Discover, rate, and discuss biology preprints.

BertST uses a transformer-based framework to identify spatial domains in spatial transcriptomics data, outperforming GNN-based methods on Visium datasets.

The paper introduces a five-dimensional functional state space to represent disease transcriptomes as interpretable fingerprints, improving human-AI interpretation and cross-condition profiling.

CircDiscoverer provides a resource for exploring circRNA-protein interactions and RNA modifications across multiple species.

The paper uses explainable AI on a genomic foundation model to identify RBP motifs influencing translation and mRNA decay in Drosophila.

This paper presents a drug repurposing pipeline that integrates knowledge graphs and network medicine, benchmarked across diseases and applied to ALS, identifying 77 prioritized candidate drugs.

The paper introduces kmcomp, a lossless compression method for k-mer matrices that enables fast random access, achieving significant storage savings with minimal query overhead.

Eva is a foundation model for tissue imaging data that learns multi-scale spatial representations and performs various tasks like cross-modal inference and patient stratification.

The paper introduces hmSEEKR, a k-mer-based hidden Markov model that identifies non-linear sequence similarities in transcriptomes, revealing functional RNA domains with biological relevance.

Celldetective is an AI-enhanced image analysis tool for analyzing dynamic cell interactions in immunology and immunotherapy assays.

BLASE uses scRNA-seq references to analyze bulk RNA-seq data, identifying trajectory progress and correcting developmental differences in differential expression analysis.

DL4DR predicts drug response across breast cancer cell lines using a residual fusion model that combines genomic data with molecular representations, achieving high accuracy and generalization.

The paper introduces D+ and D* statistics to improve detection of gene flow by addressing saturation bias in deeper phylogenies using multiallelic site patterns.

GESSO is a spatially informed gene set activity scoring method that improves accuracy, interpretability, and scalability for spatial transcriptomics data.

PINPOINT predicts protease-inhibitory activities of fungal SSPs lacking annotated inhibitor domains using protein language models and structural modeling.

This paper develops a spatio-temporal analysis framework to uncover coupled mutagenic mechanisms in radiation-induced genomic instability using whole-genome sequencing data from irradiated human cells.

TMO learns cell-state-dependent regulatory lags between chromatin accessibility and gene expression using asymmetric cross-modal attention, achieving high Lag Concordance Scores on multi-omic datasets.

The paper designs a multi-epitope vaccine against Influenza A H10N5 and H3N2 using immunoinformatics and validates it computationally, showing activation of immune responses.

FEATMAP corrects acquisition signatures in medical foundation model embeddings to improve cross-condition similarity and task generalization without retraining.

Shredtools uses multi-MUMs to enable efficient querying of pangenome coordinates, allowing rapid extraction of gene regions across multiple genomes.

The paper develops a sequence-structure fusion model to predict antibody self-association using CSI-BLI data, achieving high performance in distinguishing high and low self-interacting clones.