Briefings in Bioinformatics

Motivation: Traditional methods primarily rely on statistical features such as k-mers and GC content, making it difficult to capture complex internal relationships within sequences. Deep learning models typically rely on discrete encodings, leading to issues such as information sparsity, dimensional redun-dancy, and disruption of sequence continuity. Our previous UniWave-1 framework transforms se-quences into waveform signals; however, it lacks comprehensive physicochemical integration and com-putational efficiency. Results: We propose the UniWave-2 feature extraction framework. UniWave-2 incorporates three key advancements: (i) integration of hydrophobicity into the encoding scheme; (ii) enhancement of wave-form resolution through Fourier transform, coupled with wavelet transform to precisely capture both local details and global periodic patterns; and (iii) development of a lightweight multi-scale GRU model that facilitates cross-dimensional feature interactions, captures bidirectional temporal dependencies, and integrates long- and short-range patterns with spatial positional awareness, thereby yielding rich feature representations. UniWave-2 achieved competitive performance across multiple genomic tasks, while ISM analysis demonstrated that waveform encoding enables interpretable attribution without re-lying on gradient information. These results establish UniWave-2 as a lightweight and interpretable framework for nucleic acid sequence analysis.

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.

Over the past decade, aptamers have emerged as promising therapeutics, with cancer therapeutics as a major research area. Existing computational methods, however, either predict aptamer-target pairs or design aptamers against a specific target. In this study, we present AntiCapt, a computational method for identifying single-stranded DNA (ssDNA) anticancer aptamers (ACAs) using sequence descriptors and fine-tuned nucleotide language models. The main dataset comprises 1,021 experimentally validated ACAs and an equal number of non-aptamer sequences. Comparative analysis revealed distinct patterns in the nucleotide, dinucleotide, and trinucleotide compositions associated with ACAs. We developed machine learning (ML) models using composition, autocorrelation, binary profiles and structural features. Among these, the best-performing composition-based model achieved an AUC of 0.93 on an independent dataset, while chemical descriptor-and structural feature-based models achieved a maximum AUC of 0.89. ML models using pretrained and fine-tuned NLM embeddings were also developed, with fine-tuned HyenaDNA embeddings achieving the highest performance, with an independent AUC of 0.94. Furthermore, we developed a model for discriminating anticancer aptamers from general aptamers. The best-performing models were integrated into AntiCapt, a web server and a standalone tool for predicting, designing, and genome-scale scanning anticancer aptamers (https://webs.iiitd.edu.in/raghava/anticapt/).

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.

Interest in bacteriophages as alternative treatments for bacterial infections has grown amid the ongoing crisis of multidrug resistance. However, their narrow host range makes it labor-intensive to identify matching phages for bacterial isolates using conventional experimental methods. Although existing computational models based on biomolecular features have shown promise for phage-host matching, the performance of alternative model settings, including data embeddings, across datasets with different degrees of novelty remains underexplored. This study introduces PhageHost-GNN, a novelty-aware framework using routed models to predict strain-level Klebsiella-phage interactions, evaluated under three clinical scenarios. Using multiple embedding models to encode sequence, structural, and genomic-context features, the framework achieved AUPRCs reaching 0.784, 0.731, and 0.642 in host-unseen, phage-unseen, and both-unseen in silico evaluations, respectively. Benchmarks on experimentally validated datasets showed the clearest advantage over fixed embeddings when both hosts and phages were new to the trained model. In addition to direct inference and de novo training, the framework further implements a fine-tuning pipeline to adapt existing models to local custom datasets while reducing model-training time. Overall, PhageHost-GNN provides a flexible framework for model selection, evaluation, and local adaptation in practical phage-therapy screening.

Integrating heterogeneous molecular and clinical data into unified, clinically meaningful patient representations remains a major challenge in precision medicine. Here, we present SurvMOCA-GNN, a biologically informed multi-omics integration framework that explicitly incorporates biological network structure, cross-omics regulatory relationships, and patient-specific clinical information to generate unified patient representations. Using gene expression, microRNA expression, and clinical data from the TCGA-LAML cohort, SurvMOCA-GNN revealed patient representations that identified prognostically distinct patient groups and further stratified patients within established European LeukemiaNet 2022 risk categories. Systematic ablation analyses demonstrated that incorporation of graph-based modeling, cross-omics attention, adaptive clinical integration, and biological prior knowledge progressively improved integration quality, resulting in more structured patient representations and stronger prognostic stratification. Compared with existing multi-omics integration approaches, SurvMOCA-GNN consistently produced more biologically coherent patient representations with improved survival discrimination. Evaluation in an independent AML cohort, together with transfer learning analyses, further demonstrated the robustness and transferability of the integrated representations across heterogeneous patient populations. Together, these findings demonstrate that biologically informed multi-omics integration reveals clinically meaningful patient representations and provides a general framework for integrating heterogeneous molecular and clinical data to improve patient stratification in AML and potentially other diseases. The SurvMOCA-GNN framework is freely available at https://github.com/tjgu/SurvMOCA-GNN.git.

Accurate molecular property prediction requires effective molecular representations that can describe a molecule from multiple complementary perspectives. Existing deep learning approaches typically use SMILES strings, two-dimensional molecular graphs, or three-dimensional conformations as inputs. These representations capture different aspects of molecular information: SMILES encodes a sequential description, molecular graphs describe atom-bond connectivity, and 3D conformations provide spatial information from atomic coordinates. However, these modalities are often learned separately or merged with low effective fusion operations, which may not sufficiently capture interactions among sequential, topological, and geometric features. In this work, we propose Tri-Modality Cross-Attention (TMCA), a multimodal framework that integrates 1D SMILES, 2D molecular graphs, and 3D molecular conformations for molecular property prediction. TMCA uses pretrained encoders for the 1D and 3D branches, with a SMILES-based Transformer for the 1D branch and a recent conformation-aware pretrained model for the 3D branch. At the same time, a trainable 2D graph encoder is designed to support modality fusion. All the parameters from the pretrained models are frozen to make overall training more efficient. We evaluate TMCA on four datasets (i.e., BBBP, BACE, ClinTox, and HIV) from the MoleculeNet database for classification tasks and compare its performance with state-of-the-art competing approaches. Experimental results show that TMCA achieves the best overall performance, demonstrating the value of multi-modality integration for drug property prediction and the effectiveness of our cross-attention-based fusion strategy. The source code of TMCA is available at: https://github.com/Ay-Zhao/TMCA

Proteins function through dynamic conformational ensembles rather than a single static structure, yet accurately modeling these ensembles remains challenging. Here, we introduce TopoFlow, an evolutionarily conditioned flow-matching framework for protein conformational ensemble generation. TopoFlow uses evolutionary representations from multiple sequence alignments (MSAs) to encode global conformational heterogeneity. It combines this with structural variables from a variational autoencoder (VAE) to capture local flexibility. These complementary representations are then adaptively fused through a conditional modulation module to construct unified conditioning features that guide protein conformational ensemble generation. On the ATLAS molecular dynamics (MD) benchmark, TopoFlow improved the Jensen Shannon metrics for pairwise distances (JS-PwD), radius of gyration (JS-Rg), and time-lagged independent components (JS-TIC) by 10.73%, 5.78% and 7.75%, respectively, relative to BioEmu. Notably, this performance generalizes to experiment-based evaluation, where TopoFlow exhibits strong agreement with experimental observables, including chemical shifts (CS) and small-angle X-ray scattering (SAXS). Ablation analyses further indicated that the evolutionary representations and structural latent variables contributed complementary information to ensemble generation. Our results suggest that TopoFlow, through its integration of evolutionary and structural latent information, offers a promising strategy for generating structurally plausible and conformationally informative protein ensembles.

In the recent growing interest of AI research toward biology, genetic association studies of AI- derived phenotypes from high-content modalities such as images emerges as a powerful means for biological discovery. However, such AI-phenotyping methods still lacks a good optimization target and an efficient evaluation framework. The number of discovered loci was the major criterion for evaluating the quality of AI-derived phenotypes. However, the time and computational resources required for running the GWAS and subsequent loci-clumping are substantial, limiting the rapid development and iteration of deep learning representation algorithms. Here we present a 1000x faster and lightweight framework, sum-h2, than traditional GWAS framework for evaluating genetic discovery through total heritability. We revisit tr(P^(-1) G), previously proposed in the context of evolutionary studies, as a measure of multi-phenotype heritability. We showed that sum-h2, the sum of heritability over phenotypic Principal Components (PCs), is equivalent to the linear transformation-invariant tr(P^(-1) G), through both theoretical proof and simulation studies. Moreover, sum-h2 can be estimated rapidly with minimal information loss over a relatedness-enriched sample, while preserving the relative ranking of endophenotypes by GWAS loci counts. Based on selected UKB data and sum-h2, we set up a Genetic Discovery Arena, enabling rapid and fair comparisons for the development and optimization of deep learning-derived phenotyping methods.

Multi-task bioactivity prediction transfers information across assays, but experimental prioritization also requires uncertainty estimates that identify unreliable predictions. We benchmarked prediction accuracy and uncertainty calibration for five multi-task predictors on 100 ChEMBL27 test assays and 100 CHEMBL37 assays under the random split and the clustering split. Calibration used native, Deep Ensemble and MC Dropout uncertainty; risk ranking was compared for two Gaussian process (GP) backbones. Adaptive deep kernel fitting (ADKF) achieved the best mean calibration results, while deep kernel transfer (DKT) ranked errors more effectively than ADKF with native uncertainty. We also introduce Influence Calibrated Support Reconstruction (ICSR), a risk score for kernel-based predictors. ICSR combines measured support reconstruction errors with query-specific influence and stabilizes their weighted average toward the full-support mean to adjust Gaussian process uncertainty while preserving predicted activities. With DKT and ADKF, ICSR improved all three mean risk-ranking metrics over native uncertainty and an adapted neighborhood comparator in every panel and split. ICSR achieved a mean half-query MAE reduction (R50) of 14.2-19.6%, where R50 measures the percentage decrease in mean absolute error (MAE) after retaining the lowest-risk half of the queries. A retrospectively selected SARS-CoV-2 main protease case further illustrated its use for selecting more reliable predictions. The benchmark supports joint assessment of calibration and error ranking, while ICSR improves selective use of kernel-based bioactivity predictions.

Preprocessing of spatial transcriptomics (ST) count matrices is critical for removing technical variation while preserving biological signals, but optimal strategies remain unclear across diverse platforms. We systematically benchmarked 371 preprocessing workflows across 45 data spanning 11 mainstream ST platforms, evaluating their performance using specifically designed complementary metrics. To further assess preprocessing factors upstream of count-matrix generation, we additionally evaluated image preprocessing, including cell segmentation and post-segmentation transcript processing, across representative imaging-based ST platforms. We found that no universal count-matrix preprocessing workflow performed optimally across all platforms, while data from the same platforms exhibited striking consistency in optimal count-matrix preprocessing workflows. We identified the key data structure characteristics and count-matrix preprocessing steps that critically influence preprocessing performance. We developed platform-specific and data structure-guided recommendations to assist users in selecting the optimal count-matrix preprocessing workflow, validated across a broader range of downstream tasks. We demonstrated that our data-driven recommendations uncover subtle biologically meaningful spatial patterns in new data that were obscured by suboptimal count-matrix preprocessing. Our findings suggest best practices for ST data count-matrix preprocessing.

Drug repurposing and target discovery offer critical strategies for advancing therapeutic development by uncovering the potential biological pathways and novel associations among drugs, genes, and diseases. However, experimental discovery remains expensive and time-consuming, which limits the scalability of large-scale studies. In addition, existing computational approaches often struggle to effectively integrate heterogeneous biomedical data, capture the complex higher-order topological signatures of biological interactomes, and generalize to unseen entities. Here, we present HANAMI (Heterogeneous grAph coNtrastive leArning for drug-gene-disease Motif predIction), a multi-view deep graph learning framework designed to model complex interactions among drugs, genes, and diseases. HANAMI integrates diverse heterogeneous biomedical knowledge, including chemical structures, genomic sequences, and clinical phenotypes, and leverages relation-aware topology encoding, structure-aware aggregation, and contrastive learning to enable accurate motif prediction with biological context from the network. Systematic evaluation on benchmark datasets shows that HANAMI achieves up to 6% improvements over existing state-of-the-art methods in predicting drug-gene-disease motifs. The framework further demonstrates strong inductive generalization, maintaining an [~]18% performance advantage in zero-shot settings involving previously unseen entities. Beyond predictive performance, HANAMI effectively prioritizes drug-disease relationships investigated in Phase II or III trials while identifying candidate genes that suggest plausible mechanistic links. Together, HANAMI provides a computational framework for interpreting complex biomedical interactions, offering a scalable foundation to accelerate drug repurposing and therapeutic innovation.

Single-cell and spatial multiomics technologies are transforming disease atlas construction, but computational integration of unpaired modalities remains challenging, as existing methods overlook two biological constraints of data with matching samples. First, cells should only be mapped across modalities within the same donor or biospecimen, and second, cell recovery frequently differs substantially between modalities. To address these gaps, we present AtlasOT, an optimal-transport framework based on the fused unbalanced Gromov-Wasserstein (FUGW) formulation for multi-modal integration. AtlasOT jointly models a shared cross-modality feature space and modality-specific geometric structures while restricting transport to within-sample cell pairs. Moreover, AtlasOT relaxes strict mass conservation of the balanced optimal transport optimization to accommodate unbalanced cell numbers. We benchmark AtlasOT against state-of-the-art methods on scRNA-scATAC and scRNA-spatial transcriptomics mapping scenarios. Our results indicate that AtlasOT outperforms baselines and state-of-the-art methods in all considered scenarios. Moreover, we demonstrate that AtlasOT's transport plan can be used to improve several relevant tasks such as the detection of rare cell populations, spot-level cell-type deconvolution, spatial gene imputation, and reconstruction of transcription-factor-driven spatial regulatory dynamics. These results establish AtlasOT as a unified, biologically constrained framework for multimodal integration in disease atlas studies, with broad applicability to label transfer, imputation, and regulatory network analysis.

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.

Peptide hormones are important signaling molecules that regulate diverse physiological processes and have substantial therapeutic relevance. However, their experimental identification can be challenging because of low abundance, limited stability, and complex post-translational processing, highlighting the need for reliable computational approaches. Here, we present StackHPpred, a stacking-based ensemble framework for accurate identification of hormone peptides (HPs) from amino acid sequences. We systematically evaluated 56 feature representations, comprising 35 conventional sequence descriptors and 21 PLM/NLP-based embeddings, across 10 machine-learning classifiers, generating 560 baseline models. Prediction-level correlation analysis was subsequently employed to identify 14 complementary, non-redundant feature representations, which were integrated through feature- and classifier-level stacking strategies using out-of-fold predicted probabilities. The final StackHPpred model, based on the top three feature representations CTDD, KSCTriad, and PTAB, achieved an AUC of 0.9965, MCC of 0.9506, ACC of 0.9752, and F1-score of 0.9749 on the independent dataset. StackHPpred was further evaluated under increasingly imbalanced independent datasets with positive-to-negative ratios ranging from 1:10 to 1:50, demonstrating robust predictive performance under challenging screening conditions. t-SNE visualization and SHAP analysis further showed improved class separation and highlighted the contributions of multiple classifier-feature combinations to the final prediction. Overall, StackHPpred provides a robust and interpretable framework for distinguishing HPs from non-HPs and is freely available as a web server at https://jayasreekirthipati.org/StackHPpred/, with a standalone implementation available at https://github.com/JayasreeKirthipati04/StackHPpred.

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.

Protein complex structure prediction depends on both the breadth of candidate coverage and the ability to select accurate models within a limited output budget. Most existing approaches emphasize only one stage. End-to-end models and molecular docking workflows generate candidate structures, whereas quality-assessment models primarily rerank a predefined candidate pool. A complete strategy must account for complementary sampling across generators, differences among scoring scales and the allocation of final candidate quotas. Here, EnsPlex is presented as a multi-source framework that couples structural sampling with candidate selection for protein-protein interaction complex prediction. EnsPlex combines conformation-expanded docking, AlphaFold-Multimer, AlphaFold3 and Boltz-1 to expand the sampled conformational space. Conformation-expanded docking comprises monomer conformational expansion followed by flexible HADDOCK docking. FACET is trained to predict candidate quality using DockQ and its component metrics as supervision. Across 102 antigen-antibody systems, EnsPlex achieved up to a 27.8% relative improvement in target success rate over AlphaFold3 with its built-in ranking under matched output budgets. FACET also improved within-source ranking in the internal candidate pools and in homology-filtered external data. These findings support the utility of EnsPlex for structure prediction in the evaluated antigen-antibody systems.

Biomarker discovery from high-dimensional RNA sequencing data remains challenging. Conventional methods such as differential expression analysis, pairwise protein-protein interaction networks, and weighted gene co-expression network analysis suffer from false-positive interactions, transitivity-driven noise, and arbitrary thresholds for feature selection. Addressing these limitations, this study proposed a novel framework, Hypergraph Regularised Fuzzy Rough Gene Selection (HR-FRGS), for knowledge-driven biomarker discovery. This pipeline integrates information-rich protein clustering with entropy filtering and PPI-pruned co-expression interactions to form a dual-layer hypergraph. A hypergraph-based scoring method that combines Random Walk with Restart diffusion and Hypergraph Betweenness Centrality is used to identify genes that are central in the biological network. Furthermore an autonomous fuzzy rough set selection eliminates arbitrary threshold dependency for gene selection. HR-FRGS was applied to four TCGA cancer cohorts: Lung Adenocarcinoma, Head and Neck Cancer, Kidney Clear Cell Carcinoma, and Colon Cancer, which reduced over 60,000 transcripts to compact biomarker panels. Validation using classical Machine Learning algorithms with cross-validation demonstrated that biomarker panels matched or exceeded the classification performance of the full transcriptome (AUC > 0.99, MCC > 0.94) and confirmed generalizability (MCC $\geq$ 0.893). Benchmarking against baseline methods and an ablation study showed the contribution of major parts of the algorithm. Functional enrichment analysis captures established pan-cancer hallmarks and cohort-specific oncogenic mechanisms, confirming the biological relevance of the identified genes.

Accurate prediction of T cell receptor (TCR) specificity can potentially accelerate development of novel immunotherapies, yet public databases used to train these models contain substantial label noise. This uncertainty complicates model evaluation and limits the reliability of predicted TCR-pMHC interactions. In this work, we present a simple supervised machine learning denoising scheme to identify low-quality labels in TCR specificity data. We evaluated the framework using the TCRvdb validation dataset and further corroborated it via structural modeling with AlphaFold 3. Here, our denoising approach successfully identified non-validated TCRs, achieving an average MCC of 0.47 for filtering non-validated examples while retaining high sensitivity for validated examples. Applying our denoising approach to the complete NetTCR training dataset, we observed that data associated with known noisy attributes were significantly enriched in the filtered subset. Furthermore, TCR-pMHC complexes identified as noise consistently produced lower AlphaFold 3 confidence scores, providing independent evidence of misannotation. Experimental validation of 30 selected TCRs further supported the approach, with 80% of TCRs predicted as noise failing to demonstrate functional reactivity. Our findings demonstrate that current performance metrics may be obscured by dataset inaccuracies and provide a method for refining training data in the absence of gold-standard benchmarks.

Many biological quantities in omics are inherently periodic or directional, including circadian phase, cell-cycle position, and cellular orientation. Treating such quantities as ordinary linear variables can introduce artificial discontinuities and obscure biological interpretation. Circular methods are well developed across directional statistics, bioinformatics, and spatial statistics, but practical guidance for omics applications remains limited. We present key considerations for circular data analysis in single-cell and spatial omics, including the provenance and observational unit of the circular quantity, the inferential or predictive objective, the distinction between tissue composition and residual spatial dependence, and the choice of appropriate models. We describe relevant approaches for circular representation, inference, clustering, regression, machine learning, and spatial analysis, together with representative R software. Their practical use is illustrated through spatial transcriptomic case studies of derived circadian peak phase in a mouse model of Alzheimer's disease and latent cell-cycle position in human ovarian cancer. These examples demonstrate how the definition of the angle, its observational unit, and tissue heterogeneity shape appropriate analysis and interpretation, including when spatial modeling is not warranted. We conclude with practical recommendations and directions for future methodological development.