1199 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.
OpenGerminal presents a fully open-source implementation of the Germinal antibody design pipeline that replaces proprietary components with an open stack and adopts AbLang1, achieving significantly higher cofolding pass rates and broader deployment accessibility while maintaining equivalent structural confidence metrics.
The paper argues that the stalled predictive performance in modeling peptide skin penetration stems from an ill-posed reliance on conflated transdermal flux labels that ignore delivery vehicles and endpoints, proposing instead a reframed approach that separates intrinsic barrier-crossing potential from delivery-specific retention and risk.
The paper introduces EnzymeArt, a function-conditioned generative framework that successfully converts functional descriptions into experimentally validated enzymes with quantitative catalytic activity across multiple protein families, achieving performance that often matches or exceeds wild-type and commercial references.
This paper critically reviews and proposes homology-aware cross-validation strategies to address data leakage and ensure fair generalization assessment for both classical and deep learning-based RNA secondary structure prediction methods.
Bamsnap-LRS is an automated command-line tool designed to overcome the scalability and optimization limitations of existing visualization software by enabling high-throughput, publication-ready batch visualization of long-read sequencing alignments with support for long-read-specific features, phased SNP inspection, and diverse genomic analyses.
As part of a University of Alberta capstone course, this paper introduces three deep-learning models (ProMax, APEX, and UltraMax) trained on a hybrid dataset of approximately one million protein-ligand pairs to predict quantitative glycan-protein binding strengths, while highlighting the challenges posed by long-tail data distributions and insufficient chiral feature utilization.
ComplexDesign is a hallucination-based approach that utilizes structure-prediction-guided sequence optimization and a specialized masking mechanism to successfully design multichain protein complexes and flexible binders bridging multiple targets, outperforming existing methods in both unconditional multimer design and ternary complex generation.
fastQpick is an open-source command-line tool and Python library that enables efficient, scalable bootstrap resampling of FASTQ reads to quantify uncertainty in raw sequencing data, offering multiple memory-optimized modes for handling large libraries.
HoloCell is a 860-million-parameter generative foundation model pretrained on a massive multi-omics corpus that unifies epigenomic, transcriptomic, and proteomic data through hierarchical tokenization and iterative diffusion to enable holistic cellular representation learning and flexible cross-modal generation.
This paper demonstrates that output-head designs for multi-label prediction should be explicitly chosen to reflect the underlying biological structure of comorbidities, as evidenced by a symmetric conditional random field outperforming complex alternatives in the microvascularly linked complications of Type 2 diabetes, while no single architecture proved stable for the heterogeneous electrophysiological complications of myocardial infarction.