Fred A. van Eeuwijk

The paper develops a hierarchical Bayesian model to analyze dynamic phenotypes in multi-environment trials, quantifying genotype, environment, and G x E effects on growth parameters.

The paper models genotype-specific photosynthetic trait trajectories using functional data analysis to predict biomass, finding that early growth period traits most strongly influence final dry weight.

This paper introduces MagicImpute, a genotype imputation tool for multiparental populations that improves computational efficiency and error handling, showing accuracy across real datasets with missing parents and sequencing errors.

The paper introduces SINN, a structured multimodal deep learning model that improves genomic prediction in new environments by combining statistical decomposition with deep learning, achieving higher accuracy than existing methods.

The paper develops a Physics-Informed Neural Network (PINN) for predicting plant height growth, showing improved accuracy over traditional models with limited data.

The paper identifies organellar genetic variation in Arabidopsis thaliana and shows its impact on photosynthesis through cybrid analysis and high-throughput phenotyping.

The paper identifies a flower bud cooling mechanism that protects tomato pollen during heatwaves and shows that the qPV11 QTL enhances this process, improving pollen viability and fruit set.

This paper introduces a one-stage spatio-temporal P-spline method for analyzing high-throughput phenotyping data, showing it performs similarly to a two-stage approach in genomic selection.

The paper develops a complete chromosome substitution mapping panel in Arabidopsis thaliana to enable high power mapping of genome-wide epistasis.

The paper proposes a combined differential abundance analysis method for sparse microbiome data with group-wise zeros, validated through simulations and applied to plant microbiome datasets.

The paper maps genotype-by-ploidy effects in Arabidopsis thaliana by analyzing haploid and diploid offspring from a late flowering cross, identifying ploidy-specific QTLs and pleiotropic effects.

The paper introduces a two-stage statistical approach to analyze spatio-temporal high-throughput phenotyping data, enabling the estimation of longitudinal genetic signals in plant growth.

This study examines genetic components in diploid hybrid potato using a two-step modelling approach, finding that additive effects dominate over non-additive ones for yield traits.

The paper examines metabolic diversity in Arabidopsis thaliana using metabolomics and GWA mapping, finding that local adaptation and selection shape secondary metabolite profiles through haplotype variation.

This study simulates QTL detection in multi-parent populations of sugar beet, finding that fewer, larger crosses improve detection power, while more parents help detect rare alleles with large effects when population size increases.

The paper introduces methods to analyze multi-parent multi-environment QTL data, showing they better estimate QTL effects across environments and sub-populations than single-environment analyses.

The paper introduces PCgen, an algorithm that reconstructs genetic networks by incorporating direct and indirect effects, improving accuracy using individual data over genotypic means.

This study uses cybrids to show that plasmotype variation affects plant phenotypes through additive and epistatic interactions, with epistasis explaining more variance than additivity.