Explore the vast range of titles available.

Humans can learn several tasks in succession with minimal mutual interference but perform more poorly when trained on multiple tasks at once. The opposite is true for standard deep neural networks. Here, we propose novel computational constraints for artificial neural networks, inspired by earlier work on gating in the primate prefrontal cortex, that capture the cost of interleaved training and allow the network to learn two tasks in sequence without forgetting. We augment standard stochastic gradient descent with two algorithmic motifs, so-called “sluggish” task units and a Hebbian training step that strengthens connections between task units and hidden units that encode task-relevant information. We found that the “sluggish” units introduce a switch-cost during training, which biases representations under interleaved training towards a joint representation that ignores the contextual cue, while the Hebbian step promotes the formation of a gating scheme from task units to the hidden layer that produces orthogonal representations which are perfectly guarded against interference. Validating the model on previously published human behavioural data revealed that it matches performance of participants who had been trained on blocked or interleaved curricula, and that these performance differences were driven by misestimation of the true category boundary.

Dark septate endophytes (DSE) are a form-group of root endophytic fungi with elusive functions. Here, the genomes of two common DSE of semiarid areas, Cadophora sp. and Periconia macrospinosa were sequenced and analyzed with another 32 ascomycetes of different lifestyles. Cadophora sp. (Helotiales) and P. macrospinosa (Pleosporales) have genomes of 70.46 Mb and 54.99 Mb with 22,766 and 18,750 gene models, respectively. The majority of DSE-specific protein clusters lack functional annotation with no similarity to characterized proteins, implying that they have evolved unique genetic innovations. Both DSE possess an expanded number of carbohydrate active enzymes (CAZymes), including plant cell wall degrading enzymes (PCWDEs). Those were similar in three other DSE, and contributed a signal for the separation of root endophytes in principal component analyses of CAZymes, indicating shared genomic traits of DSE fungi. Number of secreted proteases and lipases, aquaporins, and genes linked to melanin synthesis were also relatively high in our fungi. In spite of certain similarities between our two DSE, we observed low levels of convergence in their gene family evolution. This suggests that, despite originating from the same habitat, these two fungi evolved along different evolutionary trajectories and display considerable functional differences within the endophytic lifestyle.

Hyphae represent a hallmark structure of multicellular fungi. The evolutionary origins of hyphae and of the underlying genes are, however, hardly known. By systematically analyzing 72 complete genomes, we here show that hyphae evolved early in fungal evolution probably via diverse genetic changes, including co-option and exaptation of ancient eukaryotic (e.g. phagocytosis-related) genes, the origin of new gene families, gene duplications and alterations of gene structure, among others. Contrary to most multicellular lineages, the origin of filamentous fungi did not correlate with expansions of kinases, receptors or adhesive proteins. Co-option was probably the dominant mechanism for recruiting genes for hypha morphogenesis, while gene duplication was apparently less prevalent, except in transcriptional regulators and cell wall - related genes. We identified 414 novel gene families that show correlated evolution with hyphae and that may have contributed to its evolution. Our results suggest that hyphae represent a unique multicellular organization that evolved by limited fungal-specific innovations and gene duplication but pervasive co-option and modification of ancient eukaryotic functions. Hyphae are a major innovation in fungi associated with transitions to multicellularity. Here, Kiss and colleagues use comparative genomic analyses to reconstruct the evolutionary origins of hyphae and the molecular evolution of hypha morphogenesis genes.

With ∼36,000 described species, Agaricomycetes are among the most successful groups of Fungi. Agaricomycetes display great diversity in fruiting body forms and nutritional modes. Most have pileate-stipitate fruiting bodies (with a cap and stalk), but the group also contains crust-like resupinate fungi, polypores, coral fungi, and gasteroid forms (e.g., puffballs and stinkhorns). Some Agaricomycetes enter into ectomycorrhizal symbioses with plants, while others are decayers (saprotrophs) or pathogens. We constructed a megaphylogeny of 8,400 species and used it to test the following five hypotheses regarding the evolution of morphological and ecological traits in Agaricomycetes and their impact on diversification: 1) resupinate forms are plesiomorphic, 2) pileate-stipitate forms promote diversification, 3) the evolution of gasteroid forms is irreversible, 4) the ectomycorrhizal (ECM) symbiosis promotes diversification, and 5) the evolution of ECM symbiosis is irreversible. The ancestor of Agaricomycetes was a saprotroph with a resupinate fruiting body. There have been 462 transitions in the examined morphologies, including 123 origins of gasteroid forms. Reversals of gasteroid forms are highly unlikely but cannot be rejected. Pileate-stipitate forms are correlated with elevated diversification rates, suggesting that this morphological trait is a key to the success of Agaricomycetes. ECM symbioses have evolved 36 times in Agaricomycetes, with several transformations to parasitism. Across the entire 8,400-species phylogeny, diversification rates of ectomycorrhizal lineages are no greater than those of saprotrophic lineages. However, some ECM lineages have elevated diversification rates compared to their non-ECMsister clades, suggesting that the evolution of symbioses may act as a key innovation at local phylogenetic scales.

The evolution of complex multicellularity has been one of the major transitions in the history of life. In contrast to simple multicellular aggregates of cells, it has evolved only in a handful of lineages, including animals, embryophytes, red and brown algae, and fungi. Despite being a key step toward the evolution of complex organisms, the evolutionary origins and the genetic underpinnings of complex multicellularity are incompletely known. The development of fungal fruiting bodies from a hyphal thallus represents a transition from simple to complex multicellularity that is inducible under laboratory conditions. We constructed a reference atlas of mushroom formation based on developmental transcriptome data of six species and comparisons of >200 whole genomes, to elucidate the core genetic program of complex multicellularity and fruiting body development in mushroom-forming fungi (Agaricomycetes). Nearly 300 conserved gene families and >70 functional groups contained developmentally regulated genes from five to six species, covering functions related to fungal cell wall remodeling, targeted protein degradation, signal transduction, adhesion, and small secreted proteins (including effector-like orphan genes). Several of these families, including F-box proteins, expansin-like proteins, protein kinases, and transcription factors, showed expansions in Agaricomycetes, many of which convergently expanded in multicellular plants and/or animals too, reflecting convergent solutions to genetic hurdles imposed by complex multicellularity among independently evolved lineages. This study provides an entry point to studying mushroom development and complex multicellularity in one of the largest clades of complex eukaryotic organisms.

Rapidly evolving, indel-rich phylogenetic markers play a pivotal role in our understanding of the relationships at multiple levels of the tree of life. There is extensive evidence that indels provide conserved phylogenetic signal, however, the range of phylogenetic depths for which gaps retain tree signal has not been investigated in detail. Here we address this question using the fungal internal transcribed spacer (ITS), which is central in many phylogenetic studies, molecular ecology, detection and identification of pathogenic and non-pathogenic species. ITS is repeatedly criticized for indel-induced alignment problems and the lack of phylogenetic resolution above species level, although these have not been critically investigated. In this study, we examined whether the inclusion of gap characters in the analyses shifts the phylogenetic utility of ITS alignments towards earlier divergences. By re-analyzing 115 published fungal ITS alignments, we found that indels are slightly more conserved than nucleotide substitutions, and when included in phylogenetic analyses, improved the resolution and branch support of phylogenies across an array of taxonomic ranges and extended the resolving power of ITS towards earlier nodes of phylogenetic trees. Our results reconcile previous contradicting evidence for the effects of data exclusion: in the case of more sophisticated indel placement, the exclusion of indel-rich regions from the analyses results in a loss of tree resolution, whereas in the case of simpler alignment methods, the exclusion of gapped sites improves it. Although the empirical datasets do not provide to measure alignment accuracy objectively, our results for the ITS region are consistent with previous simulations studies alignment algorithms. We suggest that sophisticated alignment algorithms and the inclusion of indels make the ITS region and potentially other rapidly evolving indel-rich loci valuable sources of phylogenetic information, which can be exploited at multiple taxonomic levels.

The invasiveness of cancer cells describes the metastasizing capability of a primary tumor. The straightforward detection and quantification of cancer cell invasion are important to predict the survival rate of a cancer patient and to test how anti-cancer compounds influence cancer progression. Digital holographic microscopy based M4 Holomonitor (HM) is a technique that allows the label-free monitoring of cell morphological and kinetical parameters in real-time. Here, a fully confluent epithelial monolayer derived from the African green monkey kidney (Vero) on a gelatin-coated surface was established, then HeLa cells were seeded on top of the monolayer, and their behavior was monitored for 24 h using HM. Several cancer cells showing invasiveness were detected during this period, while other HeLa cells did not show any signs of aggressivity. It was demonstrated that the invasion of single cancer cells is soundly observable and also quantifiable through monitoring parameters such as phase shift, optical volume, area, and motility, which parameters can easily be obtained and processed automatically. Based on the experimental data, the invasion speed of cancer cells entering the epithelial layer can be defined as the shrinking of detected single-cell volume per unit time. The invasion speed and its correlation with cell migration parameters were analyzed in depth. A clear linear relationship between migration and invasion speed was found, cancer cells with stronger migration have slower invasion speed. These results not only describe the effect of how cancer cells invade the underlying monolayer in contrast to non-invasive HeLa cells, but could help in future research to optimize drugs affecting cell invasibility in a fully automated, label-free and high-throughput manner.

In eukaryotes the entry into mitosis is initiated by activation of cyclin-dependent kinases (CDKs), which in turn activate a large number of protein kinases to induce all mitotic processes. The general view is that kinases are active in mitosis and phosphatases turn them off in interphase. Kinases activate each other by cross- and self-phosphorylation, while phosphatases remove these phosphate groups to inactivate kinases. Crucial exceptions to this general rule are the interphase kinase Wee1 and the mitotic phosphatase Cdc25. Together they directly control CDK in an opposite way of the general rule of mitotic phosphorylation and interphase dephosphorylation. Here we investigate why this opposite system emerged and got fixed in almost all eukaryotes. Our results show that this reversed action of a kinase-phosphatase pair, Wee1 and Cdc25, on CDK is particularly suited to establish a stable G2 phase and to add checkpoints to the cell cycle. We show that all these regulators appeared together in LECA (Last Eukaryote Common Ancestor) and co-evolved in eukaryotes, suggesting that this twist in kinase-phosphatase regulation was a crucial step happening at the emergence of eukaryotes.

The fluidic force microscope (FluidFM) can be considered as the nanofluidic extension of the atomic force microscope (AFM). This novel instrument facilitates the experimental procedure and data acquisition of force spectroscopy (FS) and is also used for the determination of single-cell adhesion forces (SCFS) and elasticity. FluidFM uses special probes with an integrated nanochannel inside the cantilevers supported by parallel rows of pillars. However, little is known about how the properties of these hollow cantilevers affect the most important parameters which directly scale the obtained spectroscopic data: the inverse optical lever sensitivity (InvOLS) and the spring constant ( k ). The precise determination of these parameters during calibration is essential in order to gain reliable, comparable and consistent results with SCFS. Demonstrated by our literature survey, the standard error of previously published SCFS results obtained with FluidFM ranges from 11.8% to 50%. The question arises whether this can be accounted for biological diversity or may be the consequence of improper calibration. Thus the aim of our work was to investigate the calibration accuracy of these parameters and their dependence on: (1) the aperture size (2, 4 and 8 µm) of the hollow micropipette type cantilever; (2) the position of the laser spot on the back of the cantilever; (3) the substrate used for calibration (silicon or polystyrene). It was found that both the obtained InvOLS and spring constant values depend significantly on the position of the laser spot. Apart from the theoretically expectable monotonous increase in InvOLS (from the tip to the base of the cantilever, as functions of the laser spot’s position), we discerned a well-defined and reproducible fluctuation, which can be as high as ±30%, regardless of the used aperture size or substrate. The calibration of spring constant also showed an error in the range of −13/+20%, measured at the first 40 µm of the cantilever. Based on our results a calibration strategy is proposed and the optimal laser position which yields the most reliable spring constant values was determined and found to be on the first pair of pillars. Our proposed method helps in reducing the error introduced via improper calibration and thus increases the reliability of subsequent cell adhesion force or elasticity measurements with FluidFM.

Contamination of genomes is an increasingly recognized problem affecting several downstream applications, from comparative evolutionary genomics to metagenomics. Here we introduce ContScout, a precise tool for eliminating foreign sequences from annotated genomes. It achieves high specificity and sensitivity on synthetic benchmark data even when the contaminant is a closely related species, outperforms competing tools, and can distinguish horizontal gene transfer from contamination. A screen of 844 eukaryotic genomes for contamination identified bacteria as the most common source, followed by fungi and plants. Furthermore, we show that contaminants in ancestral genome reconstructions lead to erroneous early origins of genes and inflate gene loss rates, leading to a false notion of complex ancestral genomes. Taken together, we offer here a tool for sensitive removal of foreign proteins, identify and remove contaminants from diverse eukaryotic genomes and evaluate their impact on phylogenomic analyses. It is unclear whether naturally evolved de novo proteins have stable, folded structures. Here, systematic identification and structural modeling of de novo genes, this study reveals that a small subset of these proteins may have well-folded structures, and were likely born with these structures.