but the identity of genes shaped by parallel selection and the causes of expression parallelism remain largely unknown. By comparing genomes and transcriptomes of four distinct foothill-alpine population pairs across four treatments, we addressed the genetic underpinnings, plasticity and functional consequences of gene expression parallelism in alpine adaptation. Seeds of eight populations of Arabidopsis arenosa were raised under four treatments that differed in temperature and irradiance, factors varying strongly with elevation. Parallelism in differential gene expression between the foothill and alpine ecotypes was quantified by RNA-seq in leaves of young plants. By manipulating temperature and irradiance, we also tested for parallelism in plasticity (i.e., gene-environment interaction, GEI). In spite of global non-parallel patterns transcriptome wide, we found significant parallelism in gene expression at the level of individual loci with an over-representation of genes involved in biotic stress response. In addition, we demonstrated significant parallelism in GEI, indicating a shared differential response of the originally foothill versus alpine populations to environmental variation across mountain regions. A fraction of genes showing expression parallelism also encompassed parallel outliers for genomic differentiation, with greater enrichment of such variants in cis-regulatory elements in some mountain regions. In summary, our results suggest frequent evolutionary repeatability in gene expression changes associated with the colonization of a challenging environment that combines constitutive expression differences and plastic interaction with the surrounding environment.

Speciation by triparental hybridization in genus Sorbus (Rosaceae).

Hybridization associated with polyploidization and apomixis is a frequent mechanism of speciation. Sorbus is a genus with ongoing hybridization resulting in a polyploid complex with different parental lineage. Triparens is the smallest hybridogenous subgenus of Sorbus so far known to comprise only two taxa, S. intermedia and S. × liljeforsii that combine the genomes of three taxa (S. aria agg., S. aucuparia and S. torminalis). To elucidate the origins of S. dacica, S. paxiana and S. tauricola, three new trigenomic candidates formerly believed to be of biparental origin with either S. aria agg. × S. aucuparia or S. aria agg. × S. torminalis lineage we combined data from HPLC and chloroplast DNA analysing additional 33 related taxa as well. We concluded that the 'torminalis-type' flavonoid profile and the 'aucuparia-type' plastid indicate the participation of both S. torminalis and S. aucuparia resulting in the formation of S. dacica, S. paxiana and S. tauricola. Sorbus aria agg. as the third ancestor and as a necessary link to meet genes of S. torminalis and S. aucuparia in one genome is obvious from morphological features (densely tomentose undersides of leaves). The tetraploid cytotypes and obligate pseudogamy of S. dacica and S. paxiana were determined by flow cytometry and are published here for the first time. The most probable evolutionary scenario for Triparens species is: 1. a diploid sexual S. aucuparia as pollen acceptor hybridized with a tetraploid apomictic taxon from the S. aria agg. producing a triploid apomictic taxon with 'aucuparia-type' plastid inherited maternally; 2. during a second crossing event this subgenus Soraria hybrid as maternal progenitor hybridized with the sexual diploid S. torminalis (providing gene(s) of apigenin O-glucuronide synthesis) forming a tetraploid Triparens hybrid with 'aucuparia-type' plastid and 'torminalis-type' flavonoids.