Briefings in bioinformatics最新文献

Finding Significant Hits in Networks (FISHNET) uses prior biological knowledge, represented as gene interaction networks and gene function annotations, to identify genes that do not meet the genome-wide significance threshold but replicate, nonetheless. Its input is gene-level P-values from any source, including omicsWAS, aggregation of genome-wide association studies P-values, CRISPR screens, or differential expression analysis. It is based on the idea that genes whose P-values are low purely by chance are distributed randomly across networks and functions, so genes with suggestive P-values that cluster in densely connected subnetworks and share common functions are less likely to reflect chance and more likely to replicate. FISHNET combines network and function analysis with permutation-based P-value thresholds to identify a small set of exceptional genes that we call FISHNET genes. Applied to 11 cardiovascular risk traits, FISHNET identified 19 gene-trait relationships that missed genome-wide significance thresholds but, nonetheless, replicated in an independent cohort. The replication rate of FISHNET genes matched that of genes with lower P-values. FISHNET identified a novel association between RUNX1 expression and HDL that is supported by experimental evidence that RUNX1 promotes white fat browning, which increases HDL cholesterol levels. FISHNET also identified an association between LTB expression and BMI that is supported by experimental evidence that higher LTB expression increases BMI via activation of the LTβR pathway. Both associations failed genome-wide significance thresholds, highlighting FISHNET's ability to uncover meaningful relationships missed by traditional methods. FISHNET software is freely available at https://brentlab.github.io/fishnet/.

Zinc ions play essential structural and catalytic roles in a wide range of proteins. Accurate prediction of their binding sites is crucial for structural and functional annotation. We present MoM2, a web-accessible tool for predicting zinc-binding sites in protein 3D structures. MoM2 employs a graph neural network trained exclusively on spatial features specifically, Cα and Cβ coordinates eliminating the need for templates or sequence-based heuristics. The tool efficiently processes entire proteomes within hours and demonstrates strong predictive performance. In a benchmark of 412 experimentally determined apo-structures, MoM2 outperformed existing methods, achieving the highest F1-score (55.7%) and the lowest false discovery rate (44.1%). The web interface supports input via structure files, PDB or UniProt IDs, and allows batch processing with customizable thresholds. As an independent validation, MoM2 correctly identified 18 out of 20 predicted zinc sites in SARS-CoV-2 proteins. The tool is freely available at https://mom2.cerm.unifi.it.

The rise of mRNA vaccines highlights the pivotal role of T-cell antigen identification in modern vaccinology and personalized medicine. T-cell recognition relies on the sophisticated ternary interaction between the T-cell receptor (TCR), the major histocompatibility complex (MHC) molecule, and the peptide antigen, which forms the peptide-MHC (pMHC) complex. Computational methods, particularly artificial intelligence (AI), are indispensable for accurately predicting these complex bindings. This review systematically surveys the rapidly evolving AI-driven landscape for T-cell antigen identification, providing a comprehensive categorization of methods for MHC-I, MHC-II, and the highly complex TCR-pMHC binding prediction, alongside foundational data resources. Crucially, we conduct a rigorous, standardized benchmarking of 18 state-of-the-art TCR-pMHC prediction models across diverse training data sources. Our evaluation on two distinct and challenging out-of-distribution (OOD) unseen epitope variant datasets reveals a significant and concerning generalization gap in current predictors. Notably, the overall absolute predictive gain remains marginal across all models under OOD conditions. This result underscores a severe and persistent generalization challenge when faced with novel epitope variants. To address these limitations, we emphasize the urgent need for enhanced structural modeling, the integration of multi-omics data, and the development of generative models for de novo TCR design. By advancing these computational frontiers, our community can accelerate the transition from prediction to rational design in immunoinformatics.

T cell receptor (TCR) recognition of peptide-major histocompatibility complex (pMHC) molecules is the critical first step in adaptive immune activation, shaping immunity against pathogens and tumors, as well as tolerance to self. Despite extensive structural characterization of TCR-pMHC complexes, the molecular principles underlying this process remain incompletely understood, hindered by the inherent duality of TCR specificity and cross-reactivity. Traditional structural analyses often fall short in capturing the multidimensional features that govern TCR-pMHC engagement. Here, we introduce a multimodal geometric deep learning framework that systematically extracts and learns various physicochemical and spatial features from pMHC interfaces, which encode key immunological cues for TCR recognition. Applied to a curated dataset of human leukocyte antigens HLA-A*02-peptide-TCR crystal structures, our model robustly predicts TCR binding preferences and uncovers interfacial "immunological fingerprints" that inform receptor engagement. Through an integrated explainability module, we identify critical contact residues and interaction motifs, thus providing interpretable insights into the determinants of TCR specificity. We further demonstrate the model's generalizability by analyzing HLA-B*27-peptide complexes, revealing potential TCR cross-reactivity between self-derived and bacterial peptides-highlighting its utility in probing molecular mimicry. This work establishes a scalable, structure-based approach for decoding T cell recognition and offers a powerful tool for guiding antigen design, vaccine development, and TCR-based immunotherapies.

Accounting for dependence among high-dimensional variables in omics data analysis is critical to obtain accurate and reliable statistical inference. Although latent, omics variables often exhibit structured correlation/co-expression patterns. However, there are few methods explicitly accounting for such structured dependence in the statistical analysis of omics data (e.g. differential expression analysis). To address this methodological gap, we propose a Coexpression network multivariate Regression (CoReg), which integrates co-expression network structure into multivariate regression analysis to precisely account for the inter-correlations (dependence) among omics variables. We show in simulations that CoReg substantially improves the accuracy of statistical inference and replicability across studies. These findings suggest that CoReg provides an alternative approach for omics data association analysis with dependence adjustment, analogous to the role of mixed-effects models in handling repeated measures in lower-dimensional settings.

While traditional imaging techniques, such as histopathology, are often part of clinical workflows, molecular profiling remains more difficult to conduct and is less cost-effective. Thus, the prediction of molecular 'omics' data directly from imaging has emerged as an appealing alternative. While existing reviews have mentioned image-based prediction of biomarkers within specific disease contexts, this review provides a comprehensive overview of current methods that leverage imaging to predict (i) DNA-based aberrations, (ii) bulk transcriptomic profiles, (iii) single-cell transcriptomics, and (iv) spatial transcriptomics across disease contexts and imaging modalities. To address the complexity of these predictive tasks, we find that many studies employ cutting-edge deep learning strategies for image processing, feature extraction, feature aggregation, and downstream molecular prediction. In this review, we highlight the diverse applications of both deep learning-based and modern statistical frameworks designed for image-based omics prediction. The insights gleaned from these inferred molecular data have broad clinical relevance and will continue to improve our understanding of the relationships between molecular and visual features, paving the way for new diagnostic and therapeutic applications.

Copy number variants (CNVs) have been shown to play a significant role in the pathogenesis of various human diseases. Although several tools have been developed for detecting CNVs based on whole-exome sequencing (WES) data, their performance remains suboptimal for small exon-level CNVs (exCNVs). This is primarily due to multiple technical variabilities, including probe capture efficiency, mappability, exon size, batch effects, experimental background noise bias, and control sample selection, all of which can lead to false negatives and false positives in exCNV detection. To address these challenges, we developed ML-ExonCNV, which innovatively integrates the XGBoost machine learning model with a multi-expert ensemble approach. The model was trained using 14 features derived from 22 364 real-world, quantitative polymerase chain reaction-validated rare exCNVs. Evaluation on a test set of 492 real WES and the NA12878 gold-standard dataset demonstrated that ML-ExonCNV outperformed widely used tools such as GATK-gCNV, ExomeDepth, and CNVkit. Notably, ML-ExonCNV can detect large segmental CNVs, mosaic CNVs, and breakpoint CNVs on exon region. Furthermore, our analysis revealed recurrent exCNV-associated genes and their phenotypic correlations. Neurodevelopmental and musculoskeletal abnormalities were identified as the most frequently associated phenotypes with high-recurrence exCNVs.

Single-cell multi-omics data reveal complex cellular states and deepen our understanding of tissue cell phenotypes and functions. However, data analysis remains challenging due to the discrete nature and high noise level of the data, as well as the lack of modality. Here, we propose scMultiNet, a multi-task deep adversarial neural network that can integrate different tasks to analyze single-cell multi-modal data. In particular, we achieve joint training of multi-modal integration and cross-modal prediction tasks by introducing a cross-modal bi-prediction module and a multi-head self-attention module. Data denoising is further enhanced by integrating an indicator matrix that constrains and precisely reconstructs the original expression values. Extensive simulations and real data experiments demonstrate that scMultiNet outperforms existing state-of-the-art methods in dimensionality reduction, visualization, clustering, batch elimination, data denoising, multi-modal integration, single-cell cross-modality translation, and in revealing cell type-specific biological insights. In addition, we demonstrate that scMultiNet can effectively transfer the complex relationships between modalities from one batch to another. In summary, scMultiNet stands as a comprehensive end-to-end framework, ideally suited for analyzing single-cell multi-omics data.

Introduction: Mutational signatures serve as molecular fingerprints of the biological processes and exposures that shape cancer genomes. However, accurate signal recovery remains challenging due to pervasive background variants, sequencing artifacts, technical noise, and platform-specific biases that obscure true mutagenic patterns, hampering biomarker discovery, and mechanistic interpretation.

Methods: Here we introduce SigRescueR, a rigorous, pan-system, computational framework based on Bayesian inference designed for noise correction and mutational signature identification. SigRescueR applies statistically robust baseline correction to effectively disentangle true mutational signals from confounding noise and artifacts.

Results: When applied to extensive datasets spanning experimental models and human cancers, SigRescueR reliably identified canonical mutational signatures associated with environmental mutagens such as colibactin, benzo[a]pyrene, and UV radiation, and chemotherapeutic agents, namely 5-fluorouracil and cisplatin. SigRescueR effectively operated across diverse mutation classes, including single base substitutions, insertions and deletions, and doublet base substitutions, while also integrating strand bias and duplex sequencing data for toxicology applications.

Conclusion: SigRescueR offers a unified, high-precision platform that seamlessly integrates cancer genomics, molecular toxicology, and mechanistic studies. It enables precise mapping of mutagenic processes and identification of robust genomic biomarkers of environmental and therapeutic exposures, providing a transformative framework for translational cancer research.

Availability and implementation: SigRescueR is implemented in R and provided as open-source software on GitHub at https://github.com/ZhivaguiLab/SigRescueR/.

RNA language models (LMs) are increasingly applied to RNA structure and function analysis, yet their intrinsic representational capacities remain poorly characterized. Here, we present a standardized zero-shot evaluation of 21 RNA LMs, with representative DNA LMs included as reference controls. Three complementary tasks-attention-based RNA secondary structure prediction, embedding-based RNA classification, and mutational fitness estimation from sequence likelihoods-are evaluated without downstream fine-tuning. Our results reveal substantial variability across models and clear trade-offs between structural, functional, and evolutionary representations. RNA-specific, noncoding RNA-enriched pretraining is crucial for capturing structural information, while evolutionary signals from multiple sequence alignments substantially boost performance. Although model scaling yields gains, architectural and objective choices critically influence performance across task categories. Together, this study provides a foundational benchmark, highlights inherent challenges in learning unified RNA representations, and offers insights for developing next-generation RNA foundation models.