1012 papers
Bioinformatics sits at the exciting intersection where biology meets data science, using powerful computer tools to decode the vast complexity of living systems. From mapping the human genome to tracking how viruses evolve, this field transforms raw biological information into actionable insights that drive modern medicine and research forward without requiring a supercomputer to understand the basics.
On Gist.Science, we ensure you never miss a breakthrough by processing every new preprint in this category directly from bioRxiv. Our team provides both plain-language explanations and detailed technical summaries for each paper, making cutting-edge discoveries accessible to everyone regardless of their background.
Below are the latest bioinformatics papers added from bioRxiv, ready for you to explore with clarity and depth.
This paper introduces TranScouter, a lightweight encoder-decoder framework that leverages LLM-derived gene embeddings and target condition transcriptomic profiles to competitively predict transcriptional responses to gene perturbations across both seen and unseen biological conditions.
The paper introduces SAHA, a scalable R package that accelerates cell type identification in single-cell and spatial transcriptomic studies by enabling rapid, privacy-preserving label transfer using summary statistics and user-defined hyperparameters to generate transparent, semi-automated annotation reports.
The authors present BWR-finder, a novel, database-free software tool that utilizes a parallelized BWT-based seed-and-extend algorithm to efficiently detect interspersed repeats in gigantic genomes (over 10 Gb) with improved speed and memory usage compared to existing tools while maintaining high sensitivity.
crispAIPE is a transformer-based probabilistic framework that quantifies uncertainty in prime editing outcomes by modeling competing editing events on a simplex with Dirichlet likelihood and split-conformal calibration, enabling reliable prediction of variant correction efficiency and generalizable design filters across cell types.
Although single-cell foundation models exhibit some shared structural patterns in their gene embeddings, such as ribosomal enrichment, geometric outliers are largely model-specific and do not reliably identify biologically important genes or disease-relevant targets.
RKMR is a scalable, uncertainty-aware framework that integrates nonlinear kernel modeling, spike-and-slab variable selection, and spatial dependence to identify robust, predictive marker panels from high-dimensional spatial omics data, outperforming existing methods in accuracy and interpretability.
The PG-LLM benchmark demonstrates that general-purpose language models, despite lacking access to structural data or multiple-sequence alignments, can effectively rank protein variants and outperform many established sequence-based predictors, though they still lag behind specialized biomolecular tools.
This study reveals that local ancestry inference at low genetic divergence is fundamentally constrained by a feasibility threshold and the inadequacy of current reference data rather than model architecture, demonstrating that per-site accuracy metrics mask severe structural failures and that providing richer haplotype information yields greater performance gains than architectural innovations.
This study reveals a newly discovered, evolutionarily conserved family of GFP-fold proteins spanning prokaryotes and eukaryotes that replaces the canonical fluorescent chromophore with a highly constrained, buried charge-network of two arginines, two glutamates, and a tyrosine, a motif also found convergently in the unrelated DUF2490 family.
MassGAT is an open-source, graph-based collective learning approach that integrates peak detection and annotation for LC-HRMS data by modeling ion species as a graph and utilizing a Graph Attention Network to outperform existing independent processing pipelines.