858 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 study introduces StackCBEmbed, a novel ensemble machine learning model that combines traditional sequence features with pre-trained protein language model embeddings to accurately predict protein-carbohydrate binding sites at the residue level directly from primary sequences.
The paper introduces spammR, an R package designed to facilitate end-to-end analysis and integration of diverse spatial multi-omic datasets, with a specific focus on mass spectrometry-derived measurements within a single tissue sample.
This study demonstrates that the CANDO platform's multitarget drug repurposing approach successfully identifies and validates novel therapeutic candidates for schizophrenia by analyzing compound-proteome interaction signatures, revealing both established and potential new mechanisms involving neurotransmitter systems and calcium signaling pathways.
This paper demonstrates that the widely used sum-of-pairs score often fails to identify the most accurate multiple sequence alignments and proposes a deep-learning-based scoring framework, specifically Model 2, which effectively ranks alternative alignments to improve downstream phylogenetic reconstruction.
LongPolyASE is a comprehensive, end-to-end framework designed to enable allele-specific gene and isoform analysis in polyploids using long-read RNA-seq data, addressing the current lack of specialized tools for these organisms through its integrated components for syntenic gene identification, transcript quantification, and differential expression analysis.
This study demonstrates that poly(A)-enriched RNA-seq data is unsuitable for reliable circRNA detection due to high false-positive rates and tool disagreement, establishing that total RNA sequencing with effective ribosomal RNA depletion is the necessary standard for accurate circRNA profiling.
RareCapsNet is an explainable capsule network framework that robustly identifies rare cell populations and their transcriptomic signatures in large-scale single-cell RNA-seq data, outperforming state-of-the-art methods while enabling knowledge transfer across different experimental batches.
The paper introduces GCP-VQVAE, an SE(3)-equivariant geometry-complete tokenizer that converts protein backbones into a discrete 4,096-token vocabulary while preserving chirality and orientation, achieving state-of-the-art reconstruction accuracy, robust zero-shot generalization, and significantly faster inference speeds compared to prior methods.
This study reveals that while the AlphaGenome model effectively predicts chromatin accessibility variation by capturing local regulatory grammar, it significantly underperforms in predicting gene expression variation due to the challenge of modeling the long-range regulatory integration required for inter-individual differences.
MLMarker is an interpretable machine learning framework that utilizes a Random Forest model trained on healthy tissues to compute continuous similarity scores and identify tissue origins in complex proteomics data, including sparse biofluid samples and metastatic tumors.