For more than a decade, the workhorse of genomic prediction has been a method called Genomic Best Linear Unbiased Prediction, or GBLUP. The approach treats thousands of DNA markers as a single genomic relationship matrix, essentially a map of how genetically similar each breeding line is to every other line, and uses that similarity to estimate the breeding value of untested plants. It is elegant, fast, and interpretable, but it has a fundamental limitation: it is linear. Complex traits such as yield are not governed by genes acting independently. Multiple quantitative trait loci interact with one another through epistatic effects, and the same genotype can excel in one field and disappoint in another, a phenomenon known as genotype-by-environment interaction. When the underlying biology bends and curves in these ways, a straight-line model can leave predictive signal on the table.

The research team tackled this problem by replacing the linear relationship matrix in the mixed-model framework with four alternative kernels: Laplacian, Gaussian, Bessel, and Polynomial. Each kernel is a mathematical function that converts the raw marker profiles of two genotypes into a single similarity score, but they differ in how that similarity decays with genetic distance. The Gaussian kernel, for example, uses a squared Euclidean distance and produces a smooth, bell-shaped decay of similarity, while the Laplacian kernel relies on a Manhattan distance and decays more sharply. Through the so-called kernel trick, these functions implicitly project the marker data into a higher-dimensional space where non-linear relationships between markers, including epistatic interactions of multiple orders, can be captured without ever explicitly estimating every pairwise interaction. Crucially, because the kernels simply replace the covariance structure inside a Bayesian mixed model, the researchers could still estimate variance components and heritability, the quantities that make quantitative genetics useful for breeding decisions.

The evidence base was substantial. The team analyzed the SoyNAM population, a publicly available resource of roughly 5,400 recombinant inbred lines derived from 40 biparental families that each share a common elite parent. From this resource they used a subset of 1,379 genotypes evaluated across six environments in Iowa, Illinois, Indiana, and Nebraska during 2011 and 2012, yielding 8,274 phenotypic records per trait. After filtering, 4,325 single nucleotide polymorphisms remained as predictors. Because successive generations of selfing make these lines nearly homozygous, dominance effects are effectively eliminated, leaving additive and additive-by-additive epistatic effects as the main sources of genetic variation, an ideal setting for testing whether non-linear kernels can detect interaction signal that linear models miss.

Testing was carried out under three cross-validation schemes that mirror real breeding decisions. CV1 asks whether a model can predict genotypes never observed in environments where training data exist, the classic scenario when new lines are developed. CV2 addresses a sparser situation in which genotypes have been tested in only some locations. CV0 is the hardest test of all: predicting how genotypes perform in environments entirely absent from the training set. Predictive accuracy was quantified as the Pearson correlation between observed and predicted values, with results averaged across five independent data splits for the more variable schemes.

The results revealed a striking trait-specific pattern. Grain yield proved the most stubborn target. Under a main-effects-only model, the genetic component explained on average just 17.9 percent of phenotypic variance for yield, with error dominating at 82.1 percent. When genotype-by-environment interaction terms were added, the picture flipped dramatically: the interaction component became the largest contributor at 55.6 percent, and predictive accuracy under CV1 rose from a range of roughly 0.20 to 0.55 up to approximately 0.35 to 0.70. Seed composition traits told a different story. Protein content averaged 32.6 to 34.8 percent across environments and oil content 18.8 to 20.4 percent, and both traits showed much stronger genetic control, with heritability estimates around 0.45 for protein and 0.55 for oil. For these more stable traits, adding interaction terms brought only modest gains, and oil content was where the non-linear kernels showed their clearest advantage over GBLUP, outperforming it across all validation schemes.

The hardest test, CV0, produced one of the study’s most instructive findings. When the target environment was completely unobserved, adding explicit interaction terms provided little benefit, because interaction effects for that environment cannot be learned without phenotypic observations there. Predictions instead leaned on the main genetic effects estimated from other sites. Even so, the non-linear kernels generally outperformed GBLUP under both modeling frameworks in this scenario, suggesting that their richer similarity structure helps transfer information to novel environments. Certain environments proved consistently difficult regardless of method: Illinois 2012 was a persistent weak spot for yield prediction, while Illinois 2011 and Indiana 2012 repeatedly dragged down accuracy for protein and oil, reinforcing the principle that training populations must represent the range of target environments.

Hyperparameter tuning emerged as a decisive but tractable part of the workflow. For the Gaussian kernel, accuracy peaked at the smallest bandwidth tested, sigma equal to 0.0001, while the Laplacian kernel was remarkably insensitive to bandwidth between 0.0001 and 0.01, making it easier to optimize in practice. The Bessel kernel was highly sensitive to its scale parameter, with sigma of 0.1 producing a marked collapse in performance, whereas its degree and order parameters mattered little. The Polynomial kernel showed an almost flat optimization landscape, with degree 2 sufficient to capture the underlying structure. The authors note that these small optimal bandwidths indicate performance was maximized when similarity was concentrated among closely related genotypes, emphasizing local rather than global genetic relationships.

To stress-test the approach under known epistasis, the team turned to a simulated dataset of 1,000 individuals with 4,010 markers, in which six traits were generated under epistatic architectures ranging from 8 to 480 quantitative trait loci and heritabilities from 0.70 down to 0.30. Here the hierarchy of methods depended on genetic architecture. For traits controlled by only a handful of loci, Random Forest, a tree-based machine learning method with 500 decision trees, clearly won, because tree splits excel at capturing strong marker effects and threshold-like responses. But as the number of loci grew and the signal became distributed across the genome, the kernels, particularly Laplacian and Gaussian, became competitive or superior, and under the most polygenic settings all methods converged as broad genomic similarity captured most of the predictable signal. Notably, the Laplacian kernel delivered consistently high accuracy across every architecture tested, positioning it as a promising alternative to the widely adopted Gaussian kernel.

Perhaps the study’s most consequential message is about what breeders gain by staying inside the mixed-model framework. Random Forest was competitive only under the CV0 scheme for protein and oil, and it consistently trailed the other methods for yield. More importantly, machine learning models are primarily predictive black boxes: they do not directly yield variance components, heritability estimates, or the other inferential quantities that guide selection decisions. The non-linear kernel models, by contrast, occupy a practical middle ground, combining the flexibility to capture moderate non-linearities and interaction patterns with the interpretability of quantitative genetics. With no single kernel dominating across all environments, the authors advise that kernel choice should be guided by the characteristics of each dataset rather than by a universal favorite. As breeding programs worldwide race to develop soybeans that yield more under increasingly volatile climates, this work suggests that upgrading the covariance structure, rather than abandoning statistical inference for pure machine learning, may be the smartest path to more accurate and more accountable genomic prediction.

Subject of Research: Non-linear kernel methods for genomic prediction of soybean yield and seed quality traits in multi-environment trials

Article Title: Non-linear kernel methods for genomic prediction of soybean yield and quality in multi-environment trials

Article References: Paiva, W. A. L., da Costa, W. G., Machado, L. P., Nascimento, A. C. C., Jarquin, D., & Nascimento, M. (2026). Non-linear kernel methods for genomic prediction of soybean yield and quality in multi-environment trials. Theoretical and Applied Genetics, 139(10), Article 277. https://doi.org/10.1007/s00122-026-05387-3

Image Credits: AI Generated

DOI: 10.1007/s00122-026-05387-3

Keywords: genomic prediction, soybean, kernel methods, GBLUP, genotype-by-environment interaction, epistasis, SoyNAM population, Random Forest, variance components, heritability, multi-environment trials, plant breeding

The heart of Gaussian process training lies in its hyperparameters, the knobs that define how the model believes data points relate to one another. A kernel function encodes assumptions about smoothness, correlation length, signal strength and noise level, and these quantities are typically learned by maximizing the marginal likelihood, a quantity that balances how well the model fits the data against how complex it becomes. For decades this procedure has been regarded as statistically principled and, under ideal Gaussian assumptions, it genuinely is. What Sedai and Medda show, however, is that the estimators produced by this procedure possess what statisticians call an unbounded influence function, a formal certificate that a single extreme observation can exert arbitrarily large leverage on the fitted kernel parameters.

The influence function is a classic concept from robust statistics, introduced by Frank Hampel in the 1970s as a diagnostic for measuring how much an infinitesimal speck of contamination at any point in the data distribution shifts an estimator. If the influence function is bounded, no single observation can dominate the result no matter how extreme it is. If it is unbounded, a lone outlier can drag the estimate anywhere. The authors adapt this framework to the Gaussian process setting with a subtle twist: because the joint log marginal likelihood couples all observations through the kernel covariance matrix, it does not decompose into a sum of independent per-point terms the way classical M-estimators do. Their solution is a leave-one-in contamination construction, in which the perturbation is represented by the predictive log-likelihood of an added data point conditional on the observed sample.

From this construction they derive a closed-form expression for the influence function of the hyperparameter estimator, factoring it into two interpretable pieces: the global curvature of the objective, captured by the Hessian of the log marginal likelihood, and a local score contribution from the contaminated point. That local score depends on the residual between the contaminated response and the model’s predictive mean, on the predictive variance at the contaminated input, and on where the input sits relative to the observed design. When the input lies far from the training data, the kernel coupling weakens and the geometric influence attenuates, though variance-related parameters such as signal and noise variance remain affected even then.

The central result is stark. Under the standard Gaussian likelihood, the contamination score grows quadratically in the magnitude of the response, so the influence function diverges as the outlier’s value tends toward infinity. The authors prove this formally in a proposition requiring only mild regularity conditions, and they show that the conclusion is automatic whenever the model includes a signal variance or noise variance parameter, which virtually every practical Gaussian process does. They further demonstrate that the result survives log-parameterizations and automatic relevance determination kernels, closing the escape hatches that practitioners might hope would soften the blow. In Hampel’s terminology, maximum likelihood training of Gaussian processes is nonrobust by construction.

Crucially, the paper does not stop at diagnosis. The authors extend the influence function framework to the robust alternatives that have long circulated in the Gaussian process literature, including heavy-tailed likelihoods such as the Student-t and Laplace distributions, and robustified objectives built on the Huber loss and its smooth cousin, the Pseudo-Huber or Charbonnier loss. For each case they establish the conditions under which the influence function becomes bounded. The unifying principle is elegant: if the derivative of the pointwise loss with respect to the mean prediction remains bounded over all residuals, then the contamination score is bounded, and so is the influence function, provided the local curvature of the objective stays finite and nonsingular. This connects Gaussian process training directly to the classical theory of M-estimation, recasting it as a kernelized, probabilistic analogue of robust regression.

The empirical validation is unusually thorough. The authors first verify that their analytic influence function matches an independent automatic-differentiation implementation to machine precision, that it is insensitive to the small ridge regularization used for numerical conditioning, and that it reproduces the true first-order response of the optimizer to small data perturbations as measured by finite differences. On a synthetic one-dimensional benchmark with two hundred points, they then compare Gaussian, Student-t, Huber and Pseudo-Huber training under single-point stress tests and controlled contamination. The scatter of per-point influence norms against whitened residuals tells the story visually: the Gaussian fit reaches large influence values at modest residual sizes, while Huber and Pseudo-Huber accommodate residuals several times larger within a comparable influence range, with Student-t sitting in between.

Perhaps the most surprising empirical finding concerns curvature. Clipping the score function, as Huber does, turns out to be necessary but not sufficient for stable hyperparameter behaviour. Because Huber’s score is flat in the tail, contaminated points contribute no data curvature to the Hessian, and with enough contamination the smallest eigenvalue of the curvature matrix can collapse toward zero or below, destabilizing the local solve and sometimes amplifying influence beyond even the Gaussian fit. The Student-t likelihood, whose score redescends toward zero while retaining curvature, preserved positive curvature in every contaminated run and produced the smallest hyperparameter drift, whereas Huber and Pseudo-Huber lost positive curvature in most runs. The lesson is that robustness depends on both suppressing the direct pull of outliers and maintaining a well-conditioned training objective.

The practical implications ripple outward. In experiments on real datasets including Kin8nm and superconductivity data, the authors show that robust objectives cost essentially nothing when the Gaussian model is appropriate, matching point accuracy and calibrated coverage after a single variance calibration. When contamination or heavy tails appear, however, the Gaussian fit degrades fastest, compensating by inflating its noise estimate and widening its intervals, while Student-t and Pseudo-Huber keep hyperparameters stable and uncertainty honest. The authors also verify that their influence analysis carries over to sparse variational Gaussian processes, the scalable variant used with large datasets, finding that the analytic influence function closely predicts the optimizer’s actual movement.

For a field that increasingly deploys Gaussian processes in safety-critical emulation, environmental modelling and Bayesian optimization, the message is hard to ignore. The fragility of marginal-likelihood training is a structural property of the estimator, not an artefact of finite precision or poor implementation, and it can be measured, predicted and repaired. Sedai and Medda’s work gives practitioners both the warning and the toolkit: check the influence structure of your fitted model, prefer heavy-tailed likelihoods when outliers are plausible, and remember that a bounded score alone will not save you if the curvature of your objective quietly disappears. Robustness, it turns out, is a two-part bargain, and Gaussian process training has only recently learned to sign it.

Subject of Research: Robustness analysis of Gaussian process hyperparameter estimation using influence functions

Article Title: Influence functions for Gaussian process hyperparameter estimation

Article References: Sedai, A., & Medda, F. (2026). Influence functions for Gaussian process hyperparameter estimation. International Journal of Data Science and Analytics, 22(1), Article 316. https://doi.org/10.1007/s41060-026-01285-5

Image Credits: AI Generated

DOI: 10.1007/s41060-026-01285-5

Keywords: Gaussian processes, influence functions, hyperparameter estimation, robust statistics, marginal likelihood, outliers, Student-t likelihood, Huber loss, M-estimation, Bayesian inference, kernel methods, machine learning

The researchers call the underlying problem the Euclidean barrier. In flat space, the volume of a ball grows only polynomially with its radius, meaning that each additional dimension buys a steadily shrinking amount of room. Biological sequences, however, often encode hierarchies that grow exponentially with depth: an evolutionary tree, a taxonomic classification, or a phylogenetic lineage doubles or multiplies its number of branches at every level. When such exponentially expanding structures are squeezed into a fixed number of Euclidean dimensions, distortion is unavoidable. Distances get compressed, subtle relationships between sequences blur, and downstream classifiers inherit the noise. The team’s central insight is that the mismatch is not a flaw in the algorithms but a property of the geometry itself, and that the remedy is to change the stage on which the data performs.

Hyperbolic geometry, the mathematics of spaces with constant negative curvature, has exactly the property the sequences need. The volume of a disk in hyperbolic space grows exponentially with its radius, mirroring the exponential growth of tree-like data. This means a hierarchy can be embedded with far less distortion, with ancestral or central elements placed near the center and more specialized descendants fanning outward toward the boundary. The idea has gained momentum in machine learning in recent years, notably through Poincaré embeddings for hierarchical representations, but its application to the practical problems of biological sequence classification has remained limited. The new work extends this line of research by building a complete pipeline that takes raw sequence representations and transports them into hyperboloid space, where similarity can be measured with a geometry-aware kernel.

The methodology proceeds in well-defined stages. First, biological sequences are transformed into numerical feature representations, preserving the structural information inherent in the original data. Rather than allowing a classifier to operate on these vectors in flat space, the researchers apply a transformation that projects the representations onto the hyperboloid, a model of hyperbolic geometry embedded in a higher-dimensional real space. Once the sequences reside on this curved manifold, the method computes a kernel matrix from the hyperboloid features. Kernels are the mathematical engines behind algorithms such as support vector machines: they encode pairwise similarities between data points without requiring explicit feature engineering. Here, the kernel exploits the hyperbolic dot product of the embedded representations, a quantity that captures how close two sequences are on the curved surface and therefore how similar they are in a structurally meaningful sense.

This hyperboloid-based kernel matrix becomes the foundation for classification. Because the kernel encodes pairwise sequence similarities in a way that respects hierarchical structure, standard learning algorithms fed with this matrix can distinguish sequence families and categories with greater precision than they could using Euclidean distances alone. The approach also promises benefits for exploratory analysis: sequences that share deep evolutionary ancestry cluster together naturally, while divergent lineages separate cleanly, allowing researchers to detect valuable sequence patterns that flat representations tend to smear across artificial boundaries.

Classifying biological sequences is not a niche exercise. It underpins progress in medicine and public health, from identifying SARS-CoV-2 variants and predicting which hosts a coronavirus spike protein can infect, to classifying T-cell receptor sequences for cancer immunotherapy, to recognizing anticancer peptides and predicting protein function from amino acid composition alone. Whole genome sequencing has become central to precision medicine, and the flood of data produced by high-throughput sequencing demands representations that are both faithful and computationally tractable. Every improvement in sequence classification accuracy translates into faster variant tracking during outbreaks, more reliable host-species predictions, and better-informed therapeutic design.

The new study builds on a substantial body of work by the same research group, which has systematically explored non-Euclidean and kernel-based approaches to molecular data. Earlier efforts include Spike2Vec, an efficient embedding method for COVID-19 spike sequences; Reads2Vec, which embeds raw high-throughput sequencing reads; PWM2Vec, which uses position-weight matrices for viral host specification; and Hashing2Vec, a fast embedding generator for SARS-CoV-2 spike classification. The group has also experimented with other curved geometries, publishing a Gaussian Beltrami-Klein model for protein sequence classification and an elliptic geometry-based kernel matrix, along with a Poincaré-distance approach for preserving hidden hierarchical structure in genomic data. The hyperboloid method represents a continuation and refinement of this geometric program, moving from models of hyperbolic space toward a formulation that plugs directly into kernel-based machine learning workflows.

The experimental evaluation reported in the paper suggests that the hyperboloid approach improves the precision of sequence classification relative to conventional Euclidean representations, detecting patterns that traditional machine learning methods miss when working in high-dimensional flat spaces. Traditional classifiers such as support vector machines, naive Bayes, random forests, logistic regression, decision trees, and nearest-neighbor methods all depend on the geometry of the space in which similarities are computed. When that geometry distorts the true relationships among sequences, no amount of tuning at the classifier level can fully recover the lost signal. By repairing the geometry at the representation level, the hyperbolic kernel approach lets well-established classifiers perform closer to their potential.

Importantly, the researchers position their contribution within a broader shift in representation learning for the life sciences. Neural distance embeddings for biological sequences, transfer-learning models trained on protein language, and transformer-based protein encoders have all pushed the field toward richer representations. Hyperbolic methods add a complementary dimension to this progress: rather than learning better features in flat space, they change the space itself. The tradeoffs of hyperbolic embeddings, including the computational cost of operating on curved manifolds and the care needed to optimize within them, remain active research questions, but the evidence that hierarchical biological data benefits from negative curvature continues to accumulate across tasks ranging from multi-label classification to unsupervised representation learning.

The authors note that the datasets used in the study, drawn from resources such as virology databases and public genomic data collections, cannot be shared openly but are available on request, and the code is likewise available on request. As sequencing costs continue to fall and biological datasets grow exponentially in size and complexity, geometric approaches that respect the natural structure of molecular data are likely to move from the mathematical literature into the standard toolkit of bioinformatics. By breaking the Euclidean barrier, this work suggests that the shortest path to understanding biological sequences may not be a straight line at all, but a curve through hyperbolic space.

Subject of Research: Hyperboloid-based representation of biological sequences in hyperbolic geometry for improved machine learning classification

Article Title: Breaking the euclidean barrier: hyperboloid-based biological sequence analysis

Article References: Ali, S., Mansoor, H., & Patterson, M. (2026). Breaking the euclidean barrier: hyperboloid-based biological sequence analysis. Neural Computing and Applications, 38(18), Article 746. https://doi.org/10.1007/s00521-026-12468-3

Image Credits: AI Generated

DOI: 10.1007/s00521-026-12468-3

Keywords: bio-sequence analysis, sequence classification, hyperbolic geometry, hyperboloid space, kernel methods, k-mers, spike sequence, genomic sequence analysis, machine learning, bioinformatics, feature vector representation, hierarchical embeddings