Biology subjects

Dolezal, J.

Publications and source records attributed to Dolezal, J..

24 records · Page 2Linked to original sources

bioRxiv2021

Comparative anatomy of leaf petioles in temperate trees and shrubs

Petioles are important plant organs connecting stems with leaf blades and affecting light-harvesting leaf ability as well as transport of water, nutrient and biochemical signals. Despite petioles high diversity in size, shape and anatomical settings, little information is available about their structural adaptations across evolutionary lineages and environmental conditions. To fill our knowledge gap, we investigated the variation of petiole morphology and anatomy in 95 European woody plant species using phylogenetic comparative models. Two major axes of variation were related to leaf area (from large and soft to small and tough leaves), and plant size (from cold-adapted shrubs to warm-adapted tall trees). Larger and softer leaves are found in taller trees of more productive habitats. Their petioles are longer, with a circular outline, thin cuticles without trichomes, and are anatomically characterised by the predominance of sclerenchyma, larger vessels, interfascicular areas with fibers, indistinct phloem rays, and the occurrence of prismatic crystals and druses. In contrast, smaller and tougher leaves are found in shorter trees and shrubs of colder or drier habitats. Their petioles are characterized by teret outline, thick cuticle, simple and non-glandular trichomes, epidermal cells smaller than cortex cells, phloem composed of small cells and radially arranged vessels, fiberless xylem, lamellar collenchyma, acicular crystals and secretory elements. Individual anatomical traits were linked to different internal and external drivers. The petiole length and vessel conduit size increase, while cuticle thickness decreases, with increasing leaf blade area. Epidermis cell walls are thicker in leaves with higher specific leaf area. Collenchyma becomes absent with increasing temperature, epidermis cell size increases with plant height and temperature, and petiole outline becomes polygonal with increasing precipitation. We conclude that species temperature and precipitation optima, plant height, leaf area and thickness exerted a significant control on petiole anatomical and morphological structures not confounded by phylogenetic inertia. Unrelated species with different evolutionary histories but similar thermal and hydrological requirements have converged to similar petiole anatomical structures. Our findings contribute to improving current knowledge about the functional morphoanatomy of the petiole as the key organ that plays a crucial role in the hydraulic pathways in plants.

ecology↗

bioRxiv2021

Distance decay 2.0 - a global synthesis of taxonomic and functional turnover in ecological communities.

Understanding the variation in community composition and species abundances, i.e., {beta}-diversity, is at the heart of community ecology. A common approach to examine {beta}-diversity is to evaluate directional turnover in community composition by measuring the decay in the similarity among pairs of communities along spatial or environmental distances. We provide the first global synthesis of taxonomic and functional distance decay along spatial and environmental distance by analysing 149 datasets comprising different types of organisms and environments. We modelled an exponential distance decay for each dataset using generalized linear models and extracted r2 and slope to analyse the strength and the rate of the decay. We studied whether taxonomic or functional similarity has stronger decay across the spatial and environmental distances. We also unveiled the factors driving the rate of decay across the datasets, including latitude, spatial extent, realm, and organismal features. Taxonomic distance decay was stronger along spatial and environmental distances compared with functional distance decay. The rate of taxonomic spatial distance decay was the fastest in the datasets from mid-latitudes while the rate of functional decay increased with latitude. Overall, datasets covering larger spatial extents showed a lower rate of decay along spatial distances but a higher rate of decay along environmental distances. Marine ecosystems had the slowest rate of decay. This synthesis is an important step towards a more holistic understanding of patterns and drivers of taxonomic and functional {beta}-diversity.

ecology↗

bioRxiv2021

Climate fluctuations drive the recruitment and growth of temperate grassland plants

Recent climate warming is associated with the increasing magnitude and frequency of extreme events, including heatwaves and drought periods worldwide. Such events can have major effects on the species composition of plant communities, hence on biodiversity and ecosystem functioning. Here we studied responses of Central European dry grassland plants to fluctuating temperature and precipitation over the last thirty years with monthly temporal resolution. We assessed the seasonal and annual dynamics of plant recruitment and growth based on the analysis of annual growth rings from the root collar. Although most studies so far applied such methods to trees and shrubs, we focused on typical grassland plants, two forbs and two chamaephytes. We related the recruitment and annual growth to monthly and annual precipitation, temperature and aridity between 1991 and 2019. We revealed species-specific responses, namely the (i) recruitment of deep-rooted, heavy-seeded species was positively affected by precipitation in both late winter-early spring and summer, whereas recruitment of shallow-rooted, light-seeded species was weakly influenced by climate fluctuations; (ii) growth of shallow-rooted species was more adversely affected by high summer temperature and drought than the growth of deep-rooted species. The population age structure of all the studied species was affected by the climate of the past decades. Most individuals established in the wet period of the 2000s, fewer in the precipitation-poorer 1990s, and the establishment was considerably reduced in the dry and warm period of the 2010s. Our findings indicate that the change towards warmer and drier climate has a profound effect even on drought-adapted ecosystems such as temperate dry grasslands. However, plant responses to various climatic extremes are species-specific, depending on their characteristics, such as life form or rooting depth. Consequently, the ongoing and anticipated climate warming will likely result in complex changes in species composition and other ecosystem properties of temperate grasslands

ecology↗

bioRxiv2020

The Impact of Digital Histopathology Batch Effect on Deep Learning Model Accuracy and Bias

The Cancer Genome Atlas (TCGA) is one of the largest biorepositories of digital histology. Deep learning (DL) models have been trained on TCGA to predict numerous features directly from histology, including survival, gene expression patterns, and driver mutations. However, we demonstrate that these features vary substantially across tissue submitting sites in TCGA for over 3,000 patients with six cancer subtypes. Additionally, we show that histologic image differences between submitting sites can easily be identified with DL. This site detection remains possible despite commonly used color normalization and augmentation methods, and we quantify the digital image characteristics constituting this histologic batch effect. As an example, we show that patient ethnicity within the TCGA breast cancer cohort can be inferred from histology due to site-level batch effect, which must be accounted for to ensure equitable application of DL. Batch effect also leads to overoptimistic estimates of model performance, and we propose a quadratic programming method to guide validation that abrogates this bias.

bioinformatics↗

bioRxiv2020

Does rainforest biodiversity stand on the shoulders of giants? Effect of disturbances by forest elephants on trees and insects on Mount Cameroon

Natural disturbances are essential for dynamics of tropical rainforests, contributing to their tremendous biodiversity. In the Afrotropical rainforests, megaherbivores have played a key role before their recent decline. Although the influence of savanna elephants on ecosystems has been documented, their close relatives, forest elephants, remain poorly studied. Few decades ago, in the unique natural enclosure experiment on Mount Cameroon, West/Central Africa, the rainforests were divided by lava flows which are not crossed by the local population of forest elephants. We assessed communities of trees, butterflies and two ecological guilds of moths in disturbed and undisturbed forests split by the longest lava flow at upland and montane elevations. Altogether, we surveyed 32 forest plots resulting in records of 2,025 trees of 97 species, and 7,853 butterflies and moths of 437 species. The disturbed forests differed in reduced tree density, height, and high canopy cover, and in increased DBH. Forest elephants also decreased tree species richness and altered their composition, probably by selective browsing and foraging. The elephant disturbance also increased species richness of butterflies and had various effects on species richness and composition of all studied insect groups. These changes were most probably caused by disturbance-driven alterations of (micro)habitats and species composition of trees. Moreover, the abandonment of forests by elephants led to local declines of range-restricted butterflies. Therefore, the current appalling decline of forest elephant populations across the Afrotropics most probably causes important changes in rainforest biodiversity and should be reflected by regional conservation authorities.

ecology↗

bioRxiv2019

Deep learning detects virus presence in cancer histology

Oncogenic viruses like human papilloma virus (HPV) or Epstein Barr virus (EBV) are a major cause of human cancer. Viral oncogenesis has a direct impact on treatment decisions because virus-associated tumors can demand a lower intensity of chemotherapy and radiation or can be more susceptible to immune check-point inhibition. However, molecular tests for HPV and EBV are not ubiquitously available.\n\nWe hypothesized that the histopathological features of virus-driven and non-virus driven cancers are sufficiently different to be detectable by artificial intelligence (AI) through deep learning-based analysis of images from routine hematoxylin and eosin (HE) stained slides. We show that deep transfer learning can predict presence of HPV in head and neck cancer with a patient-level 3-fold cross validated area-under-the-curve (AUC) of 0.89 [0.82; 0.94]. The same workflow was used for Epstein-Barr virus (EBV) driven gastric cancer achieving a cross-validated AUC of 0.80 [0.70; 0.92] and a similar performance in external validation sets. Reverse-engineering our deep neural networks, we show that the key morphological features can be made understandable to humans.\n\nThis workflow could enable a fast and low-cost method to identify virus-induced cancer in clinical trials or clinical routine. At the same time, our approach for feature visualization allows pathologists to look into the black box of deep learning, enabling them to check the plausibility of computer-based image classification.

cancer biology↗