Global Genotype by Environment Prediction Competition Reveals That Diverse Modeling Strategies Can Deliver Satisfactory Maize Yield Estimates.

Predicting phenotypes from a combination of genetic and environmental factors is a grand challenge of modern biology. Slight improvements in this area have the potential to save lives, improve food and fuel security, permit better care of the planet, and create other positive outcomes. In 2022 and 2023 the first open-to-the-public Genomes to Fields (G2F) initiative Genotype by Environment (GxE) prediction competition was held using a large dataset including genomic variation, phenotype and weather measurements and field management notes, gathered by the project over nine years. The competition attracted registrants from around the world with representation from academic, government, industry, and non-profit institutions as well as unaffiliated. These participants came from diverse disciplines include plant science, animal science, breeding, statistics, computational biology and others. Some participants had no formal genetics or plant-related training, and some were just beginning their graduate education. The teams applied varied methods and strategies, providing a wealth of modeling knowledge based on a common dataset. The winner's strategy involved two models combining machine learning and traditional breeding tools: one model emphasized environment using features extracted by Random Forest, Ridge Regression and Least-squares, and one focused on genetics. Other high-performing teams' methods included quantitative genetics, classical machine learning/deep learning, mechanistic models, and model ensembles. The dataset factors used, such as genetics; weather; and management data, were also diverse, demonstrating that no single model or strategy is far superior to all others within the context of this competition.

Yield prediction through integration of genetic, environment, and management data through deep learning.

Accurate prediction of the phenotypic outcomes produced by different combinations of genotypes, environments, and management interventions remains a key goal in biology with direct applications to agriculture, research, and conservation. The past decades have seen an expansion of new methods applied toward this goal. Here we predict maize yield using deep neural networks, compare the efficacy of 2 model development methods, and contextualize model performance using conventional linear and machine learning models. We examine the usefulness of incorporating interactions between disparate data types. We find deep learning and best linear unbiased predictor (BLUP) models with interactions had the best overall performance. BLUP models achieved the lowest average error, but deep learning models performed more consistently with similar average error. Optimizing deep neural network submodules for each data type improved model performance relative to optimizing the whole model for all data types at once. Examining the effect of interactions in the best-performing model revealed that including interactions altered the model's sensitivity to weather and management features, including a reduction of the importance scores for timepoints expected to have a limited physiological basis for influencing yield-those at the extreme end of the season, nearly 200 days post planting. Based on these results, deep learning provides a promising avenue for the phenotypic prediction of complex traits in complex environments and a potential mechanism to better understand the influence of environmental and genetic factors.

Timing dependent potentiation and depression of electrical synapses contributes to network stability in the crustacean cardiac ganglion.

Central pattern generators produce many rhythms necessary for survival (e.g., chewing, breathing, locomotion) and doing so often requires coordination of neurons through electrical synapses. Because even neurons of the same type within a network are often differentially tuned, uniformly applied neuromodulators or toxins can result in uncoordinated activity. In the crab (Cancer borealis) cardiac ganglion, potassium channel blockers and serotonin cause increased depolarization of the five electrically coupled motor neurons as well as loss of the normally completely synchronous activity. Given time, compensation occurs that restores excitability and synchrony. One of the underlying mechanisms of this compensation is an increase in coupling among neurons. However, the salient physiological signal that initiates increased coupling has not been determined. Using male C. borealis, we show that it is the loss of synchronous voltage signals between coupled neurons that is at least partly responsible for plasticity in coupling. Shorter offsets in naturalistic activity across a gap junction enhance coupling, while longer delays depress coupling. We also provide evidence as to why a desynchronization-specific potentiation or depression of the synapse could ultimately be adaptive through using a hybrid network created by artificially coupling two cardiac ganglia. Specifically, a stray neuron may be "brought back" in line by increasing coupling if its activity is closer to the remainder of the network. However, if a neuron's activity is far outside network parameters, it is detrimental to increase coupling and therefore depression of the synapse removes a potentially harmful influence on the network.SIGNIFICANCE STATEMENTUnderstanding how neural networks maintain output over years despite environmental and physiological challenges requires understanding the regulatory principles of these networks. Here we study how cells that are synchronously active at baseline respond to becoming desynchronized. In this system, a loss of synchrony causes different parts of the heart to receive uncoordinated stimulation. We find a calcium-dependent control mechanism which alters the strength of electrical connections between motor neurons. While others have described similar control mechanisms, here we demonstrate that voltage changes are sufficient to elicit regulation. Furthermore, we demonstrate that strong connections in a sufficiently perturbed network can prevent any neuron from producing its target activity, thus suggesting why the connections are not constitutively as strong as possible.

Molecular profiling of single neurons of known identity in two ganglia from the crab Cancer borealis.

Understanding circuit organization depends on identification of cell types. Recent advances in transcriptional profiling methods have enabled classification of cell types by their gene expression. While exceptionally powerful and high throughput, the ground-truth validation of these methods is difficult: If cell type is unknown, how does one assess whether a given analysis accurately captures neuronal identity? To shed light on the capabilities and limitations of solely using transcriptional profiling for cell-type classification, we performed 2 forms of transcriptional profiling-RNA-seq and quantitative RT-PCR, in single, unambiguously identified neurons from 2 small crustacean neuronal networks: The stomatogastric and cardiac ganglia. We then combined our knowledge of cell type with unbiased clustering analyses and supervised machine learning to determine how accurately functionally defined neuron types can be classified by expression profile alone. The results demonstrate that expression profile is able to capture neuronal identity most accurately when combined with multimodal information that allows for post hoc grouping, so analysis can proceed from a supervised perspective. Solely unsupervised clustering can lead to misidentification and an inability to distinguish between 2 or more cell types. Therefore, this study supports the general utility of cell identification by transcriptional profiling, but adds a caution: It is difficult or impossible to know under what conditions transcriptional profiling alone is capable of assigning cell identity. Only by combining multiple modalities of information such as physiology, morphology, or innervation target can neuronal identity be unambiguously determined.

Studying gap junctions with PARIS.

A new genetically encoded system manipulates the pH inside cells to detect whether they are coupled to each other.

Dopamine maintains network synchrony via direct modulation of gap junctions in the crustacean cardiac ganglion.

The Large Cell (LC) motor neurons of the crab cardiac ganglion have variable membrane conductance magnitudes even within the same individual, yet produce identical synchronized activity in the intact network. In a previous study we blocked a subset of K+ conductances across LCs, resulting in loss of synchronous activity (Lane et al., 2016). In this study, we hypothesized that this same variability of conductances makes LCs vulnerable to desynchronization during neuromodulation. We exposed the LCs to serotonin (5HT) and dopamine (DA) while recording simultaneously from multiple LCs. Both amines had distinct excitatory effects on LC output, but only 5HT caused desynchronized output. We further determined that DA rapidly increased gap junctional conductance. Co-application of both amines induced 5HT-like output, but waveforms remained synchronized. Furthermore, DA prevented desynchronization induced by the K+ channel blocker tetraethylammonium (TEA), suggesting that dopaminergic modulation of electrical coupling plays a protective role in maintaining network synchrony.

Variability in neural networks.

Experiments on neurons in the heart system of the leech reveal why rhythmic behaviors differ between individuals.

The Hillary Climber trumps manual testing: an automatic system for studying Drosophila climbing.

Climbing or negative geotaxis is an innate behavior of the fruit fly Drosophila melanogaster. There has been considerable interest in using this simple behavior to gain insights into the changes in brain function associated with aging, influence of drugs, mutated genes, and human neurological disorders. At present, most climbing tests are conducted manually and there is a lack of a simple and automatic device for repeatable and quantitative analysis of fly climbing behavior. Here we present an automatic fly climbing system, named the Hillary Climber (after Sir Edmund Hillary), that can replace the human manual tapping of vials with a mechanical tapping mechanism to provide more consistent force and reduce variability between the users and trials. Following tapping the HC records fly climbing, tracks the fly climbing path, and analyzes the velocity of individual flies and the percentage of successful climbers. The system is relatively simple to build, easy to operate, and efficient and reliable for climbing tests.