Biology subjects

Roth, L.

Publications and source records attributed to Roth, L..

At least 19 recordsLinked to original sources

bioRxiv2026

OneGrow: Unified Temporal Plant Image and Mask Generation

Image-based crop phenotyping benefits from image series that capture plant development together with organ-level labels. Such paired data are limited because organ annotation is expensive, and following the same plants over time requires repeated, registered imaging. Existing generative models for plants either synthesize temporal imagery without structural labels or generate labeled images without a temporal dimension. We introduce OneGrow, a latent flow-matching model that jointly models wheat images and their organ-segmentation masks over time. Images and masks share a single frozen image autoencoder. A reveal specifies which content is observed context, so the same model covers tasks such as mask-to-image synthesis, image-to-mask segmentation, and temporal forecasting. For sequences longer than the training window, a sliding-window roll-out generates each new image from the preceding frames, keeping long sequences temporally consistent. We train jointly on a large single-frame wheat dataset and a multi-year temporal dataset, using pseudo-labels from a pretrained segmentation model. We evaluate segmentation and image quality against held-out references and assess the multi-task and temporal behavior qualitatively.

plant biology↗

bioRxiv2026

Neural Processes with Normalizing Flows for Wheat Height Estimation

In this work, we investigate modeling plant traits over time using neural processes, a class of machine learning models that learn distributions over functions. Plant growth is an inherently stochastic process with complex dynamics measured mostly at irregular times throughout the growing seasons. While individual trait trajectories may be simple, their distributions are shaped by complex interactions between genotype, environment, and other factors. In particular, we focus on plant height in wheat, a deceptively simple-looking trait with complex dynamics. To model these trajectory distributions, we evaluate neural processes and in particular extensions using normalizing flows, with different combinations of genotype and environmental covariates. For controlled evaluations, we generate synthetic wheat height trajectories calibrated against Swiss weather station records and the FIP1 dataset. To fully evaluate these trajectory distributions, we use signatures, vector representations of sequential data, together with Sig-MMD and the recently introduced CSig-MMD. Sig-MMD enables direct pathwise comparison of predicted and simulator trajectory distributions, while CSig-MMD focuses this comparison on the tail, including lodged trajectories. Together, these metrics allow us to assess whether the models capture the full distribution of growth trajectories, including rare outcomes.

Plant Biology↗

bioRxiv2026

HAP40 functions as a proteostasis regulator by controlling huntingtin interactions and its release into the extracellular space

Huntingtin-associated protein 40 (HAP40) forms a stable protein complex with huntingtin (HTT). Its cellular function and how HAP40 loss influences mutant HTT (mHTT) abundance and pathobiology are currently unclear. Here, using diverse cellular models and OMICs methods, we demonstrate that HAP40 is an obligate interaction partner of full-length HTT and through its binding controls the abundance of HTT-associated proteins, indicating that it functions as HTT interaction regulatory unit. Also, loss of HAP40 in mHTT-expressing striatal cells impairs autophagosome-lysosome flux, triggers massive transcriptional dysregulation, including the activation of the CLEAR network, demonstrating that it functions as proteostasis regulator that acts on quality control pathways. Finally, mHTT-expressing cells lacking HAP40 showed increased secretion of mHTT through the ER-to-Golgi route, indicating that striatal cells reduce intracellular mHTT-induced proteotoxicity through activation of secretory pathways. Together, these results establish HAP40 as a critical proteostasis regulator that through controlling HTT interactions maintains cellular homeostasis.

cell biology↗

bioRxiv2025

Astrocytic lysosome deficits reduce alpha-synuclein degradation and induce spread of pathology

Parkinsons Disease (PD) is a neurodegenerative disorder caused by the loss of dopaminergic neurons in the substantia nigra due to Lewy body aggregates, primarily composed of misfolded alpha-synuclein (Syn). While PD progression is thought to be driven by a prion-like spread of Syn aggregates between neurons, the role of astrocytes remains unclear. Observations of Syn pathology in PD patient astrocytes suggest their potential involvement in processing aggregates. To investigate this, we studied astrocytes interactions with Syn pre-formed fibrils (PFFs) and their effects in astrocyte-neuron co-cultures on the spread of seed-competent Syn. Primary astrocytes quickly internalized and degraded Syn PFFs. However, degradation was significantly hindered by lysosome-compromising agents like chloroquine, Leupeptin, or CA-074. Adding astrocytes to neuron cultures reduced endogenous Syn aggregation, indicating their role in mitigating Syn pathology. When lysosome efficiency in astrocytes was compromised, their anti-seeding effect diminished. Moreover, lysosome-compromised astrocytes preloaded with Syn PFFs enhanced Syn pathology in neurons, whereas unimpaired astrocytes did not. These findings suggest astrocytes can modulate and contribute to Syn pathology spread, playing a significant role in PD pathogenesis.

neuroscience↗

bioRxiv2025

Tracking Lesion Growth in the Field: Imaging and Deep Learning Reveal Components of Quantitative Resistance

O_LIMeasuring individual components of pathogen reproduction is key to understanding mechanisms underlying rate-reducing quantitative resistance (QR). Simulation models predict that lesion expansion plays a key role in seasonal epidemics of foliar diseases, but measuring lesion growth with sufficient precision and scale to test these predictions under field conditions has remained impractical. C_LIO_LIWe used deep learning-based image analysis to track 6889 individual lesions caused by Zymoseptoria tritici on 14 wheat cultivars across two field seasons, enabling 27,218 precise and objective measurements of lesion growth in the field. C_LIO_LILesion appearance traits reflecting specific interactions between particular host and pathogen genotypes were consistently associated with lesion growth, whereas overall effects of host genotype and environment were modest. Both host cultivar and cultivar-by-environment interaction effects on lesion growth were highly significant and moderately heritable (h2 [≥] 0.40). After excluding a single outlier cultivar, a strong and statistically significant association between lesion growth and overall QR was found. C_LIO_LILesion expansion appears to be an important component of QR to STB in most--but not all--wheat cultivars, underscoring its potential as a selection target. By facilitating the dissection of individual resistance components, our approach can support more targeted, knowledge-based breeding for durable QR. C_LI

plant biology↗

bioRxiv2025

The plant-time-bender model: predicting yield through wheat's perception of time

To address challenges in food security, a better understanding of crop performance under varying and changing environmental conditions is required. Plant Time Warping (PTW) is a deep learning model that integrates high-throughput field phenotyping data with genomic and environmental information to predict wheat yield. PTW leverages image time series, genetic markers, and environmental covariates to learn genotype-specific physiological responses to temperature and vapor pressure deficit. Compared to mere genomic prediction models, PTW demonstrates superior performance when predicting yield in unseen environments across 48 year-locations in Europe. The PTW model captures non-linear growth responses varying with phenological stages and identifies distinct patterns associated with yield performance and stability. Specifically, varieties with higher yield stability exhibit reduced sensitivity to vapor pressure deficit around 1.5 kPa and distinctive temperature responses during emergence and senescence. The learned response pattern enable retrospective and prospective yield predictions, providing a foundation for location-specific variety recommendations and targeted breeding strategies. The integration of phenomic, genomic, and enviromic data has the potential to substantially advance research in climate adaptation strategies for crop production by addressing generalization challenges of predictions to novel environmental conditions. HighlightWe present a novel deep learning model that seamlessly combines high-throughput image data, genomic data, and weather data, enabling better crop predictions for future climates.

genomics↗

bioRxiv2025

Quantifying cancer- and drug-induced changes in Shannon information capacity of RTK signaling

Signaling pathways transmit and process information, enabling cells to respond accurately to external cues. Disease states like cancer can corrupt signal transmission, though the magnitude to which they reduce information capacity has not been quantified. Here we apply pseudo-random pulsatile optogenetic stimulation, live-cell imaging, and information theory to compare the information capacity of receptor tyrosine kinase (RTK) signaling pathways in EML4-ALK-driven lung cancer cells (STE-1) and non-transformed lung epithelial cells (BEAS-2B). The information rate through the RTK/ERK pathway in STE-1 cells was below 0.5 bit/hour but increased to 3 bit/hour after oncogene inhibition. Information was transmitted by only 50-70% of cells, whose channel capacity (maximum information rate) was estimated through in silico protocol optimization. Although oncogene inhibition increased the capacity of the RTK/ERK pathway in STE-1 cells (6 bit/hour), capacity remained lower than in BEAS-2B (11 bit/hour). The capacity of the parallel RTK/calcineurin pathway in BEAS-2B exceeded 15 bit/hour. This study highlights information capacity as a sensitive metric for identifying disease-associated dysfunction and evaluating effects of targeted interventions.

systems biology↗

bioRxiv2025

The Global Wheat Full Semantic Organ Segmentation (GWFSS) dataset

Computer vision is increasingly used in farmers fields and agricultural experiments to quantify important traits. Imaging setups with a sub-millimetre ground sampling distance enable the detection and tracking of plant features, including size, shape, and colour. Although todays AI-driven foundation models segment almost any object in an image, they still fail for complex plant canopies. To improve model performance, the global wheat dataset consortium assembled a diverse set of images from experiments around the globe. After the head detection dataset (GWHD), the new dataset targets a full semantic segmentation (GWFSS) of wheat organs (leaves, stems and spikes) covering all developmental stages. Images were collected by 11 institutions using a wide range of imaging setups. Two datasets are provided: i) a set of 1096 diverse images in which all organs were labelled at the pixel level, and (ii) a dataset of 52,078 images without annotations available for additional training. The labelled set was used to train segmentation models based on DeepLabV3Plus and Segformer. Our Segformer model performed slightly better than DeepLabV3Plus with a mIOU for leaves and spikes of ca. 90%. However, the precision for stems with 54% was rather lower. The major advantages over published models are: i) the exclusion of weeds from the wheat canopy, ii) the detection of all wheat features including necrotic and senescent tissues and its separation from crop residues. This facilitates further development in classifying healthy vs. unhealthy tissue to address the increasing need for accurate quantification of senescence and diseases in wheat canopies.

plant biology↗

bioRxiv2025

Formation of amyloid-like HTTex1 aggregates in neurons, downregulation of synaptic proteins and early mortality of Huntington's disease flies are causally linked

Amyloidogenic mutant huntingtin exon-1 (mHTTex1) protein aggregates with pathogenic polyglutamine (polyQ) tracts are the potential root cause of Huntingtons disease (HD). Here, we assessed the gain-of-function toxicity of mHTTex1 aggregation in neurons of HD transgenic flies. We show that the rate of mHTTex1 aggregation in neurons and early mortality of HD transgenic flies are correlated. We observed sequestration of key synaptic proteins into amyloid-like mHTTex1 aggregates and a concomitant decrease of their transcript levels, suggesting that progressive mHTTex1 aggregate stress in neurons leads to an impairment of synaptic function. Machine learning-based data analysis revealed that the abundance of synaptic proteins such as the vesicular monoamine transporter Vmat in the brain is predictive of fly survival. RNAi knockdown of Vmat-encoding transcripts in neurons with pathogenic amyloid-like HTTex1Q97 aggregates further shortened the lifespan of HD flies, supporting the hypothesis that mHTTex1 aggregation drives impairment of synaptic processes and pathogenesis of HD.

molecular biology↗

bioRxiv2024

The FIP 1.0 Data Set: Highly Resolved Annotated Image Time Series of 4,000 Wheat Plots Grown in Six Years

BackgroundUnderstanding genotype-environment interactions of plants is crucial for crop improvement, yet limited by the scarcity of quality phenotyping data. This data note presents the Field Phenotyping Platform 1.0 data set, a comprehensive resource for winter wheat research that combines imaging, trait, environmental, and genetic data. FindingsWe provide time series data for more than 4,000 wheat plots, including aligned high-resolution image sequences totaling more than 153,000 aligned images across six years. Measurement data for eight key wheat traits is included, namely canopy cover values, plant heights, wheat head counts, senescence ratings, heading date, final plant height, grain yield, and protein content. Genetic marker information and environmental data complement the time series. Data quality is demonstrated through heritability analyses and genomic prediction models, achieving accuracies aligned with previous research. ConclusionsThis extensive data set offers opportunities for advancing crop modeling and phenotyping techniques, enabling researchers to develop novel approaches for understanding genotype-environment interactions, analyzing growth dynamics, and predicting crop performance. By making this resource publicly available, we aim to accelerate research in climate-adaptive agriculture and foster collaboration between plant science and machine learning communities.

plant biology↗

bioRxiv2024

Engineering a novel probiotic toolkit in Escherichia coli Nissle1917 for sensing and mitigating gut inflammatory diseases

Inflammatory Bowel Disease (IBD) is characterized by chronic intestinal inflammation with no cure and limited treatment options that often have systemic side effects. In this study, we developed a target-specific system to potentially treat IBD by engineering the probiotic bacterium Escherichia coli Nissle 1917 (EcN). Our modular system comprises three components: a transcription factor-based sensor (NorR) capable of detecting the inflammation biomarker nitric oxide, a type 1 hemolysin secretion system, and a therapeutic cargo consisting of a library of humanized anti-TNF nanobodies. Despite a reduction in sensitivity, our system demonstrated a concentration-dependent response to nitric oxide, successfully secreting functional nanobodies with binding affinities comparable to the commonly used drug Adalimumab, as confirmed by ELISA and in vitro assays. This newly validated nanobody library expands EcN therapeutic capabilities. The adopted secretion system, also characterized for the first time in EcN, can be further adapted as a platform for screening and purifying proteins of interest. Additionally, we provided a mathematical framework to assess critical parameters in engineering probiotic systems, including the production and diffusion of relevant molecules, bacterial colonization rates, and particle interactions. This integrated approach expands the synthetic biology toolbox for EcN-based therapies, providing novel parts, circuits, and a model for tunable responses at inflammatory hotspots. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=166 HEIGHT=200 SRC="FIGDIR/small/599326v1_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@173d556org.highwire.dtl.DTLVardef@1bb2ae1org.highwire.dtl.DTLVardef@fc1fcborg.highwire.dtl.DTLVardef@8b5f7f_HPS_FORMAT_FIGEXP M_FIG C_FIG Graphical Table of Contents. The engineered probiotic system: Inflamed intestinal cells release the inflammatory regulator TNF (depicted as red squares), which promotes inflammation through a positive feedback loop. Concurrently, these cells produce large amounts of nitric oxide (NO, represented by yellow circles) during inflammation. Our custom-engineered EcN biosensor can detect NO using a NorR-based sensor (in purple) and subsequently trigger the production of nanobodies (in turquoise). These nanobodies are then released into the extracellular environment via a specially engineered secretion system in the bacterial host (shown in dark blue). Once outside the cell, the nanobodies attach to TNF, effectively sequestering them and reducing inflammation. The graph at the bottom of this panel illustrates the general behavior of our system: nanobody production starts upon reaching a certain NO concentration threshold and continues in an NO-dependent fashion. As nanobodies are produced, they capture TNF, leading to a reduction in inflammation and a decrease in NO production. This decrease in NO then halts the nanobody production. SignificanceProbiotics can be engineered to detect and act upon extracellular disease indicators, optimizing therapeutic outcomes. Particularly, self-regulating sense-and-respond genetic circuits have the potential to enhance the accuracy, efficacy, and adaptability of treatment interventions. In this study, we developed and characterized a new integrated and modular toolkit that detects a gut inflammation biomarker, specifically nitric oxide, and responds to it in an inducible manner by secreting humanized nanobodies targeting the pro-inflammatory molecule TNF. We also develop a coarse-grained mathematical framework for modelling engineered probiotic activity in the gut. This novel system contributes to current efforts to develop new engineered probiotic systems and holds promise for inspiring new treatments for gut inflammation associated with various autoimmune diseases.

synthetic biology↗

bioRxiv2024

Mitochondrial reverse electron transport in myeloid cells perpetuates neuroinflammation

Sustained smouldering, or low grade, activation of myeloid cells is a common hallmark of several chronic neurological diseases, including multiple sclerosis (MS)1. Distinct metabolic and mitochondrial features guide the activation and the diverse functional states of myeloid cells2. However, how these metabolic features act to perpetuate neuroinflammation is currently unknown. Using a multiomics approach, we identified a new molecular signature that perpetuates the activation of myeloid cells through mitochondrial complex II (CII) and I (CI) activity driving reverse electron transport (RET) and the production of reactive oxygen species (ROS). Blocking RET in pro-inflammatory myeloid cells protected the central nervous system (CNS) against neurotoxic damage and improved functional outcomes in animal disease models in vivo. Our data show that RET in myeloid cells is a potential new therapeutic target to foster neuroprotection in smouldering inflammatory CNS disorders3.

neuroscience↗

bioRxiv2023

Temporal resolution trumps spectral resolution in UAV-based monitoring of canopy senescence dynamics

Senescence is a dynamic process that is affected by many environmental, genetic, and physiological factors. Quantifying this process is important for breeding wheat varieties with high yield and of high quality. We present a method that allows up-scaling of the state of the art method - visual scoring - by using image sequences acquired from Unmanned Aerial Vehicles (UAV). This reduces measurement time and environmental changes during the measurement as well as rater bias. We compared the potential of a widely used multispectral sensor and a cheaper high-resolution RGB camera to track the dynamics of senescence. A UAV each was equipped with one of these sensors and used to measure canopy reflectance throughout the senescence process that lasted several weeks, for more than 400 winter wheat cultivars across three field seasons. Multiple spectral and RGB indices were calculated at the experimental plot level and used to model the dynamics of senescence. Model fits were further processed to extract key time points of the senescence phase. By comparing the results of the two sensors with each other and with the visual evaluation, respectively, we show that both sensors allow monitoring of senescence dynamics and measure key time points of the phase with a precision close to that of more sophisticated proximal sensing approaches. Optimal timing of measurements proved to be more important than the choice of sensor, confirming that timely and frequent measurements should be prioritized over more expensive sensors that provide a higher spectral resolution.

plant biology↗

bioRxiv2023

Field phenotyping reveals the importance of including cultivar-specific per se temperature response in phenology modeling

Predicting plant development, a longstanding goal in plant physiology, involves two interwoven components: continuous growth and the progression of growth stages (phenology). Current models, like thermal time, assume species-level growth responses to temperature. We challenge this assumption, suggesting that cultivar-specific temperature responses significantly affect phenology. To investigate, we collected field-based growth and phenology data in winter wheat and soybean over multiple years. We used diverse models, from linear to neural networks, to assess growth responses to temperature at various trait and covariate levels. Cultivar-specific non-linear models best explained phenology-related cultivar-environment interactions. With cultivar-specific models, additional relations to other stressors than temperature were found. The availability of the presented field phenotyping tools allows incorporating cultivar-specific temperature response functions in future plant physiology studies, which will deepen our understanding of key factors that influence plant development. Consequently, this work has implications for crop breeding and cultivation under adverse climatic conditions.

physiology↗

bioRxiv2022

Image-based phenomic prediction can provide valuable decision support in wheat breeding

Traditionally, breeders selection decisions in early generations are largely based on visual observations in the field. With the advent of affordable genome sequencing and high-throughput phenotyping technologies, enhancing breeders ratings with such information became attractive. In this research, it is hypothesized that GxE interactions of secondary traits (i.e., growth dynamics traits) are less complex than those of related target traits (e.g., yield). Thus, phenomic selection (PS) may allow selecting for genotypes with beneficial response-pattern in a defined population of environments. A set of 45 winter wheat varieties was grown at five year-sites and analyzed with linear and factor-analytic (FA) mixed models to estimate GxE interactions of secondary and target traits. The dynamic development of drone-derived plant height, leaf area and tiller density estimations was used to estimate the timing of key stages, quantities at defined time points, and temperature dose-response curve parameters. Most of these secondary traits and grain protein content showed little GxE interactions. In contrast, the modeling of GxE for yield required a FA model with two factors. A trained PS model predicted overall yield performance, yield stability and grain protein content with correlations of 0.43, 0.30 and 0.34. While these accuracies are modest and do not outperform well-trained GS models, PS additionally provided insights into the physiological basis of target traits. An ideotype was identified that potentially avoids the negative pleiotropic effects between yield and protein content. Key messageGenotype-by-environment interactions of secondary traits based on high-throughput field phenotyping are less complex than those of target traits, allowing for a phenomic selection in unreplicated early generation trials.

physiology↗

bioRxiv2022

High-throughput field phenotyping reveals that selection in breeding has affected the phenology and temperature response of wheat in the stem elongation phase

Crop breeders increasingly need to mitigate the effects of climate change. Ideally, their selection strategies are based on an understanding of crop responses to environmental covariates such as temperature. In this study, the height of 352 varieties (European and Swiss) was repeatedly measured in multiple years. P-splines were used to model plant height as a function of time, from which the phenology parameters jointing (start) and end of stem elongation were derived. An asymptotic model was used to estimate the base-temperature of growth (Tmin), the steepness of the response (lrc), and the growth at optimum temperature (rmax). Parameter rmax had the largest effect on final height, whereas the temperature-response parameters in the narrow sense (Tmin, lrc) were closely connected to phenology. Final height and rmax decreased from the more continental, eastern European countries towards the more maritime, western countries. For genotypes registered in Great Bitain, Tmin was distinctly lower compared to most other regions. Integrating such analysis of in-season responsiveness to fluctuating environmental conditions in breeding will help to improve the genetic gain for climate adaptation. It can only be achieved based on high-throughput assessment of phenotypes in the field throughout the season. HighlightWheat phenology and environmental responses have been affected by breeders selections since 1970. High throughput field phenotyping methods reveal these developments, allowing to better adapt future varieties to climate change.

physiology↗

bioRxiv2022

Oncogenic protein condensates modulate cell signal perception and drug tolerance

Drug resistance remains a central challenge towards durable cancer therapy, including for cancers driven by the EML4-ALK oncogene. EML4-ALK and related fusion oncogenes form cytoplasmic protein condensates that transmit oncogenic signals through the Ras/Erk pathway. However, whether such condensates play a role in drug response is unclear. Here, we used optogenetics to find that condensates suppress signaling through endogenous RTKs including EGFR. Notably, ALK inhibition hypersensitized RTK signals, which are known to drive resistance. Suppression of RTKs occurred because condensates sequestered downstream adapter proteins that are required for RTK signal transmission. Strikingly, EGFR hypersensitization resulted in rapid and pulsatile Erk signal reactivation, which originated from neighboring apoptotic cells. Paracrine signals promoted survival during ALK inhibition, and blockade of paracrine signals suppressed drug tolerance. Our results uncover a regulatory role for RTK fusion condensates in cancer drug response and demonstrate the potential of optogenetics for uncovering functional biomarkers of cancer cells.

cancer biology↗

bioRxiv2021

Phenomics data processing: Extracting temperature dose-response curves from repeated measurements

Temperature is a main driver of plant growth and development. New phenotyping tools enable quantifying the temperature response of hundreds of genotypes. Yet, for field-derived data, temperature response modeling bears flaws and pitfalls concerning the interpretation of derived parameters. In this study, climate data from five growing seasons with differing temperature distributions served as starting point for a growth simulation of wheat stem elongation, based on a four-parametric temperature response function (Wang-Engel) including all cardinal temperatures. In a novel approach, we re-extracted dose-responses from the simulation by combining high-resolution (hours) temperature courses with low-resolution (days) height data. The collection of such data is common in field phenotyping platforms. To take advantage of the lack of supra-optimal temperatures during the stem elongation, simpler (linear and asymptotic) models to predict temperature-response parameters were investigated. The asymptotic model extracted the base temperature of growth and the maximum absolute growth rate with high precision, whereas simpler, linear models failed to do so. Additionally, the asymptotic model provided a proxy estimate for the optimum temperature. However, when including seasonally changing cardinal temperatures, the prediction accuracy of the asymptotic model was strongly reduced. In a field study with three winter wheat varieties, significant differences were found for all three asymptotic dose-response curve parameters. We conclude that the asymptotic model based on high-resolution temperature courses is suitable to extract meaningful parameters from field-based data.

plant biology↗