Frontiers in Bioinformatics

Background: Machine Learning (ML) has shown promise in histopathological diagnosis, but rare tumours, like duodenal epithelial neoplasms, are challenging because large, labelled datasets are difficult to assemble. We propose a data-efficient ML pipeline for detecting duodenal epithelial neoplasia in whole-slide images. Methods: We assembled a five-hospital dataset of haematoxylin and eosin-stained duodenal biopsy whole slide images (WSIs) at 10x magnification, comprising 8,027 normal and 401 neoplastic biopsies, including adenomas, carcinomas and neuroendocrine tumours. WSIs were divided into patches and encoded using the Hibou-B foundation model. We compared global standardisation with unsupervised site-specific standardisation of patch embeddings. Our pipeline combined principal component analysis, a Gaussian mixture model and a Random Forest classifier. We assessed generalisability using leave-one-hospital-out cross-validation across four hospitals and on an independent test set from a fifth hospital. Results: In leave-one-hospital-out cross-validation, the pipeline achieved a mean accuracy of 90%, sensitivity of 96%, specificity of 83% and an AUC of 0.96. On the external test hospital, it achieved 94% accuracy, 93% sensitivity, 95% specificity, and AUC of 0.97. Site-specific standardisation improved mean cross-validation accuracy by 20 percentage points compared with global standardisation, while the Gaussian mixture pipeline improved accuracy by 12 percentage points compared with mean pooling of patch embeddings. Conclusions: We present a data-efficient ML approach for detecting duodenal epithelial neoplasia from WSIs. Strong performance was maintained across hospitals and on an independent external test cohort despite the limited number of neoplastic training cases, supporting the feasibility of foundation-model-based approaches for rare histopathological diagnoses.

Rapid and accurate identification of viral sequences underpins clinical diagnostics, epidemiological surveillance, and the safety testing of biological products, yet the established alignment-based methods such as BLAST are inherently closed-set, with accuracy tied to how well a query is already represented in the reference database. Recent advances in genomic foundation models (GFMs) have enabled alignment-free approaches to sequence classification. However, their performance relative to traditional methods remains incompletely characterised. We evaluated GFMs, specifically the Nucleotide Transformer (NT) and DNABERT families, for viral sequence classification using a dataset derived from the Reference Viral Database (RVDB). Sequence embeddings were generated with mean, max, and CLS pooling, classified by nearest-neighbour search in an embedding index and compared with BLAST as a reference. Matching BLAST provided a particularly stringent benchmark here, because as a closed-set method it compares each query directly against the exact reference sequences it stores, whereas a foundation model must rely on representations shaped by broad, general-purpose pre-training. On sequences of standard length, the best model, NTv3-650M, reached an accuracy of 0.93, approaching the 0.97 achieved by BLAST. On long genomic sequences, evaluated on a much smaller test set, the difference narrowed further, with NTv3-650M reaching 0.95 against 0.98 for BLAST. Robustness analyses showed that all GFMs were markedly more sensitive to base masking than BLAST. Both methods indexed the same labelled reference sequences, and BLAST compared queries with them directly at the nucleotide level, so approaching it from frozen, general-purpose representations without any parameter update is a meaningful result. NTv3-650M emerged as the most promising candidate for further optimisation and deployment in bioinformatics applications.

A critical step in the emergence of a virus is the ability of the viral protein to bind a host receptor and mediate cell entry. For many coronaviruses, this interaction occurs between the Spike S1 subunit and the human ACE2 receptor. Whether this binding interface can be computationally distinguished across unstudied viruses without experimentally resolved protein structures remains an open question. We predicted how 28 emerging coronaviruses may bind to human ACE2 using structural predictions, static interaction prediction programs, and molecular dynamics simulations. To screen the emerging coronaviruses, we predicted a library of S1 structures using AlphaFold. These predicted structures were then used to model the S1-ACE2 interaction with AlphaFold, ClusPro, and HADDOCK. We used known ACE2-binding sarbecoviruses as positive controls and coronaviruses that bind other receptors as negative controls to threshold predicted binding. Contact analysis quantified the predicted binding and revealed that these static interaction prediction methods varied in discriminative power. Less restrained static predictions separated binders from non-binders, whereas heavily restrained docking did not, potentially forcing an interaction where none should exist. This analysis highlighted an emerging coronavirus, Zhejiang2013, as a potential ACE2 binder. We used molecular dynamics simulations to further assess the static predictions and model the interaction over time. Overall, our results indicate that Zhejiang2013 exhibits dynamic interaction patterns consistent with ACE2 binding. Given that two ACE2-binding coronaviruses have caused global pandemics within the past two decades, identifying potential ACE2 binders is critical for early warning and pandemic preparedness.

Accurate classification of lung adenocarcinoma growth patterns from hematoxylin and eosin (H\&E)-stained whole-slide images is essential for treatment planning and prognosis. We develop and evaluate an exponentially weighted ensemble deep learning approach that combines five architectures (EfficientNet-B3, DeiT3, Swin Transformer, Vision Transformer, ConvNeXt) with weights determined by a softmax transformation of per-model Quadratic Weighted Kappa (QWK) scores. Using 25,545 image patches at 40x magnification from the Dartmouth-Hitchcock dataset, we evaluated performance with stratified 5-fold cross-validation. Individual model performance ranged from Cohen's kappa kappa = 0.5563 (ConvNeXt) to kappa = 0.9561 (EfficientNet-B3). The ensemble achieved 96.52% accuracy and kappa = 0.9648, exceeding the best individual model by delta kappa = +0.0087 with statistical significance (McNemar's test, chi^2 = 50.84, p < 10^-12; bootstrap 95% CI on delta kappa [+0.0056, +0.0117]. Pairwise disagreement rates of 16.6--43.2% between architectures indicate that the models capture different aspects of tissue morphology, and the framework automatically down-weights unreliable components, as illustrated by ConvNeXt receiving only 5.7% ensemble weight despite a single-fold training failure. The approach offers a clinically grounded method for ensemble aggregation in digital pathology.

Quantitative characterization of cell proliferation is central to preclinical drug discovery. Here, we evaluated Random Forest (RF) regression for predicting confluence-based cell growth trends using data from human cancer cell lines and benchmarked its performance against the widely used logistic and Gompertz models. The RF model achieved higher predictive accuracy within the evaluated dataset, particularly at higher initial seeding densities. We further extended the RF framework to predict growth curves across seeding densities. Collectively, these findings highlight the potential of data-driven modeling to complement conventional mathematical approaches and support more informed cell-culture design.

This paper predicts antimicrobial resistance (AMR) from matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra using DRIAMS, evaluated across 13 species-antibiotic datasets spanning sensitive/resistant imbalance ratios of 2.95:1 to 141.21:1 and raw spectral similarities of 0.9166-0.9946. We propose HetSAGE, a heterogeneous graph neural network (GNN) that combines full-spectrum "raw" edges with consensus "biomarker" edges derived from a 4-method feature-selection vote and applies per-edge-type dropout to regularize the two views separately. HetSAGE outperforms a plain multilayer perceptron (MLP) only when spectral similarity is high, consistent with oversmoothing effects reported in the broader GNN literature; the biomarker view converges on features close to independently established clinical markers: within {+/-}1 Da in Staphylococcus aureus, Oxacillin, and within range in Klebsiella pneumoniae, Meropenem, confirming it is not a black box. At the same time, per-edge-type dropout protects this smaller signal from dilution by regularization tuned for the noisier raw spectrum. Together, these results indicate that heterogeneous graph structure is not a universal win for MALDI-TOF AMR prediction but a conditional one, tied to a measurable property of the data, spectral similarity, rather than to architecture alone. This caveat is largely unaddressed in current DRIAMS-based GNN work. It provides a concrete signal for practitioners deciding when a graph model is worth the added complexity over a plain MLP.

Curated protein records and their supporting publications organize biological evidence at different levels of detail. Here, we used AI to independently answer a set of biological questions using either UniProt records or the full texts of the literature cited by those records, and then used AI-based evaluation to compare the resulting answers in terms of information coverage and semantic agreement. Fifty human proteins were selected from ten groups in a kinase sequence collection. Target-identity filtering retained 1,749 protein-publication records representing 1,655 distinct publications. For each protein, an AI language model generated one answer set from UniProt and another from the associated literature. A separate AI-based semantic evaluation compared all 2,500 pairwise combinations of the 50 UniProt-derived and 50 literature-derived answer sets, producing 87,500 question-level comparisons. Literature provided answers to 1,211 of 1,750 protein-question combinations, compared with 1,151 for UniProt. After correction for multiple testing across the 35 questions, literature coverage was significantly higher for autoinhibition and inhibitor type. Mean AI-assigned semantic agreement scores were higher for matched protein pairs (0.692) than for unmatched proteins within the same kinase group (0.384) and for proteins from different kinase groups (0.279). The matching literature-derived answer set ranked first for 49 of the 50 UniProt-derived queries, even after questions about gene symbol and kinase family were excluded. The sole exception was GUCY2C, whose UniProt-derived answer set showed slightly greater agreement with the NPR2 literature-derived answer set than with the GUCY2C literature-derived set. TYR proteins had the greatest overall answer coverage, whereas CK1 and RGC proteins had comparatively limited coverage in UniProt and the literature, respectively. Overall, AI analysis showed that UniProt records and their cited publications contain complementary yet strongly protein-specific biological information. Reporting gaps, question scope, and score aggregation influenced apparent agreement, underscoring that the framework evaluates consistency between AI-derived answers, not their biological accuracy.

Recent biological research has shown significant interest and progress towards adoption of agentic AI systems. Domain specific agentic platforms with access to large collections of biological tools with code generation and/or querying capabilities, such as Biomni (illustrative case in point), are representative of the capability to orchestrate automated research workflows. Such platforms enable collaborative AI workspace for drug discovery, genomics, bioinformatics, literature research, and experimental planning for biologists and researchers. When such platforms are used for the aforementioned workflows, individually validating every agentic decision and the research process by which it was reached becomes challenging due to the scale of decisions involved and oftentimes, insufficient disclosure of assumptions and methodological choices, limited traceability of evidence, and interpretive judgments embedded throughout the workflow, giving rise to verification debt. Verification debt in agentic environments could result in unexpected failures and consequences in downstream experiments. This study focuses on empirical qualitative characterization of verification debt in agentic (prompt to outcomes) workflows. Biomni was chosen the testing AI agentic platform in this study mainly due to the descriptive traces it provides to track every agentic step. Verification debt is examined through two complementary angles: the individual agentic decision steps that carry claims and evidence, and the underlying research process by which the agent arrives at and reports them. This study attempts to characterize verification debt with 10 test cases (covering proteomics, genomics and drug discovery) with varying degree of biological nuance across three identical and independent runs. Verification debt is represented through 9 decision-step metrics and 13 research-process metrics, and also propose the VERIFY framework for evaluating scientific verifiability and validity in agentic workflows.

Cell-type marker genes are definitional characteristics of a cell type, and in single-cell RNA sequencing atlases they are commonly nominated through differential expression (DE) testing. The Wilcoxon rank-sum test is the default method for marker gene discovery in Seurat and a recommended method in Scanpy, two widely used computational software packages for scRNA-seq analysis. Structurally, Wilcoxon's standardized statistic grows with sample size at a fixed underlying effect, meaning computed rankings across clusters of unequal size are incomparable. We quantified this effect across three atlas datasets - lung, kidney, and brain. Mean Wilcoxon score scaled with log cluster size at Pearson correlation r = 0.90-0.95 in each organ. In contrast, a classification-based comparator, NS-Forest, showed low correlation with cluster size (r = 0.10-0.28). This DE-cluster size relationship remained essentially unchanged after controlling for classification performance through partial correlation, decayed monotonically under cluster size downsampling, and was reproduced in simulation with marker quality fixed. The Wilcoxon rank-sum test can be transformed into a rank-based effect size by normalizing by the target-background pair count, removing the positive size dependence. However, this effect size nominates an identical set of candidate genes as the original statistic, indicating the cluster size dependence and poor specificity of DE nominated genes are separate problems. Genes nominated under default settings were substantially less specific to their target clusters, with median on-target fractions of 0.204 (lung), 0.166 (kidney), and 0.083 (brain), compared to NS-Forest's marker genes (0.580, 0.687, 0.364), revealing the relatively poor specificity that highly scoring DE genes exhibit.

When an ontology-based biomedical knowledge graph (a KG, containing information about genes, pathways, phenotypes, and molecular interactions) is analyzed using a computational reasoning system, the system can return unwanted generic concepts. Such generic concepts, in the context of a query about potential drug treatments for a disease, might include 'pharmaceutical preparations' or 'antibiotics'. Answers can be filtered based on hand-curated block-lists or ontology Information Content (IC), but to our knowledge, (1) neither filtering method has been validated against expert judgments of generality, and (2) IC is not available for the majority of nodes in the representative KG used in this work. In this study, three domain experts rated representative lists of KG concepts for generality. We built a semantic precision score (SPS) from a concept's name and description alone, using a biomedical language model (BioLORD). For the purpose of filtering generic answers from biomedical query results, we compared (using queries for 200 diseases) the SPS against a curated concept block-list and two IC thresholds. IC agreed moderately with the experts (Spearman's {rho} = 0.59), but correlations were negligible or negative within IC bands. The SPS was well-correlated with the experts ({rho} = 0.76) globally and within SPS bands, and remained informative on concepts without IC ({rho} = 0.59). Hard filtering of query results using the SPS removed 1,502 of 33,161 query answers, nearly three times the block-list's 513, while removing fewer of the query-specific DrugCentral indications. Hard filtering by IC removed either few answers or many approved drugs. We additionally compared the filtering methods' performance on an independent evaluation set of queries for 200 diseases. The SPS filter removed 5,135 of 91,203 answers compared with the block-list's 1,419 and lost none of 1,296 indications compared with 17. Further, SPS filtering removed approved drugs no more often from the half of DrugCentral not used in training than from the half used in training. Semantic meaning captures expert judgments of specificity more faithfully than ontology position, and semantic scoring can replace IC and manually maintained block-lists for filtering drug-repurposing results.

Introduction: Endothelial-mesenchymal transition (EndoMT) is a dynamic endothelial cell-state transition increasingly implicated in fibrosis, vascular dysfunction, inflammation and cancer. Despite expansion of EndoMT-focused transcriptomic research, EndoMT representation remains limited to broader epithelial-mesenchymal transition (EMT) frameworks in standard gene set repositories. Methods: To address the absence of a general-purpose EndoMT gene set for transcriptomic enrichment analysis, we developed an open EndoMT gene set resource and accompanying R package. A literature-informed EndoMT gene set was curated and directionally assigned to EndoMT UP and EndoMT DOWN signatures according to reported or hypothesised expression during partial EndoMT, defined here as an active transitional state rather than a complete mesenchymal endpoint, to improve capture of EndoMT-associated transcriptional changes across different stages of the process. Genes were annotated by marker subtype and implemented as Gene Matrix Transposed files distributed through the R package, together with exported R objects and gene-level metadata. Benchmarking was performed using bulk RNA-seq data sets GSE118446, comprising seven-day IL-1l{beta}/TGF-{beta}2-treated human umbilical vein endothelial cells (HUVECs) and human pulmonary arterial endothelial cells, and GSE342578, comprising three-day TGF-{beta}1/2/3-treated HUVECs; single-cell RNA-seq data set GSE237895, comprising seven-day TGF-{beta}1-treated and SNAI1-overexpressing HUVECs; and endothelial cell and cancer-associated fibroblast populations from human metastatic melanoma single-cell RNA-seq data set GSE72056. Results: In GSE118446, EndoMT UP was positively enriched and EndoMT DOWN negatively enriched in chronically IL-1{beta}/TGF-{beta}2-treated endothelial cells, concordant with experimentally validated EndoMT induction, while partial EndoMT-associated enrichment was detected in single ligand-treatment. In contrast, EndoMT UP was negatively enriched following acute TGF-{beta}1/2/3 treatment, suggesting temporal differences between acute TGF-{beta} stimulation and chronic EndoMT induction. In GSE237895 mesenchymal-like endothelial cells, EndoMT UP was positively enriched and EndoMT DOWN negatively enriched in both TGF-{beta}1-treated and SNAI1-overexpressing models. EndoMT gene sets also discriminated endothelial and mesenchymal/stromal transcriptional states in GSE72056 metastatic melanoma. Marker subtype and Gene Ontology Biological Process analyses provided additional exploratory resolution of EndoMT-associated components beyond MSigDB Hallmark EMT. Discussion: The EndoMT R package provides a dedicated, accessible and interoperable resource for EndoMT-focused transcriptomic enrichment analysis. Benchmarking demonstrates utility for detecting established, partial and divergent EndoMT-associated states, supporting broad applicability across EndoMT biology research.

Public data sharing in immunological and clinical research is essential for transparency, reproducibility, and secondary analysis. However, many datasets in repositories such as ImmPort lack sufficient structured data to enable reuse. ImmPort's Assessments and Lab Tests templates -- which capture critical clinical information -- are often incomplete or omitted due to their complexity and the manual effort required. To address this challenge, an application with a graphical user interface requiring no programming was developed to facilitate population of these templates. The tool streamlines the submission process by reducing time and complexity for data contributors while ensuring template formatting. It integrates ImmPort files, study-specific data files, and a curated data dictionary to automate the mapping of study variables to the required fields within each template. The tool was evaluated on seven ImmPort studies with diverse designs. It successfully generated submission-ready templates that passed ImmPort's internal validation and will be incorporated into the repository. By improving the feasibility and completeness of templates, this approach enhances the quality, findability, and reusability of data within ImmPort.

This work presents DSSNA (Define Spatial Structure of Nucleic Acids) v2026, a free and open-source standalone GROMACS module for the automated analysis of nucleic acid structure based on atomic coordinates. DSSNA reproduces the core functionality of the X3DNA and DSSR approach while extending it to the analysis of molecular dynamics trajectories through integration with the GROMACS software package. The algorithm identifies and classifies structural features, including canonical and noncanonical base pairs, hydrogen-bond networks, stacking interactions, helices, and many other elements of nucleic acid secondary and tertiary structure. Its modular output system enables users to select the structural metrics required for a particular analysis, while the use of neighbor-search algorithms and data-processing capabilities facilitates the analysis of large trajectory datasets. The performance and applicability of DSSNA were demonstrated using a molecular dynamics trajectory of a GUAA tetraloop mutant of the sarcin-ricin domain from Escherichia coli 23S ribosomal RNA (PDB ID: 1MSY). The analysis showed that only a subset of the observed base pairs and stacking interactions remains stable throughout the trajectory. Persistent canonical base pairs formed the structural core responsible for maintaining the secondary-structure profile, whereas most noncanonical base pairs and stacking interactions were transient. The results demonstrate that DSSNA can be used to characterize both static nucleic acid structures and the temporal evolution of structural interactions in molecular dynamics simulations. The work also provides a glossary of terms relevant to the analysis and interpretation of nucleic acid structures.

Single-cell foundation models (scFMs) are commonly benchmarked using final-layer embeddings, although useful depth may depend on the task. We evaluate Geneformer, scBERT, scGPT, and scFoundation as frozen feature extractors for cell-type and perturbation-response prediction using layer-wise probes, leave-one-dataset-out (LODO) selection, readout controls, representation diagnostics, centered kernel alignment (CKA), and retrieval/collapse analyses. Intermediate layers significantly improve cell-type accuracy for scGPT and scBERT in several settings, while Geneformer and scFoundation generally show performance comparable to their final layers. For perturbation prediction, LODO identifies layer 1 as an earliest-on-par candidate across all four models, without establishing universal superiority. Paired scGPT CLS and mean-pooling analyses show that readout affects preferred depth. CKA and representation metrics reveal model- and task-dependent geometric changes, while retrieval and collapse checks support the robustness of perturbation findings. We recommend reporting extraction depth, readout, selection protocol, and task-specific diagnostics. A validated non-final layer may also reduce computation, although efficiency was not directly benchmarked.

Background Multi-omics integration has emerged as a powerful strategy to decode the molecular complexity of biological systems. However, the diversity of available methods, each designed with distinct assumptions, objectives, and computational requirements, makes method selection, usage and interpretation challenging for nonexpert users. Here we present BiomiX 3.0, an updated version of the BiomiX platform that extends its integration capabilities with four additional methods: Similarity Network Fusion (SNF), NEighborhood-based Multi-Omics clustering (NEMO), Data Integration Analysis for Biomarker discovery using Latent variable approaches for Omics studies (DIABLO), and PRAMIGO (Phenotyping netwoRk Application for Multi-omics InteGratiOn), a novel supervised heterogeneous graph transformer (HGT) introduced in this work. Results We benchmarked all five methods, MOFA, DIABLO, SNF, NEMO, and PRAMIGO, on two independent multiomics datasets derived from a Chronic Lymphocytic Leukemia (CLL) cohort comparing IGHV-mutated and unmutated patients, and a pulmonary tuberculosis (PTB) cohort versus healthy controls. Supervised methods (DIABLO, PRAMIGO) consistently achieved higher condition-specific discrimination as measured by the Adjusted Rand Index (ARI) and the Adjusted Mutual Information (AMI). In contrast, unsupervised methods (SNF, NEMO) revealed alternative patient stratifications driven by independent sources of biological variance while MOFA performed in a semi-supervised way occupies an intermediate position, capturing latent factors that explain both disease-associated and orthogonal sources of variance. Systematic gene-centric analysis of the top-ranked features prioritized by each method was supported by manual biological annotation of shared and method-specific signals. Across cohorts, we annotated 154 shared features (116 genes, 38 metabolites) and 60 method-unique features per cohort, demonstrating that no single integration strategy captures the full landscape of biologically relevant signals. In the CLL cohort, shared features spanned B-cell receptor biology, innate immune signaling, RAS/MAPK activation, and epigenetic regulation, while methodunique features revealed supervised-method-specific insights into vesicle trafficking (DIABLO), immune checkpoints (MOFA), and ncRNA regulation (PRAMIGO). In the PTB cohort, a convergent interferon/innate immune signature dominated shared features across all methods. Still, method-unique analysis uncovered DIABLO-specific acylcarnitine metabolic reprogramming, MOFA-specific restoration of lysosomal trafficking, and PRAMIGO-specific {gamma}{delta} T-cell and immunoglobulin repertoire diversity. Across both cohorts, SNF and NEMO proved useful for detecting biological and technical sources of variation that were orthogonal to the primary condition of interest. Ultimately, PRAMIGO uniquely enables the construction of heterogeneous graphs modeling cross-modal molecular interactions, uncovering epigenetic co-regulation programs in CLL and multi-omics inflammatory modules in PTB that are difficult to identify using conventional integration approaches. Conclusions BiomiX 3.0 provides a graphical user interface (GUI) multi-method integration environment that democratizes access to state-of-the-art multi-omics analysis. By combining both unsupervised and supervised integration strategies within a unified platform and introducing graph-based learning through PRAMIGO, BiomiX 3.0 enables researchers across disciplines with complementary tools to interrogate the biological sources of variation in their data, without requiring bioinformatics expertise.

In this work, we focus on predicting fold-switching phenomena observed in certain protein structures. Proteins that undergo fold-switching are called metamorphic proteins. We model fold-switch prediction in proteins as a position/residue-wise classification task. We use various features extracted from a protein's sequence as well as structure to train the classifiers. We propose a novel class of handcrafted parametrised features that encode the structural context of a residue. We also use embeddings from existing protein language models (PLMs) to train the classifiers. We train binary and 3-class classifiers, with the 3-class classifiers providing granular information on the type of fold-switch. In our computations, higher classification performance is achieved for binary prediction as compared to 3-class granular prediction. The handcrafted features perform comparably with the PLM embeddings on our standard test set. A rigorous error analysis by us during the review period revealed a critical data-leakage vulnerability in the metamorphic protein test data. This analysis also revealed that the PLM-based classifier suffers sequence-memorisation bias and fails to generalise to novel fold-switching sequences. On the other hand, one of our domain-informed handcrafted feature based classifier detects fold-switching without relying on sequence memory.

In recent years, the issue of reproducibility in scientific studies has been at the forefront of scientific discussion. Many of these issues arise due to poor employment of statistical methods, making it crucial to address these knowledge gaps for frequently used methods. Elastic net regression (ENET) is one such algorithm used frequently in biology and public health due to its ability to select important predictors (also known as "features") from large data sets. However, ENET struggles with consistent feature selection and hyperparameter choice in these applications due to low observation counts and noisy data. This limitation hinders ENET's effectiveness in biological research due to low model reproducibility and predictor interpretability. We propose the Instability Quotient Criterion (IQC) as a metric to quantify the consistency of selected features for feature selection models such as ENET, with a specific emphasis on biological applications. IQC leverages ensemble modeling and Shannon's entropy in order to determine the stability of selected features derived from ENET models. We demonstrate IQC's robustness via simulation studies, showing that IQC is capable of detecting significant decreases in model stability with lower observation counts, higher dimensionality, and higher proportions of noisy signal. We additionally utilize simulation to provide a naive threshold for low versus high stability. Finally, we applied IQC to a case study to illustrate how our method can be integrated into a larger analysis pipeline when looking at microarray gene expression data. IQC indicated stability issues for differentially-expressed genes when classifying acute leukemia subtypes in spite of high classification success. Lastly, we discuss how model instability impedes interpretation of biologically significant genes for discerning cancer subtypes. Overall, we present IQC as a robust evaluation metric that can easily be incorporated into high-dimensional biological analyses.

Next-generation sequencing data is compositional in nature and ignoring this fact can significantly impact downstream data analyses. This is especially true in co-occurrence (correlation) networks because compositional data is prone to many false negative correlations. The centered log-ration (clr) transformation is a commonly used compositional data analysis (CoDA) approach that converts compositionally constrained data into real number space. The clr transformed values can subsequently be used in machine learning and statistical analyses such as Spearman or Pearson correlations used to generate co-occurrence networks. Here we show that the necessary zero substitution performed prior to clr transformation can create numerous multicollinearity artefacts. We also show that a newer CoDA approach called the robust centered log-ratio (rclr) eliminates these artefacts in correlation-based network analysis.

Background: Conventional discrimination metrics, such as the area under the receiver operating characteristic curve (AUROC), do not characterize the composition of outcomes among patients flagged as high risk by clinical prediction models. This limitation is particularly relevant in critical care, where competing deaths and incomplete follow-up can complicate the interpretation of prediction alerts. We developed the Balanced Risk Integrity Index (BRII), a stratum-level metric defined as the proportion of target events among target events, competing deaths, and early-censored observations in a flagged high-risk stratum. BRII is algebraically related to positive predictive value (PPV) and is reported alongside its component decomposition to characterize the outcome composition of flagged patients. Methods: We evaluated BRII using seven acute kidney injury (AKI) prediction models developed in MIMIC-IV and externally validated in MIMIC-III and eICU-CRD (116,010 patients in total). Additional simulations examined the effects of competing deaths and early censoring on BRII and conventional model performance metrics. Results: For 48-hour AKI prediction in the top 10% risk stratum, BRII ranged from 0.90 to 0.96 in MIMIC-III, compared with less than 0.26 across all models in eICU-CRD, where 61.5% of evaluated patients were censored before the prediction horizon. In the eICU-CRD top 10% strata for the same target, 37.0% to 50.2% of patients were censored early, 12.2% to 15.5% had AKI, and 3.4% to 8.1% died before AKI. BRII and PPV ordered at least one model pair differently in 34 of 48 cohort-outcome-top-k groups. Under simulated random censoring for 48-hour AKI prediction, BRII declined by up to 0.80 while binary-label AUROC changed by less than 0.02. Conclusions: BRII provided complementary information on high-risk stratum composition, yielding different model rankings from PPV and revealing substantial differences across external validation cohorts. Interpreted alongside PPV and its component decomposition, BRII offers a standardized approach to evaluating clinical prediction alerts across models and observation environments.

Several tools are available to detect differential gene expression. The predominant approach of different statistical methods is to consider changes in gene expression independent from their genomic context. Previously, DELocal was introduced that takes into consideration gene expression dynamics in the genomic neighbourhood. Whereas DELocal was shown to identify developmentally regulated genes in developing mouse tooth, its performance in other biological conditions remains undetermined. To benchmark DElocal together with other methods, we develop another tool 'broadSeq'. It provides easy to use interface for methods like DELocal , deseq2, limma, edger, EBSeq and many more. Here we report benchmarking results on different data sets that comprise wild type developing mouse organs from different embryonic time points, and also human cancer data. The examined datasets are from the heart, kidney, liver, brain, forebrain, and hindbrain. We found that most methods produce relatively comparable lists of differentially expressed genes. Nevertheless, DELocal identifies partially distinct sets of genes compared to the other methods. Evaluating the identified genes using gene and disease ontology terms show that the uniquely detected genes by DELocal are biologically relevant and also highly expressed compared to the genes identified by the other methods. Because consistently highly expressed genes show relatively subtle fold changes, the neighbourhood concept of DELocal appears to correct for the difficulty in detecting these differential expression of highly expressed genes. Finally, we provide the tools as combined bioconductor packages, allowing researchers to explore and decide the best methods for their data and questions.