Discover, rate, and discuss biology preprints.

CellTFusion identifies functional multicellular states from bulk RNA-seq data by integrating cell type deconvolution and transcriptional regulatory network analysis, revealing TME programs linked to immunotherapy responses.

The paper maps olfactory receptors to odor percepts using docking energy scores, showing a direct relationship between receptors and specific smells.

This paper benchmarks registration methods for spatial transcriptomics under tissue tearing and introduces a supervised graph model, finding that tearing degrades performance across methods and that a learned model performs better in-sample but not across donors.

The paper adapts a pre-trained biological foundation model to learn perturbation-centric representations by fine-tuning it on perturbation profiles, improving classification and uncovering new biological relationships.

The paper identifies trinucleotide symmetry in G4 motifs and introduces MSI and nMSI to quantify it, showing higher symmetry in functional G4s compared to mutants and non-G4 sequences.

GTcomplex improves structural alignment of macromolecular complexes by directly aligning entire complexes rather than individual chains, achieving higher accuracy and speed.

This paper trains a deep learning model to predict gene expression from regulatory sequences in plants and shows it captures positional regulatory effects and functional context.

This paper evaluates active learning strategies for whole-cell antibiotic discovery, finding an optimal approach that significantly improves hit rates and generalizability across bacterial species.

This paper introduces atroplex, a framework for indexing and comparing isoform complexity across pan-transcriptomes, enabling cross-cohort structural analysis of alternative splicing.

VirProtRAG improves viral protein function annotation by integrating literature retrieval and evidence-based generation, introducing 32.53% new annotations and supporting sequence-inferred functions.

This paper introduces TCR-FramePose, a geometric framework for analyzing TCR-pMHC recognition, showing it captures novel structural features beyond conventional descriptors.

This paper introduces U-Pert, a framework for modeling unbalanced perturbation dynamics in single-cell data, enabling forward prediction and inverse design of cellular interventions.

The paper creates a high-confidence human methylation atlas and develops a deep learning model for methylated peptide detection.

The paper develops a framework to correct splicing metric estimation for NMD-sensitive transcripts using statistical models and linear fractional transformations.

AART improves cross-platform proteomic translation by integrating ridge regression and residual learning, enhancing reproducibility and accuracy across multiple diseases and platforms.

The paper introduces Ab-CASLR, a method for antigen-specific antibody retrieval under strict out-of-distribution conditions, achieving better performance than existing methods.

This paper introduces Classi4RPE, a computational algorithm for segmenting and classifying retinal pigment granules in FLIM data into three categories with high accuracy.

GLproxScape reconstructs spatial chromatin occupancy landscapes from genomic locus proteomics data by modeling per-region enrichments as spatial measurements.

This paper benchmarks AI-based drug-resistance prediction models for Mycobacterium tuberculosis using a large unified dataset, finding XGBoost performs best in precision-recall and F1 metrics, while neural models excel in AUROC.

The paper identifies a novel mechanism for large tandem duplications in cancer by analyzing replication fork directionality and develops a classifier to distinguish SV phenotypes based on replication features.