Biology subjects

Varela, S.

Publications and source records attributed to Varela, S..

6 recordsLinked to original sources

bioRxiv2026

ROOTQUANT: AUTOMATED ROOT TRAIT QUANTIFICATION FROMMINIRHIZOTRON IMAGES USING DEEP LEARNING

AO_SCPLOWBSTRACTC_SCPLOWQuantifying root traits such as root length (RL) and root surface area (RSA) from minirhizotron imagery is a valuable approach for overcoming the phenotyping bottleneck that limits understanding and improvement of crop productivity, resource use efficiency and resilience in field experiments. However, current approaches remain labor-intensive, and deep learning (DL) methods suffer from limited generalization ability. We present RootQuant, an end-to-end DL model that simultaneously predicts RL and RSA directly from minirhizotron images using only whole-image trait values as supervision, thereby eliminating the need for pixel-level annotations. The models generalization ability was evaluated across species and fine-tuning configurations. The practical applicability of the model was further assessed under field conditions by converting image-derived RL estimates into volumetric root length density (vRLD). Using 118,191 maize and soybean images collected between 2009 and 2020, RootQuant trained on both species achieved an R2 of 0.90 and an RMSE of 2.9 mm for RL, and an R2 of 0.88 and an RMSE of 4.2 mm2 for RSA. The same mixed-species model generalized strongly across species, yielding an 8% relative improvement in R2 and a 30% lower RMSE on maize compared with the same architecture trained on a single species and applied zero-shot. Image-derived RL predictions converted to vRLD showed the expected depth-dependent decline in vRLD, as was also found by coincident destructive quantification of roots washed out of soil cores. By providing a generalist backbone model trained on a large dataset from two major crop species, RootQuant enables high-throughput simultaneous estimation of two relevant root traits directly from raw imagery without task-specific fine-tuning, thereby accelerating in situ root system analysis and phenotyping applications.

plant biology↗

bioRxiv2025

Breaking the barrier of human-annotated training data for machine-learning-aided plant research using aerial imagery

Machine learning (ML) can accelerate biological research. However, the adoption of such tools to facilitate phenotyping based on sensor data has been limited by: (1) the need for a large amount of human-annotated training data for each context in which the tool is used; and (2) phenotypes varying across contexts defined in terms of genetics and environment. This is a major bottleneck because acquiring training data is generally costly and time-consuming. This study demonstrates how a ML approach can address these challenges by minimizing the amount of human supervision needed for tool building. A case study comparing ML approaches that examine images collected by an uncrewed aerial vehicle was performed to determine the presence of panicles (i.e., "heading") across thousands of field plots containing genetically diverse breeding populations of two Miscanthus species. Automated analysis of aerial imagery enabled the identification of heading approximately nine times faster than in-field visual inspection by humans. Leveraging an Efficiently Supervised Generative and Adversarial Network (ESGAN) learning strategy reduced the requirement for human-annotated data by one to two orders of magnitude compared to traditional, fully supervised learning approaches. The ESGAN model learned the salient features of the dataset by using thousands of unlabeled images to inform the discriminative ability of a classifier so that it required minimal human-labeled training data. This method can accelerate the phenotyping of heading date as a measure of flowering time in Miscanthus across diverse contexts (e.g., in multi-state trials) and opens avenues to promote the broad adoption of ML tools. One-sentence summaryMachine learning approach accelerates plant phenotyping by reducing the need for large amounts of human-annotated data, enabling faster and more efficient detection of a plant trait using aerial imagery The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/General-Instructions) is Sebastian Varela.

bioinformatics↗

bioRxiv2024

Optimistic and pessimistic cognitive judgement bias modulates the stress response and cancer progression in zebrafish

Cognitive judgement bias in decision-making under ambiguity occurs both in animals and humans, with some individuals interpreting ambiguous stimulus as positive (optimism) and others as negative (pessimism). We hypothesize that judgement bias is a personality trait and that individuals with a pessimistic bias would be more reactive to stressors and therefore more susceptible to stress-related diseases than optimistic ones. Here, we show that zebrafish judgment bias is a consistent behavioral trait over time, and that pessimistic and optimistic fish express phenotype-specific neurogenomic responses to stress. Furthermore, both phenotypes show differential activation of the hypothalamic-pituitary-interrenal axis in response to chronic stress, suggesting that optimists have a lower stress reactivity. Accordingly, optimists seem to be more resilient to disease than pessimists, as shown by a lower tumorigenesis in a zebrafish melanoma line [Tg(mtifa:HRAS-GFP)]. Together these results indicate that judgement bias is paralleled by differences in the stress response with implications for disease resilience.

animal behavior and cognition↗

bioRxiv2024

Single-nucleus and spatial landscape of the sub-ventricular zone in human glioblastoma

The sub-ventricular zone (SVZ) is the most well-characterized neurogenic area in the mammalian brain. We previously showed that in 65% of patients with glioblastoma (GBM), the SVZ is a reservoir of cancer stem-like cells that contribute to treatment resistance and emergence of recurrence. Here, we built a single-nucleus RNA-sequencing-based microenvironment landscape of the tumor mass (T_Mass) and the SVZ (T_SVZ) of 15 GBM patients and 2 histologically normal SVZ (N_SVZ) samples as controls. We identified a mesenchymal signature in the T_SVZ of GBM patients: tumor cells from the T_SVZ relied on the ZEB1 regulatory network, whereas tumor cells in the T_Mass relied on the TEAD1 regulatory network. Moreover, the T_SVZ microenvironment was predominantly characterized by tumor-supportive microglia, which spatially co-exist and establish heterotypic interactions with tumor cells. Lastly, differential gene expression analyses, predictions of ligand-receptor and incoming/outgoing interactions, and functional assays revealed that the IL-1{beta}/IL-1RAcP and Wnt-5a/Frizzled-3 pathways are therapeutic targets in the T_SVZ microenvironment. Our data provide insights into the biology of the SVZ in GBM patients and identify specific targets of this microenvironment.

cancer biology↗

bioRxiv2023

Early Jurassic origin of avian endothermy and thermophysiological diversity in Dinosauria

A fundamental question in dinosaur evolution is how they adapted to substantial long-term shifts in Earth System during the Mesozoic and when they developed environmentally independent, avian-style acclimatization due to the evolution of an endothermic physiology. Combining fossil occurrences with macroevolutionary and paleoclimatic models, we unveil distinct evolutionary pathways in the main dinosaur lineages: ornithischians and theropods diversified across broader climatic landscapes, trending toward cooler niches. An Early Jurassic shift to colder climates in Theropoda suggests an early adoption of endothermic thermophysiology. Conversely, sauropodomorphs exhibited prolonged climatic conservatism associated with higher thermal conditions. Paleo-biome mapping emphasizes temperature, rather than plant productivity, as the primary driver of this pattern, suggesting poikilothermic physiology with a stronger dependence on higher temperatures in sauropods since the Early Jurassic. One-Sentence SummaryDinosaur climatic evolution reveals early endothermy emergence in theropods, ornithischians but heterotherm sauropodomorphs.

paleontology↗

bioRxiv2023

Bioregionalization: From Wallace and Humboldt to deep-time paleoregion dynamics

Bioregionalization methods allow us to classify and map biogeographic units using data on species composition and traits. Here, we reviewed the evolution of the field during the last 70 years, seeking to summarize its history, and identify gaps and future avenues for research. Our results show that the aim of the studies using bioregionalization methods changed in time. First, bioregionalization were used to unveil the drivers of the observed spatial patterns of biodiversity on Earth, and to understand the role of dispersal limitations on the evolutionary history of clades, but recently, these methods are mostly used for conservation management. Further, data used to map biodiversity regions, the ones that we are now defining conservation strategies, are taxonomically and geographically biased, with a large percentage of the papers using vertebrate data from developed continents/countries. Finally, we show how key papers in the field, the ones with most citations, heavily depend on expert criteria and non-reproducible workflows, preventing direct comparison of maps of bioregions from different papers. Following our findings, we identified 3 gaps for the advance in the field, 1) We need to move beyond maps of vertebrate composition. Ideally, we need to increase the taxonomic diversity of the studies, but also to add other type of information, like data on species traits, genetic diversity, or phylogenetic distances. 2) we need reproducible and standardized methods 3) we need to further explore the temporal dimension of bioregions, to understand how they evolved through time.

ecology↗