850 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.
The paper introduces hmSEEKR, a k-mer-based hidden Markov model that scans transcriptomes to identify non-linear, domain-level sequence similarities in long noncoding RNAs, thereby enabling the discovery of functionally related RNA domains and their associated protein interaction networks without requiring prior knowledge of sequence alignment.
The paper introduces AllTheBacteria, a comprehensive community resource that uniformly processes nearly 2.5 million public bacterial genomes to enable large-scale biological research and successfully demonstrates its utility by discovering and validating novel peptide antibiotics through AI-driven mining.
The paper introduces PHI, a user-friendly and automated Galaxy-based workflow that streamlines reproducible prophage identification, host interaction prediction, and functional characterization while generating standardized reports to lower barriers for both experts and non-experts in viral genomics research.
This paper introduces BGC-MLM, a foundation model pretrained on unlabeled biosynthetic gene cluster sequences that significantly improves the prediction of natural product properties, bioactivity, and structural similarity, thereby enhancing the efficiency of genome mining for novel natural products.
The 2026 update to Scop3P presents a major expansion of this proteomics-informed knowledgebase by integrating 152,350 uniformly reprocessed human phosphorylation sites with comprehensive structural, biophysical, evolutionary, and mutational annotations to provide a scalable, provenance-aware resource for interpreting phosphorylation in functional and disease contexts.
This study evaluates the performance of AlphaFold2, AlphaFold3, and enhanced sampling protocols in modeling antibody and TCR antigen recognition, demonstrating that while increased sampling and AlphaFold3 generally improve accuracy, success varies significantly across complex types and can be further boosted by pooling complementary models.
This paper defines and analyzes the NP-complete problem of selecting genomic blocks from multiple source genomes to optimize polygenic traits, proposing a suite of algorithms—including a certified Branch-and-Bound solver, a fast Block-Coordinate-Descent heuristic, and a semidefinite-programming relaxation—that collectively provide optimal or near-optimal solutions with theoretical guarantees and empirical validation on yeast-scale simulations.
GTcomplex is a novel, spatial indexing-powered algorithm that enables accurate and efficient holistic alignment of macromolecular complexes by directly deriving chain assignments from optimal global superpositions, outperforming existing methods in both speed and precision across diverse structural datasets.
This study demonstrates that a calibrated active learning strategy, optimized through retrospective simulations and validated in a closed-loop *Borrelia burgdorferi* screening campaign, significantly enhances experimental hit rates and enables the training of generalizable machine learning models capable of accurately predicting antibiotic activity for out-of-distribution compounds.
This paper introduces a novel method for representing sorted k-mers as virtual super-k-mer lists, implemented in the sklib tool, which achieves high-throughput set operations and significantly reduced memory usage compared to existing tools like KMC while maintaining competitive query performance.