TL;DR
This work proposes post-pretrained lasso selective inference (PPL-SI), a novel selective inference method designed to provide statistically valid p values for the pretrained lasso that reliably controls false discoveries and significantly improves the detection of biologically relevant features compared to traditional approaches.
Learning from high-dimensional, low-sample-size (HDLSS) data remains a persistent challenge in machine learning, as models must infer reliable patterns from limited observations while handling an excessive number of variables—a scenario particularly prevalent in biomedical applications. Such data structures render predictive modeling highly vulnerable to erratic optimization and overfitting. To address this challenge, we propose the Global Interaction Network with Dynamic Generalization Control (GINet-DGC), an artificial intelligence (AI) framework that integrates feature-wise structural priors with dynamic generalization monitoring. Rather than directly learning an unconstrained first-layer weight matrix, GINet-DGC generates task-specific weights from multi-view feature descriptors, encompassing latent semantic, global distributional, local topological, and hierarchical representations. This structure-constrained weight generation strategy effectively narrows the feature-interaction search space and acts as an inductive regularizer against noise and redundant molecular features. Furthermore, we introduce an Overfitting-aware Index (OFI) to monitor the training trajectory and effectively identify the generalization saturation point for adaptive termination. Empirical evaluations on eight public real-world biomedical HDLSS gene-expression datasets, using a repeated stratified 5 × 5 cross-validation protocol, demonstrate that GINet-DGC achieves competitive and stable performance against 17 baselines. These findings support the effectiveness of the proposed framework within the evaluated public biomedical HDLSS benchmark setting.
Overfitting is the excessive adaptation of a machine learning model to its training data and remains a persistent challenge in biomedical informatics. In supervised learning, models may capture patterns “too well,” failing to generalize to unseen data and yielding overly optimistic performance estimates. The problem is especially acute in biomedical settings, where datasets are high-dimensional, heterogeneous, and often of few samples. Numerous strategies have been proposed to mitigate overfitting in bioinformatics and health informatics. However, even established techniques can produce misleading results if misapplied, for example through data leakage, excessive hyperparameter tuning, inappropriate preprocessing, or inadequate validation. To address these pitfalls, we present eleven practical tips for reducing unintentional overfitting in supervised biomedical machine learning studies. The recommendations stress principled data splitting, domain-informed preprocessing, controlled model complexity, systematic tuning, comprehensive performance evaluation, and robustness analysis. Rather than offering an exhaustive treatment, we provide an accessible, practice-oriented guide to support more reliable and reproducible machine learning research. Although developed for biomedical informatics, these quick tips are broadly applicable across disciplines using supervised machine learning.
Survival risk prediction often suffers from challenges such as rare event rates, small effective sample sizes, high-dimensional feature spaces, weak signals, population heterogeneity, and concerns over patient privacy. To overcome these obstacles and improve the precision of prognostic modeling, we introduce the survkl software, which enables the incorporation of external summary-level information with newly collected time-to-event data to support more robust and accurate predictions in survival analysis. Our method adaptively adjusts the weight given to external information, down-weighting heterogeneous information and highlighting more informative ones. The proposed tool accommodates both low-dimensional and high-dimensional data, offering unpenalized estimation and computationally efficient lasso, ridge, and elastic net penalties. The survkl software also provides auxiliary evaluation and plotting functions for model assessment. survkl is freely available to the public at https://github.com/UM-KevinHe/survkl and published under General Public License version 3 license. Supplementary data are available at Bioinformatics Advances online.
The Normalised Sensitivity Ratio (NSR) is introduced, a post-hoc, model-agnostic diagnostic for this question under a structured-shift regime: environments differ primarily in the mean of spurious features while the causal mechanism and causal marginals remain stable, as in multi-site clinical data or multi-batch genomics.
We consider the problem of high-dimensional inference with the lasso estimator. Different methods including'double selection'techniques and multiple versions of the'debiased lasso'have been proposed for this task with noticeable success. However, most guarantees assume strong hypotheses on the underlying data process and the errors in the linear regression model, such as subgaussian designs and independence between errors and the data itself. We show that'standardizing'one's dataset -- a natural procedure in practical penalized regression -- leads to the same results under much weaker hypotheses, paying only a small price for not assuming light tails. The key technical point allowed by this step is exploiting the concentration properties of self-normalized processes. Importantly, we prove our results for two different methods closely related to the'debiased lasso'. The second method performs valid inference even for a misspecified linear model, under mild sparsity conditions similar to the'double selection'literature.
Accurate identification and estimation of variables associated with outcomes or disease states are critical for advancing diagnosis, prognosis, and precision medicine in biomedical research. Regularized regression techniques, such as lasso, are widely employed to enhance interpretability by reducing model complexity and identifying significant variables. However, these methods face two major challenges: (1) the exclusion of important variables due to high correlation with included predictors, and (2) the presence of skewness in human biomedical datasets, which violates key statistical assumptions. Current approaches that fail to address these issues simultaneously may lead to biased interpretations and unreliable coefficient estimates. To overcome these limitations, we propose an enhanced two-step approach, the Bootstrap-Enhanced Regularization Method (BERM). BERM outperformed existing regularization methods in variable selection, achieving the highest overall balanced accuracy while maintaining competitive coefficient estimation performance across a range of simulated sparsity, noise, and dimensionality scenarios. We further demonstrated the effectiveness of BERM by applying it to a human immunophenotyping dataset to identify important immune parameters in the autoimmune disease, type 1 diabetes. BERM is a robust approach for variable selection and coefficient estimation in complex biomedical datasets. Its consistent performance across a wide range of data conditions supports more reliable identification of important variables. An open-source implementation of BERM is available as an R package on GitHub ( https://github.com/xiaorudong/berm ).
