Computational Biology

Genomics, bioinformatics, and quantitative methods in biology

Includes:Genomics(q-bio.GN)Quantitative Methods(q-bio.QM)

Trending in Computational Biology

2511.19068

The TAG array of a multiple sequence alignment

Modern genomic analyses increasingly rely on pangenomes, that is, representations of the genome of entire populations. The simplest representation of a pangenome is a set of individual genome sequences. Compared to e.g. sequence graphs, this has the advantage that efficient exact search via indexes based on the Burrows-Wheeler Transform (BWT) is possible, that no chimeric sequences are created, and that the results are not influenced by heuristics. However, such an index may report a match in thousands of positions even if these all correspond to the same locus, making downstream analysis unnecessarily expensive. For sufficiently similar sequences (e.g. human chromosomes), a multiple sequence alignment (MSA) can be computed. Since an MSA tends to group similar strings in the same columns, it is likely that a string occurring thousands of times in the pangenome can be described by very few columns in the MSA. We describe a method to tag entries in the BWT with the corresponding column in the MSA and develop an index that can map matches in the BWT to columns in the MSA in time proportional to the output. As a by-product, we can efficiently project a match to a designated reference genome, a capability that current pangenome aligners based on the BWT lack.

2511.190681

Nov 2025Genomics

CellStream: Dynamical Optimal Transport Informed Embeddings for Reconstructing Cellular Trajectories from Snapshots Data

Single-cell RNA sequencing (scRNA-seq), especially temporally resolved datasets, enables genome-wide profiling of gene expression dynamics at single-cell resolution across discrete time points. However, current technologies provide only sparse, static snapshots of cell states and are inherently influenced by technical noise, complicating the inference and representation of continuous transcriptional dynamics. Although embedding methods can reduce dimensionality and mitigate technical noise, the majority of existing approaches typically treat trajectory inference separately from embedding construction, often neglecting temporal structure. To address this challenge, here we introduce CellStream, a novel deep learning framework that jointly learns embedding and cellular dynamics from single-cell snapshot data by integrating an autoencoder with unbalanced dynamical optimal transport. Compared to existing methods, CellStream generates dynamics-informed embeddings that robustly capture temporal developmental processes while maintaining high consistency with the underlying data manifold. We demonstrate CellStream's effectiveness on both simulated datasets and real scRNA-seq data, including spatial transcriptomics. Our experiments indicate significant quantitative improvements over state-of-the-art methods in representing cellular trajectories with enhanced temporal coherence and reduced noise sensitivity. Overall, CellStream provides a new tool for learning and representing continuous streams from the noisy, static snapshots of single-cell gene expression.

2511.137861

Nov 2025Genomics

Protein Structure Tokenization via Geometric Byte Pair Encoding

Protein structure is central to biological function, and enabling multimodal protein models requires joint reasoning over sequence, structure, and function. A key barrier is the lack of principled protein structure tokenizers (PSTs): existing approaches fix token size or rely on continuous vector codebooks, limiting interpretability, multi-scale control, and transfer across architectures. We introduce GeoBPE, a geometry-grounded PST that transforms continuous, noisy, multi-scale backbone conformations into discrete ``sentences'' of geometry while enforcing global constraints. Analogous to byte-pair encoding, GeoBPE generates a hierarchical vocabulary of geometric primitives by iteratively (i) clustering Geo-Pair occurrences with k-medoids to yield a resolution-controllable vocabulary; (ii) quantizing each Geo-Pair to its closest medoid prototype; and (iii) reducing drift through differentiable inverse kinematics that optimizes boundary glue angles under an $\mathrm{SE}(3)$ end-frame loss. GeoBPE offers compression ($>$10x reduction in bits-per-residue at similar distortion rate), data efficiency ($>$10x less training data), and generalization (maintains test/train distortion ratio of $1.0-1.1$). It is architecture-agnostic: (a) its hierarchical vocabulary provides a strong inductive bias for coarsening residue-level embeddings from large PLMs into motif- and protein-level representations, consistently outperforming leading PSTs across $12$ tasks and $24$ test splits; (b) paired with a transformer, GeoBPE supports unconditional backbone generation via language modeling; and (c) tokens align with CATH functional families and support expert-interpretable case studies, offering functional meaning absent in prior PSTs. Code is available at https://github.com/shiningsunnyday/PT-BPE/.

2511.117581

Nov 2025Quantitative Methods

Fast, memory-efficient genomic interval tokenizers for modern machine learning

Introduction: Epigenomic datasets from high-throughput sequencing experiments are commonly summarized as genomic intervals. As the volume of this data grows, so does interest in analyzing it through deep learning. However, the heterogeneity of genomic interval data, where each dataset defines its own regions, creates barriers for machine learning methods that require consistent, discrete vocabularies. Methods: We introduce gtars-tokenizers, a high-performance library that maps genomic intervals to a predefined universe or vocabulary of regions, analogous to text tokenization in natural language processing. Built in Rust with bindings for Python, R, CLI, and WebAssembly, gtars-tokenizers implements two overlap methods (BITS and AIList) and integrates seamlessly with modern ML frameworks through Hugging Face-compatible APIs. Results: The gtars-tokenizers package achieves top efficiency for large-scale datasets, while enabling genomic intervals to be processed using standard ML workflows in PyTorch and TensorFlow without ad hoc preprocessing. This token-based approach bridges genomics and machine learning, supporting scalable and standardized analysis of interval data across diverse computational environments. Availability: PyPI and GitHub: https://github.com/databio/gtars.

2511.015551

Nov 2025Genomics

scMRDR: A scalable and flexible framework for unpaired single-cell multi-omics data integration

Advances in single-cell sequencing have enabled high-resolution profiling of diverse molecular modalities, while integrating unpaired multi-omics single-cell data remains challenging. Existing approaches either rely on pair information or prior correspondences, or require computing a global pairwise coupling matrix, limiting their scalability and flexibility. In this paper, we introduce a scalable and flexible generative framework called single-cell Multi-omics Regularized Disentangled Representations (scMRDR) for unpaired multi-omics integration. Specifically, we disentangle each cell's latent representations into modality-shared and modality-specific components using a well-designed $β$-VAE architecture, which are augmented with isometric regularization to preserve intra-omics biological heterogeneity, adversarial objective to encourage cross-modal alignment, and masked reconstruction loss strategy to address the issue of missing features across modalities. Our method achieves excellent performance on benchmark datasets in terms of batch correction, modality alignment, and biological signal preservation. Crucially, it scales effectively to large-level datasets and supports integration of more than two omics, offering a powerful and flexible solution for large-scale multi-omics data integration and downstream biological discovery.

2510.249872

Oct 2025Quantitative Methods

Physics-Informed Deep Learning for Improved Input Function Estimation in Motion-Blurred Dynamic [${}^{18}$F]FDG PET Images

Kinetic modeling enables \textit{in vivo} quantification of tracer uptake and glucose metabolism in [${}^{18}$F]Fluorodeoxyglucose ([${}^{18}$F]FDG) dynamic positron emission tomography (dPET) imaging of mice. However, kinetic modeling requires the accurate determination of the arterial input function (AIF) during imaging, which is time-consuming and invasive. Recent studies have shown the efficacy of using deep learning to directly predict the input function, surpassing established methods such as the image-derived input function (IDIF). In this work, we trained a physics-informed deep learning-based input function prediction model (PIDLIF) to estimate the AIF directly from the PET images, incorporating a kinetic modeling loss during training. The proposed method uses a two-tissue compartment model over two regions, the myocardium and brain of the mice, and is trained on a dataset of 70 [${}^{18}$F]FDG dPET images of mice accompanied by the measured AIF during imaging. The proposed method had comparable performance to the network without a physics-informed loss, and when sudden movement causing blurring in the images was simulated, the PIDLIF model maintained high performance in severe cases of image degradation. The proposed physics-informed method exhibits an improved robustness that is promoted by physically constraining the problem, enforcing consistency for out-of-distribution samples. In conclusion, the PIDLIF model offers insight into the effects of leveraging physiological distribution mechanics in mice to guide a deep learning-based AIF prediction network in images with severe degradation as a result of blurring due to movement during imaging.

2510.212811

Oct 2025Quantitative Methods

Drug-disease networks and drug repurposing

Repurposing existing drugs to treat new diseases is a cost-effective alternative to de novo drug development, but there are millions of potential drug-disease combinations to be considered with only a small fraction being viable. In silico predictions of drug-disease associations can be invaluable for reducing the size of the search space. In this work we present a novel network of drugs and the diseases they treat, compiled using a combination of existing textual and machine-readable databases, natural-language processing tools, and hand curation, and analyze it using network-based link prediction methods to identify potential drug-disease combinations. We measure the efficacy of these methods using cross-validation tests and find that several methods, particularly those based on graph embedding and network model fitting, achieve impressive prediction performance, significantly better than previous approaches, with area under the ROC curve above 0.95 and average precision almost a thousand times better than chance.

2510.199481

Oct 2025Quantitative Methods

Triangle Multiplication Is All You Need For Biomolecular Structure Representations

AlphaFold has transformed protein structure prediction, but emerging applications such as virtual ligand screening, proteome-wide folding, and de novo binder design demand predictions at a massive scale, where runtime and memory costs become prohibitive. A major bottleneck lies in the Pairformer backbone of AlphaFold3-style models, which relies on computationally expensive triangular primitives-especially triangle attention-for pairwise reasoning. We introduce Pairmixer, a streamlined alternative that eliminates triangle attention while preserving higher-order geometric reasoning capabilities that are critical for structure prediction. Pairmixer substantially improves computational efficiency, matching state-of-the-art structure predictors across folding and docking benchmarks, delivering up to 4x faster inference on long sequences while reducing training cost by 34%. Its efficiency alleviates the computational burden of downstream applications such as modeling large protein complexes, high-throughput ligand and binder screening, and hallucination-based design. Within BoltzDesign, for example, Pairmixer delivers over 2x faster sampling and scales to sequences ~30% longer than the memory limits of Pairformer. Code is available at https://github.com/genesistherapeutics/pairmixer.

2510.188701

Oct 2025Quantitative Methods

Identifying multi-omics interactions for lung cancer drug targets discovery using Kernel Machine Regression

Cancer exhibits diverse and complex phenotypes driven by multifaceted molecular interactions. Recent biomedical research has emphasized the comprehensive study of such diseases by integrating multi-omics datasets (genome, proteome, transcriptome, epigenome). This approach provides an efficient method for identifying genetic variants associated with cancer and offers a deeper understanding of how the disease develops and spreads. However, it is challenging to comprehend complex interactions among the features of multi-omics datasets compared to single omics. In this paper, we analyze lung cancer multi-omics datasets from The Cancer Genome Atlas (TCGA). Using four statistical methods, LIMMA, the T test, Canonical Correlation Analysis (CCA), and the Wilcoxon test, we identified differentially expressed genes across gene expression, DNA methylation, and miRNA expression data. We then integrated these multi-omics data using the Kernel Machine Regression (KMR) approach. Our findings reveal significant interactions among the three omics: gene expression, miRNA expression, and DNA methylation in lung cancer. From our data analysis, we identified 38 genes significantly associated with lung cancer. From our data analysis, we identified 38 genes significantly associated with lung cancer. Among these, eight genes of highest ranking (PDGFRB, PDGFRA, SNAI1, ID1, FGF11, TNXB, ITGB1, ZIC1) were highlighted by rigorous statistical analysis. Furthermore, in silico studies identified three top-ranked potential candidate drugs (Selinexor, Orapred, and Capmatinib) that could play a crucial role in the treatment of lung cancer. These proposed drugs are also supported by the findings of other independent studies, which underscore their potential efficacy in the fight against lung cancer.

2510.160932

Oct 2025Genomics

AGNES: Adaptive Graph Neural Network and Dynamic Programming Hybrid Framework for Real-Time Nanopore Seed Chaining

Nanopore sequencing enables real-time long-read DNA sequencing with reads exceeding 10 kilobases, but inherent error rates of 12-15 percent present significant computational challenges for read alignment. The critical seed chaining step must connect exact k-mer matches between reads and reference genomes while filtering spurious matches, yet state-of-the-art methods rely on fixed gap penalty functions unable to adapt to varying genomic contexts including tandem repeats and structural variants. This paper presents RawHash3, a hybrid framework combining graph neural networks with classical dynamic programming for adaptive seed chaining that maintains real-time performance while providing statistical guarantees. We formalize seed chaining as graph learning where seeds constitute nodes with 12-dimensional feature vectors and edges encode 8-dimensional spatial relationships including gap consistency. Our architecture employs three-layer EdgeConv GNN with confidence-based method selection that dynamically switches between learned guidance and algorithmic fallback. Comprehensive evaluation on 1,000 synthetic nanopore reads with 5,200 test seeds demonstrates RawHash3 achieves 99.94 percent precision and 40.07 percent recall, representing statistically significant 25.0 percent relative improvement over baseline with p less than 0.001. The system maintains median inference latency of 1.59ms meeting real-time constraints, while demonstrating superior robustness with 100 percent success rate under 20 percent label corruption versus baseline degradation to 30.3 percent. Cross-validation confirms stability establishing graph neural networks as viable approach for production genomics pipelines.

2510.160131

Oct 2025Genomics

Phenome-Wide Multi-Omics Integration Uncovers Distinct Archetypes of Human Aging

Aging is a highly complex and heterogeneous process that progresses at different rates across individuals, making biological age (BA) a more accurate indicator of physiological decline than chronological age. While previous studies have built aging clocks using single-omics data, they often fail to capture the full molecular complexity of human aging. In this work, we leveraged the Human Phenotype Project, a large-scale cohort of 10,000 adults aged 40-70 years, with extensive longitudinal profiling that includes clinical, behavioral, environmental, and multi-omics datasets spanning transcriptomics, lipidomics, metabolomics, and the microbiome. By employing advanced machine learning frameworks capable of modeling nonlinear biological dynamics, we developed and rigorously validated a multi-omics aging clock that robustly predicts diverse health outcomes and future disease risk. Unsupervised clustering of the integrated molecular profiles from multi-omics uncovered distinct biological subtypes of aging, revealing striking heterogeneity in aging trajectories and pinpointing pathway-specific alterations associated with different aging patterns. These findings demonstrate the power of multi-omics integration to decode the molecular landscape of aging and lay the groundwork for personalized healthspan monitoring and precision strategies to prevent age-related diseases.

2510.123841

Oct 2025Genomics

PRISM: Enhancing Protein Inverse Folding through Fine-Grained Retrieval on Structure-Sequence Multimodal Representations

Designing protein sequences that fold into a target three-dimensional structure, known as the inverse folding problem, is central to protein engineering but remains challenging due to the vast sequence space and the importance of local structural constraints. Existing deep learning approaches achieve strong recovery rates, yet they lack explicit mechanisms to reuse fine-grained structure-sequence patterns that are conserved across natural proteins. We present PRISM, a multimodal retrieval-augmented generation framework for inverse folding that retrieves fine-grained representations of potential motifs from known proteins and integrates them with a hybrid self-cross attention decoder. PRISM is formulated as a latent-variable probabilistic model and implemented with an efficient approximation, combining theoretical grounding with practical scalability. Across five benchmarks (CATH-4.2, TS50, TS500, CAMEO 2022, and the PDB date split), PRISM establishes new state of the art in both perplexity and amino acid recovery, while also improving foldability metrics (RMSD, TM-score, pLDDT), demonstrating that fine-grained multimodal retrieval is a powerful and efficient paradigm for protein sequence design.

2510.117501

Oct 2025Quantitative Methods

BioVERSE: Representation Alignment of Biomedical Modalities to LLMs for Multi-Modal Reasoning

Recent advances in large language models (LLMs) and biomedical foundation models (BioFMs) have achieved strong results in biological text reasoning, molecular modeling, and single-cell analysis, yet they remain siloed in disjoint embedding spaces, limiting cross-modal reasoning. We present BIOVERSE (Biomedical Vector Embedding Realignment for Semantic Engagement), a two-stage approach that adapts pretrained BioFMs as modality encoders and aligns them with LLMs through lightweight, modality-specific projection layers. The approach first aligns each modality to a shared LLM space through independently trained projections, allowing them to interoperate naturally, and then applies standard instruction tuning with multi-modal data to bring them together for downstream reasoning. By unifying raw biomedical data with knowledge embedded in LLMs, the approach enables zero-shot annotation, cross-modal question answering, and interactive, explainable dialogue. Across tasks spanning cell-type annotation, molecular description, and protein function reasoning, compact BIOVERSE configurations surpass larger LLM baselines while enabling richer, generative outputs than existing BioFMs, establishing a foundation for principled multi-modal biomedical reasoning.

2510.014281

Oct 2025Quantitative Methods

MorphGen: Controllable and Morphologically Plausible Generative Cell-Imaging

Simulating in silico cellular responses to interventions is a promising direction to accelerate high-content image-based assays, critical for advancing drug discovery and gene editing. To support this, we introduce MorphGen, a state-of-the-art diffusion-based generative model for fluorescent microscopy that enables controllable generation across multiple cell types and perturbations. To capture biologically meaningful patterns consistent with known cellular morphologies, MorphGen is trained with an alignment loss to match its representations to the phenotypic embeddings of OpenPhenom, a state-of-the-art biological foundation model. Unlike prior approaches that compress multichannel stains into RGB images -- thus sacrificing organelle-specific detail -- MorphGen generates the complete set of fluorescent channels jointly, preserving per-organelle structures and enabling a fine-grained morphological analysis that is essential for biological interpretation. We demonstrate biological consistency with real images via CellProfiler features, and MorphGen attains an FID score over 35% lower than the prior state-of-the-art MorphoDiff, which only generates RGB images for a single cell type. Code is available at https://github.com/czi-ai/MorphGen.

2510.012981

Oct 2025Quantitative Methods

Commutative algebra neural network reveals genetic origins of diseases

Genetic mutations can disrupt protein structure, stability, and solubility, contributing to a wide range of diseases. Existing predictive models often lack interpretability and fail to integrate physical and chemical interactions critical to molecular mechanisms. Moreover, current approaches treat disease association, stability changes, and solubility alterations as separate tasks, limiting model generalizability. In this study, we introduce a unified framework based on multiscale commutative algebra to capture intrinsic physical and chemical interactions for the first time. Leveraging Persistent Stanley-Reisner Theory, we extract multiscale algebraic invariants to build a Commutative Algebra neural Network (CANet). Integrated with transformer features and auxiliary physical features, we apply CANet to tackle three key domains for the first time: disease-associated mutations, mutation-induced protein stability changes, and solubility changes upon mutations. Across six benchmark tasks, CANet and its gradient boosting tree counterpart, CATree, consistently attain state-of-the-art performance, achieving up to 7.5% improvement in predictive accuracy. Our approach offers multiscale, mechanistic, interpretable,and generalizable models for predicting disease-mutation associations.

2509.265664

Sep 2025Quantitative Methods

scUnified: An AI-Ready Standardized Resource for Single-Cell RNA Sequencing Analysis

Single-cell RNA sequencing (scRNA-seq) technology enables systematic delineation of cellular states and interactions, providing crucial insights into cellular heterogeneity. Building on this potential, numerous computational methods have been developed for tasks such as cell clustering, cell type annotation, and marker gene identification. To fully assess and compare these methods, standardized, analysis-ready datasets are essential. However, such datasets remain scarce, and variations in data formats, preprocessing workflows, and annotation strategies hinder reproducibility and complicate systematic evaluation of existing methods. To address these challenges, we present scUnified, an AI-ready standardized resource for single-cell RNA sequencing data that consolidates 13 high-quality datasets spanning two species (human and mouse) and nine tissue types. All datasets undergo standardized quality control and preprocessing and are stored in a uniform format to enable direct application in diverse computational analyses without additional data cleaning. We further demonstrate the utility of scUnified through experimental analyses of representative biological tasks, providing a reproducible foundation for the standardized evaluation of computational methods on a unified dataset.

2509.258841

Sep 2025Genomics

HEIST: A Graph Foundation Model for Spatial Transcriptomics and Proteomics Data

Single-cell transcriptomics and proteomics have become a great source for data-driven insights into biology, enabling the use of advanced deep learning methods to understand cellular heterogeneity and gene expression at the single-cell level. With the advent of spatial-omics data, we have the promise of characterizing cells within their tissue context as it provides both spatial coordinates and intra-cellular transcriptional or protein counts. Proteomics offers a complementary view by directly measuring proteins, which are the primary effectors of cellular function and key therapeutic targets. However, existing models either ignore the spatial information or the complex genetic and proteomic programs within cells. Thus they cannot infer how cell internal regulation adapts to microenvironmental cues. Furthermore, these models often utilize fixed gene vocabularies, hindering their generalizability unseen genes. In this paper, we introduce HEIST, a hierarchical graph transformer foundation model for spatial transcriptomics and proteomics. HEIST models tissues as hierarchical graphs. The higher level graph is a spatial cell graph, and each cell in turn, is represented by its lower level gene co-expression network graph. HEIST achieves this by performing both intra-level and cross-level message passing to utilize the hierarchy in its embeddings and can thus generalize to novel datatypes including spatial proteomics without retraining. HEIST is pretrained on 22.3M cells from 124 tissues across 15 organs using spatially-aware contrastive and masked autoencoding objectives. Unsupervised analysis of HEIST embeddings reveals spatially informed subpopulations missed by prior models. Downstream evaluations demonstrate generalizability to proteomics data and state-of-the-art performance in clinical outcome prediction, cell type annotation, and gene imputation across multiple technologies.

2506.111522

Jun 2025Genomics

AbBiBench: A Benchmark for Antibody Binding Affinity Maturation and Design

We introduce AbBiBench (Antibody Binding Benchmarking), a benchmarking framework for antibody binding affinity maturation and design. Unlike previous strategies that evaluate antibodies in isolation, typically by comparing them to natural sequences with metrics such as amino acid recovery rate or structural RMSD, AbBiBench instead treats the antibody-antigen (Ab-Ag) complex as the fundamental unit. It evaluates an antibody design's binding potential by measuring how well a protein model scores the full Ab-Ag complex. We first curate, standardize, and share more than 184,500 experimental measurements of antibody mutants across 14 antibodies and 9 antigens-including influenza, lysozyme, HER2, VEGF, integrin, Ang2, and SARS-CoV-2-covering both heavy-chain and light-chain mutations. Using these datasets, we systematically compare 15 protein models including masked language models, autoregressive language models, inverse folding models, diffusion-based generative models, and geometric graph models by comparing the correlation between model likelihood and experimental affinity values. Additionally, to demonstrate AbBiBench's generative utility, we apply it to antibody F045-092 in order to introduce binding to influenza H1N1. We sample new antibody variants with the top-performing models, rank them by the structural integrity and biophysical properties of the Ab-Ag complex, and assess them with in vitro ELISA binding assays. Our findings show that structure-conditioned inverse folding models outperform others in both affinity correlation and generation tasks. Overall, AbBiBench provides a unified, biologically grounded evaluation framework to facilitate the development of more effective, function-aware antibody design models.

2506.042354

May 2025Quantitative Methods

ChromFound: Towards A Universal Foundation Model for Single-Cell Chromatin Accessibility Data

The advent of single-cell Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq) offers an innovative perspective for deciphering regulatory mechanisms by assembling a vast repository of single-cell chromatin accessibility data. While foundation models have achieved significant success in single-cell transcriptomics, there is currently no foundation model for scATAC-seq that supports zero-shot high-quality cell identification and comprehensive multi-omics analysis simultaneously. Key challenges lie in the high dimensionality and sparsity of scATAC-seq data, as well as the lack of a standardized schema for representing open chromatin regions (OCRs). Here, we present ChromFound, a foundation model tailored for scATAC-seq. ChromFound utilizes a hybrid architecture and genome-aware tokenization to effectively capture genome-wide long contexts and regulatory signals from dynamic chromatin landscapes. Pretrained on 1.97 million cells from 30 tissues and 6 disease conditions, ChromFound demonstrates broad applicability across 6 diverse tasks. Notably, it achieves robust zero-shot performance in generating universal cell representations and exhibits excellent transferability in cell type annotation and cross-omics prediction. By uncovering enhancer-gene links undetected by existing computational methods, ChromFound offers a promising framework for understanding disease risk variants in the noncoding genome.

2505.126382

May 2025Genomics

scSiameseClu: A Siamese Clustering Framework for Interpreting single-cell RNA Sequencing Data

Single-cell RNA sequencing (scRNA-seq) reveals cell heterogeneity, with cell clustering playing a key role in identifying cell types and marker genes. Recent advances, especially graph neural networks (GNNs)-based methods, have significantly improved clustering performance. However, the analysis of scRNA-seq data remains challenging due to noise, sparsity, and high dimensionality. Compounding these challenges, GNNs often suffer from over-smoothing, limiting their ability to capture complex biological information. In response, we propose scSiameseClu, a novel Siamese Clustering framework for interpreting single-cell RNA-seq data, comprising of 3 key steps: (1) Dual Augmentation Module, which applies biologically informed perturbations to the gene expression matrix and cell graph relationships to enhance representation robustness; (2) Siamese Fusion Module, which combines cross-correlation refinement and adaptive information fusion to capture complex cellular relationships while mitigating over-smoothing; and (3) Optimal Transport Clustering, which utilizes Sinkhorn distance to efficiently align cluster assignments with predefined proportions while maintaining balance. Comprehensive evaluations on seven real-world datasets demonstrate that scSiameseClu outperforms state-of-the-art methods in single-cell clustering, cell type annotation, and cell type classification, providing a powerful tool for scRNA-seq data interpretation.

2505.126266

May 2025Genomics

175 papers

Identifying expanding TCR clonotypes with a longitudinal Bayesian mixture model and their associations with cancer patient prognosis, metastasis-directed therapy, and VJ gene enrichment

Examination of T-cell receptor (TCR) clonality has become a way of understanding immunologic response to cancer and its interventions in recent years. An aspect of these analyses is determining which receptors expand or contract statistically significantly as a function of an exogenous perturbation such as therapeutic intervention. We characterize the commonly used Fisher's exact test approach for such analyses and propose an alternative formulation that does not necessitate pairwise, within-patient comparisons. We develop this flexible Bayesian longitudinal mixture model that accommodates variable length patient followup and handles missingness where present, not omitting data in estimation because of structural practicalities. Once clones are partitioned by the model into dynamic (expanding or contracting) and static categories, one can associate their counts or other characteristics with disease state, interventions, baseline biomarkers, and patient prognosis. We apply these developments to a cohort of prostate cancer patients who underwent randomized metastasis-directed therapy or not. Our analyses reveal a significant increase in clonal expansions among MDT patients and their association with later progressions both independent and within strata of MDT. Analysis of receptor motifs and VJ gene enrichment combinations using a high-dimensional penalized log-linear model we develop also suggests distinct biological characteristics of expanding clones, with and without inducement by MDT.

2601.04536Jan 2026

View

Biomechanically Informed Image Registration for Patient-Specific Aortic Valve Strain Analysis

Aortic valve (AV) biomechanics play a critical role in maintaining normal cardiac function. Pathological variations, particularly in bicuspid aortic valves (BAVs), alter leaflet loading, increase strain, and accelerate disease progression. Accurate, patient-specific characterization of valve geometry and deformation is essential for predicting disease progression and guiding durable repair. Current imaging and computational methods often fail to capture rapid valve motion and complex patient-specific features. To address these challenges, we combined image registration with finite element method (FEM) to enhance AV tracking and biomechanical assessment. Patient-specific valve geometries from 4D transesophageal echocardiography (TEE) and CT were used in FEM to model AV closure and generate intermediate deformation states. The FEM-generated states facilitated leaflet tracking, while the registration algorithm corrected mismatches between simulation and image. Across 20 patients, FEM-augmented registration improved accuracy by 40% compared with direct registration (33% for TEE, 46% for CT). This improvement enabled more reliable strain estimation directly from imaging and reducing uncertainties from boundary conditions and material assumptions. Areal and Green-Lagrange strains, as well as effective strain, were quantified in adult trileaflet/bicuspid, and pediatric patients. Trileaflet adults showed uniform deformation, BAVs exhibited asymmetric strain, and pediatric valves had low mean areal strain with high variability. Convergence between trileaflet adult and pediatric valves in mean effective strain suggests volumetric deformation drives age- and size-related differences. The FEM-augmented registration framework enhances geometric tracking and provides clinically relevant insights into patient-specific AV deformation, supporting individualized intervention planning.

2601.04375Jan 2026

View

Tool Choice Matters: Evaluating edgeR vs. DESeq2 for Sensitivity, Robustness, and Cross-Study Performance

Differential gene expression (DGE) analysis is foundational to transcriptomic research, yet tool selection can substantially influence results. This study presents a comprehensive comparison of two widely used DGE tools, edgeR and DESeq2, using real and semi-simulated bulk RNA-Seq datasets spanning viral, bacterial, and fibrotic conditions. We evaluated tool performance across three key dimensions: (1) sensitivity to sample size and robustness to outliers; (2) classification performance of uniquely identified gene sets within the discovery dataset; and (3) generalizability of tool-specific gene sets across independent studies. First, both tools showed similar responses to simulated outliers, with Jaccard similarity between the DEG sets from perturbed and original (unperturbed) data decreasing as more outliers were added. Second, classification models trained on tool-specific genes showed that edgeR achieved higher F1 scores in 9 of 13 contrasts and more frequently reached perfect or near-perfect precision. Dolan-More performance profiles further indicated that edgeR maintained performance closer to optimal across a greater proportion of datasets. Third, in cross-study validation using four independent SARS-CoV-2 datasets, gene sets uniquely identified by edgeR yielded higher AUC, precision, and recall in classifying samples from held-out datasets. This pattern was consistent across folds, with some test cases achieving perfect separation using edgeR-specific genes. In contrast, DESeq2-specific genes showed lower and more variable performance across studies. Overall, our findings highlight that while DESeq2 may identify more DEGs even under stringent significance conditions, edgeR yields more robust and generalizable gene sets for downstream classification and cross-study replication, which underscores key trade-offs in tool selection for transcriptomic analyses.

2601.04122Jan 2026

View

DNACHUNKER: Learnable Tokenization for DNA Language Models

DNA language models have emerged as powerful tools for decoding the complex language of DNA sequences. However, the performance of these models is heavily affected by their tokenization strategy, i.e., a method used to parse DNA sequences into a shorter sequence of chunks. In this work, we propose DNACHUNKER, which integrates a learnable dynamic DNA tokenization mechanism and is trained as a masked language model. Adopting the dynamic chunking procedure proposed by H-Net, our model learns to segment sequences into variable-length chunks. This dynamic chunking offers two key advantages: it's resilient to shifts and mutations in the DNA, and it allocates more detail to important functional areas. We demonstrate the performance of DNACHUNKER by training it on the human reference genome (HG38) and testing it on the Nucleotide Transformer and Genomic benchmarks. Further ablative experiments reveal that DNACHUNKER learns tokenization that grasps biological grammar and uses smaller chunks to preserve detail in important functional elements such as promoters and exons, while using larger chunks for repetitive, redundant regions.

2601.03019Jan 2026

View

MetagenBERT: a Transformer-based Architecture using Foundational genomic Large Language Models for novel Metagenome Representation

Metagenomic disease prediction commonly relies on species abundance tables derived from large, incomplete reference catalogs, constraining resolution and discarding valuable information contained in DNA reads. To overcome these limitations, we introduce MetagenBERT, a Transformer based framework that produces end to end metagenome embeddings directly from raw DNA sequences, without taxonomic or functional annotations. Reads are embedded using foundational genomic language models (DNABERT2 and the microbiome specialized DNABERTMS), then aggregated through a scalable clustering strategy based on FAISS accelerated KMeans. Each metagenome is represented as a cluster abundance vector summarizing the distribution of its embedded reads. We evaluate this approach on five benchmark gut microbiome datasets (Cirrhosis, T2D, Obesity, IBD, CRC). MetagenBERT achieves competitive or superior AUC performance relative to species abundance baselines across most tasks. Concatenating both representations further improves prediction, demonstrating complementarity between taxonomic and embedding derived signals. Clustering remains robust when applied to as little as 10% of reads, highlighting substantial redundancy in metagenomes and enabling major computational gains. We additionally introduce MetagenBERT Glob Mcardis, a cross cohort variant trained on the large, phenotypically diverse MetaCardis cohort and transferred to other datasets, retaining predictive signal including for unseen phenotypes, indicating the feasibility of a foundation model for metagenome representation. Robustness analyses (PERMANOVA, PERMDISP, entropy) show consistent separation of different states across subsamples. Overall, MetagenBERT provides a scalable, annotation free representation of metagenomes pointing toward future phenotype aware generalization across heterogeneous cohorts and sequencing technologies.

2601.03295Jan 2026

View

Apparent Selection Pressure for Channel Capacity in Bacterial Chemotactic Sensors

Bacterial chemotactic sensing converts noisy chemical signals into running and tumbling. We analyze the static sensing limits of mixed Tar/Tsr chemoreceptor clusters in individual Escherichia coli cells using a heterogeneous Monod-Wyman-Changeux (MWC) model. By sweeping a seven-dimensional parameter space, we compute three sensing performance metrics-channel capacity, effective Hill coefficient, and dynamic range. Across E. coli-like parameter regimes, we consistently observe pronounced local maxima of channel capacity, whereas neither the effective Hill coefficient nor the dynamic range exhibit comparable optimization. The capacity-achieving input distribution is bimodal, which implies that individual cells maximize information by sampling both low- and high concentration regimes. Together, these results suggest that, at the individual-cell level, channel capacity may be selected for in E. coli receptor clusters.

2601.02446Jan 2026

View

A New Framework for Explainable Rare Cell Identification in Single-Cell Transcriptomics Data

The detection of rare cell types in single-cell transcriptomics data is crucial for elucidating disease pathogenesis and tissue development dynamics. However, a critical gap that persists in current methods is their inability to provide an explanation based on genes for each cell they have detected as rare. We identify three primary sources of this deficiency. First, the anomaly detectors often function as "black boxes", designed to detect anomalies but unable to explain why a cell is anomalous. Second, the standard analytical framework hinders interpretability by relying on dimensionality reduction techniques, such as Principal Component Analysis (PCA), which transform meaningful gene expression data into abstract, uninterpretable features. Finally, existing explanation algorithms cannot be readily applied to this domain, as single-cell data is characterized by high dimensionality, noise, and substantial sparsity. To overcome these limitations, we introduce a framework for explainable anomaly detection in single-cell transcriptomics data which not only identifies individual anomalies, but also provides a visual explanation based on genes that makes an instance anomalous. This framework has two key ingredients that are not existed in current methods applied in this domain. First, it eliminates the PCA step which is deemed to be an essential component in previous studies. Second, it employs the state-of-art anomaly detector and explainer as the efficient and effective means to find each rare cell and the relevant gene subspace in order to provide explanations for each rare cell as well as the typical normal cell associated with the rare cell's closest normal cells.

2601.01358Jan 2026

View

Comparative Analysis of Formula and Structure Prediction from Tandem Mass Spectra

Liquid chromatography mass spectrometry (LC-MS)-based metabolomics and exposomics aim to measure detectable small molecules in biological samples. The results facilitate hypothesis-generating discovery of metabolic changes and disease mechanisms and provide information about environmental exposures and their effects on human health. Metabolomics and exposomics are made possible by the high resolving power of LC and high mass measurement accuracy of MS. However, a majority of the signals from such studies still cannot be identified or annotated using conventional library searching because existing spectral libraries are far from covering the vast chemical space captured by LC-MS/MS. To address this challenge and unleash the full potential of metabolomics and exposomics, a number of computational approaches have been developed to predict compounds based on tandem mass spectra. Published assessment of these approaches used different datasets and evaluation. To select prediction workflows for practical applications and identify areas for further improvements, we have carried out a systematic evaluation of the state-of-the-art prediction algorithms. Specifically, the accuracy of formula prediction and structure prediction was evaluated for different types of adducts. The resulting findings have established realistic performance baselines, identified critical bottlenecks, and provided guidance to further improve compound predictions based on MS.

2601.00941Jan 2026

View

Benchmarking Preprocessing and Integration Methods in Single-Cell Genomics

Single-cell data analysis has the potential to revolutionize personalized medicine by characterizing disease-associated molecular changes at the single-cell level. Advanced single-cell multimodal assays can now simultaneously measure various molecules (e.g., DNA, RNA, Protein) across hundreds of thousands of individual cells, providing a comprehensive molecular readout. A significant analytical challenge is integrating single-cell measurements across different modalities. Various methods have been developed to address this challenge, but there has been no systematic evaluation of these techniques with different preprocessing strategies. This study examines a general pipeline for single-cell data analysis, which includes normalization, data integration, and dimensionality reduction. The performance of different algorithm combinations often depends on the dataset sizes and characteristics. We evaluate six datasets across diverse modalities, tissues, and organisms using three metrics: Silhouette Coefficient Score, Adjusted Rand Index, and Calinski-Harabasz Index. Our experiments involve combinations of seven normalization methods, four dimensional reduction methods, and five integration methods. The results show that Seurat and Harmony excel in data integration, with Harmony being more time-efficient, especially for large datasets. UMAP is the most compatible dimensionality reduction method with the integration techniques, and the choice of normalization method varies depending on the integration method used.

2601.00277Jan 2026

View

MethConvTransformer: A Deep Learning Framework for Cross-Tissue Alzheimer's Disease Detection

Alzheimer's disease (AD) is a multifactorial neurodegenerative disorder characterized by progressive cognitive decline and widespread epigenetic dysregulation in the brain. DNA methylation, as a stable yet dynamic epigenetic modification, holds promise as a noninvasive biomarker for early AD detection. However, methylation signatures vary substantially across tissues and studies, limiting reproducibility and translational utility. To address these challenges, we develop MethConvTransformer, a transformer-based deep learning framework that integrates DNA methylation profiles from both brain and peripheral tissues to enable biomarker discovery. The model couples a CpG-wise linear projection with convolutional and self-attention layers to capture local and long-range dependencies among CpG sites, while incorporating subject-level covariates and tissue embeddings to disentangle shared and region-specific methylation effects. In experiments across six GEO datasets and an independent ADNI validation cohort, our model consistently outperforms conventional machine-learning baselines, achieving superior discrimination and generalization. Moreover, interpretability analyses using linear projection, SHAP, and Grad-CAM++ reveal biologically meaningful methylation patterns aligned with AD-associated pathways, including immune receptor signaling, glycosylation, lipid metabolism, and endomembrane (ER/Golgi) organization. Together, these results indicate that MethConvTransformer delivers robust, cross-tissue epigenetic biomarkers for AD while providing multi-resolution interpretability, thereby advancing reproducible methylation-based diagnostics and offering testable hypotheses on disease mechanisms.

2601.00143Jan 2026

View

Deep Learning Framework for RNA Inverse Folding with Geometric Structure Potentials

RNA's diverse biological functions stem from its structural versatility, yet accurately predicting and designing RNA sequences given a 3D conformation (inverse folding) remains a challenge. Here, I introduce a deep learning framework that integrates Geometric Vector Perceptron (GVP) layers with a Transformer architecture to enable end-to-end RNA design. I construct a dataset consisting of experimentally solved RNA 3D structures, filtered and deduplicated from the BGSU RNA list, and evaluate performance using both sequence recovery rate and TM-score to assess sequence and structural fidelity, respectively. On standard benchmarks and RNA-Puzzles, my model achieves state-of-the-art performance, with recovery and TM-scores of 0.481 and 0.332, surpassing existing methods across diverse RNA families and length scales. Masked family-level validation using Rfam annotations confirms strong generalization beyond seen families. Furthermore, inverse-folded sequences, when refolded using AlphaFold3, closely resemble native structures, highlighting the critical role of geometric features captured by GVP layers in enhancing Transformer-based RNA design.

2601.00895Dec 2025

View

INSIGHT: Spatially resolved survival modelling from routine histology crosslinked with molecular profiling reveals prognostic epithelial-immune axes in stage II/III colorectal cancer

Routine histology contains rich prognostic information in stage II/III colorectal cancer, much of which is embedded in complex spatial tissue organisation. We present INSIGHT, a graph neural network that predicts survival directly from routine histology images. Trained and cross-validated on TCGA (n=342) and SURGEN (n=336), INSIGHT produces patient-level spatially resolved risk scores. Large independent validation showed superior prognostic performance compared with pTNM staging (C-index 0.68-0.69 vs 0.44-0.58). INSIGHT spatial risk maps recapitulated canonical prognostic histopathology and identified nuclear solidity and circularity as quantitative risk correlates. Integrating spatial risk with data-driven spatial transcriptomic signatures, spatial proteomics, bulk RNA-seq, and single-cell references revealed an epithelium-immune risk manifold capturing epithelial dedifferentiation and fetal programs, myeloid-driven stromal states including $\mathrm{SPP1}^{+}$ macrophages and $\mathrm{LAMP3}^{+}$ dendritic cells, and adaptive immune dysfunction. This analysis exposed patient-specific epithelial heterogeneity, stratification within MSI-High tumours, and high-risk routes of CDX2/HNF4A loss and CEACAM5/6-associated proliferative programs, highlighting coordinated therapeutic vulnerabilities.

2512.22262Dec 2025

View

Scalable Agentic Reasoning for Designing Biologics Targeting Intrinsically Disordered Proteins

Intrinsically disordered proteins (IDPs) represent crucial therapeutic targets due to their significant role in disease -- approximately 80\% of cancer-related proteins contain long disordered regions -- but their lack of stable secondary/tertiary structures makes them "undruggable". While recent computational advances, such as diffusion models, can design high-affinity IDP binders, translating these to practical drug discovery requires autonomous systems capable of reasoning across complex conformational ensembles and orchestrating diverse computational tools at scale.To address this challenge, we designed and implemented StructBioReasoner, a scalable multi-agent system for designing biologics that can be used to target IDPs. StructBioReasoner employs a novel tournament-based reasoning framework where specialized agents compete to generate and refine therapeutic hypotheses, naturally distributing computational load for efficient exploration of the vast design space. Agents integrate domain knowledge with access to literature synthesis, AI-structure prediction, molecular simulations, and stability analysis, coordinating their execution on HPC infrastructure via an extensible federated agentic middleware, Academy. We benchmark StructBioReasoner across Der f 21 and NMNAT-2 and demonstrate that over 50\% of 787 designed and validated candidates for Der f 21 outperformed the human-designed reference binders from literature, in terms of improved binding free energy. For the more challenging NMNAT-2 protein, we identified three binding modes from 97,066 binders, including the well-studied NMNAT2:p53 interface. Thus, StructBioReasoner lays the groundwork for agentic reasoning systems for IDP therapeutic discovery on Exascale platforms.

2512.15930Dec 2025

View

Cycles Communities from the Perspective of Dendrograms and Gradient Sampling

Identifying and comparing topological features, particularly cycles, across different topological objects remains a fundamental challenge in persistent homology and topological data analysis. This work introduces a novel framework for constructing cycle communities through two complementary approaches. First, a dendrogram-based methodology leverages merge-tree algorithms to construct hierarchical representations of homology classes from persistence intervals. The Wasserstein distance on merge trees is introduced as a metric for comparing dendrograms, establishing connections to hierarchical clustering frameworks. Through simulation studies, the discriminative power of dendrogram representations for identifying cycle communities is demonstrated. Second, an extension of Stratified Gradient Sampling simultaneously learns multiple filter functions that yield cycle barycenter functions capable of faithfully reconstructing distinct sets of cycles. The set of cycles each filter function can reconstruct constitutes cycle communities that are non-overlapping and partition the space of all cycles. Together, these approaches transform the problem of cycle matching into both a hierarchical clustering and topological optimization framework, providing principled methods to identify similar topological structures both within and across groups of topological objects.

2512.12974Dec 2025

View

Multiscale Cross-Modal Mapping of Molecular, Pathologic, and Radiologic Phenotypes in Lipid-Deficient Clear Cell Renal CellCarcinoma

Clear cell renal cell carcinoma (ccRCC) exhibits extensive intratumoral heterogeneity on multiple biological scales, contributing to variable clinical outcomes and limiting the effectiveness of conventional TNM staging, which highlights the urgent need for multiscale integrative analytic frameworks. The lipid-deficient de-clear cell differentiated (DCCD) ccRCC subtype, defined by multi-omics analyses, is associated with adverse outcomes even in early-stage disease. Here, we establish a hierarchical cross-scale framework for the preoperative identification of DCCD-ccRCC. At the highest layer, cross-modal mapping transferred molecular signatures to histological and CT phenotypes, establishing a molecular-to-pathology-to-radiology supervisory bridge. Within this framework, each modality-specific model is designed to mirror the inherent hierarchical structure of tumor biology. PathoDCCD captured multi-scale microscopic features, from cellular morphology and tissue architecture to meso-regional organization. RadioDCCD integrated complementary macroscopic information by combining whole-tumor and its habitat-subregions radiomics with a 2D maximal-section heterogeneity metric. These nested models enabled integrated molecular subtype prediction and clinical risk stratification. Across five cohorts totaling 1,659 patients, PathoDCCD reliably recapitulated molecular subtypes, while RadioDCCD provided reliable preoperative prediction. The consistent predictions identified patients with the poorest clinical outcomes. This cross-scale paradigm unifies molecular biology, computational pathology, and quantitative radiology into a biologically grounded strategy for preoperative noninvasive molecular phenotyping of ccRCC.

2512.14750Dec 2025

View

2512.12272

Accurate de novo sequencing of the modified proteome with OmniNovo

Post-translational modifications (PTMs) serve as a dynamic chemical language regulating protein function, yet current proteomic methods remain blind to a vast portion of the modified proteome. Standard database search algorithms suffer from a combinatorial explosion of search spaces, limiting the identification of uncharacterized or complex modifications. Here we introduce OmniNovo, a unified deep learning framework for reference-free sequencing of unmodified and modified peptides directly from tandem mass spectra. Unlike existing tools restricted to specific modification types, OmniNovo learns universal fragmentation rules to decipher diverse PTMs within a single coherent model. By integrating a mass-constrained decoding algorithm with rigorous false discovery rate estimation, OmniNovo achieves state-of-the-art accuracy, identifying 51\% more peptides than standard approaches at a 1\% false discovery rate. Crucially, the model generalizes to biological sites unseen during training, illuminating the dark matter of the proteome and enabling unbiased comprehensive analysis of cellular regulation.

2512.12272Dec 2025

View

Graph AI generates neurological hypotheses validated in molecular, organoid, and clinical systems

Neurological diseases are the leading global cause of disability, yet most lack disease-modifying treatments. We present PROTON, a heterogeneous graph transformer that generates testable hypotheses across molecular, organoid, and clinical systems. To evaluate PROTON, we apply it to Parkinson's disease (PD), bipolar disorder (BD), and Alzheimer's disease (AD). In PD, PROTON linked genetic risk loci to genes essential for dopaminergic neuron survival and predicted pesticides toxic to patient-derived neurons, including the insecticide endosulfan, which ranked within the top 1.29% of predictions. In silico screens performed by PROTON reproduced six genome-wide $α$-synuclein experiments, including a split-ubiquitin yeast two-hybrid system (normalized enrichment score [NES] = 2.30, FDR-adjusted $p < 1 \times 10^{-4}$), an ascorbate peroxidase proximity labeling assay (NES = 2.16, FDR $< 1 \times 10^{-4}$), and a high-depth targeted exome sequencing study in 496 synucleinopathy patients (NES = 2.13, FDR $< 1 \times 10^{-4}$). In BD, PROTON predicted calcitriol as a candidate drug that reversed proteomic alterations observed in cortical organoids derived from BD patients. In AD, we evaluated PROTON predictions in health records from $n = 610,524$ patients at Mass General Brigham, confirming that five PROTON-predicted drugs were associated with reduced seven-year dementia risk (minimum hazard ratio = 0.63, 95% CI: 0.53-0.75, $p < 1 \times 10^{-7}$). PROTON generated neurological hypotheses that were evaluated across molecular, organoid, and clinical systems, defining a path for AI-driven discovery in neurological disease.

2512.13724Dec 2025

View

DCAF-Net: Dual-Channel Attentive Fusion Network for Lower Limb Motion Intention Prediction in Stroke Rehabilitation Exoskeletons

Rehabilitation exoskeletons have shown promising results in promoting recovery for stroke patients. Accurately and timely identifying the motion intentions of patients is a critical challenge in enhancing active participation during lower limb exoskeleton-assisted rehabilitation training. This paper proposes a Dual-Channel Attentive Fusion Network (DCAF-Net) that synergistically integrates pre-movement surface electromyography (sEMG) and inertial measurement unit (IMU) data for lower limb intention prediction in stroke patients. First, a dual-channel adaptive channel attention module is designed to extract discriminative features from 48 time-domain and frequency-domain features derived from bilateral gastrocnemius sEMG signals. Second, an IMU encoder combining convolutional neural network (CNN) and attention-based long short-term memory (attention-LSTM) layers is designed to decode temporal-spatial movement patterns. Third, the sEMG and IMU features are fused through concatenation to enable accurate recognition of motion intention. Extensive experiment on 11 participants (8 stroke subjects and 3 healthy subjects) demonstrate the effectiveness of DCAF-Net. It achieved a prediction accuracies of 97.19% for patients and 93.56% for healthy subjects. This study provides a viable solution for implementing intention-driven human-in-the-loop assistance control in clinical rehabilitation robotics.

2512.12184Dec 2025

View

Boosted Random Forests for Predicting Treatment Failure of Chemotherapy Regimens

Cancer patients may undergo lengthy and painful chemotherapy treatments, comprising several successive regimens or plans. Treatment inefficacy and other adverse events can lead to discontinuation (or failure) of these plans, or prematurely changing them, which results in a significant amount of physical, financial, and emotional toxicity to the patients and their families. In this work, we build treatment failure models based on the Real World Evidence (RWE) gathered from patients' profiles available in our oncology EMR/EHR system. We also describe our feature engineering pipeline, experimental methods, and valuable insights obtained about treatment failures from trained models. We report our findings on five primary cancer types with the most frequent treatment failures (or discontinuations) to build unique and novel feature vectors from the clinical notes, diagnoses, and medications that are available in our oncology EMR. After following a novel three axes - performance, complexity and explainability - design exploration framework, boosted random forests are selected because they provide a baseline accuracy of 80% and an F1 score of 75%, with reduced model complexity, thus making them more interpretable to and usable by oncologists.

2512.10995Dec 2025

View

Understanding temperature tuning in energy-based models

Generative models of complex systems often require post-hoc parameter adjustments to produce useful outputs. For example, energy-based models for protein design are sampled at an artificially low ''temperature'' to generate novel, functional sequences. This temperature tuning is a common yet poorly understood heuristic used across machine learning contexts to control the trade-off between generative fidelity and diversity. Here, we develop an interpretable, physically motivated framework to explain this phenomenon. We demonstrate that in systems with a large ''energy gap'' - separating a small fraction of meaningful states from a vast space of unrealistic states - learning from sparse data causes models to systematically overestimate high-energy state probabilities, a bias that lowering the sampling temperature corrects. More generally, we characterize how the optimal sampling temperature depends on the interplay between data size and the system's underlying energy landscape. Crucially, our results show that lowering the sampling temperature is not always desirable; we identify the conditions where \emph{raising} it results in better generative performance. Our framework thus casts post-hoc temperature tuning as a diagnostic tool that reveals properties of the true data distribution and the limits of the learned model.

2512.09152Dec 2025

View

Page 1 of 9