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 Tox21mer, a 43.5-million-parameter transformer foundation model pretrained on 2.5 million Tox21 concentration-response curves via masked-response reconstruction, which generates high-quality 768-dimensional embeddings that achieve state-of-the-art performance in predicting assay outcomes and AC50 values while enabling extrapolation to untested compounds.
This study demonstrates that in a Congo Basin subsistence community, children's DNA methylation patterns are significantly associated with parental conflict and father caregiving, linking these family dynamics to genes involved in stress, immunity, and development, thereby suggesting that the biological embedding of family environments is a universal phenomenon across diverse socio-ecological contexts.
MorphoStat is an open-source Python pipeline that automates the normalization, distribution-aware statistical testing, and multiple-testing correction of high-content imaging data to facilitate robust morphological profiling analysis.
The paper introduces ROSEATE, a novel framework that accurately predicts bacterial optimal growth temperatures by leveraging MSA Transformer-derived structural signatures from a single ubiquitous protein, adenylate kinase, enabling robust, phylogenetically generalizable, and community-level thermal inference across diverse environments.
MetaHarmonizer is a robust, fully local, and deterministic automated system for biomedical metadata harmonization that combines a multi-stage cascade with controlled vocabularies to prevent hallucinations and inflated benchmark performance, achieving state-of-the-art accuracy in schema and ontology mapping while enabling principled human-in-the-loop triage.
VLab4Mic is a simulation platform that predicts the structural resolvability of protein assemblies across various super-resolution microscopy modalities by modeling probe placement and steric constraints, thereby enabling researchers to assess experimental feasibility before conducting physical experiments.
THEOBROMA is an open database aggregating over 1.13 million natural products from 29 sources that distinguishes itself through per-compound license auditing, a three-tier classification system, and stereochemistry-aware deduplication to enable license-compliant virtual screening and isomer-specific bioactivity analysis.
The authors present gemsparcl, a highly scalable and efficient tool that clusters over 5.6 million bacterial genomes into 92,954 species-level genomic cohesive units in approximately 14 hours, thereby overcoming computational bottlenecks to enable comprehensive, reference-free analysis of prokaryotic diversity and taxonomy.
This study demonstrates that the biological meaning of protein language model embeddings is not fixed but is resolution-dependent, where different alignment levels to biological hierarchies selectively preserve distinct organizational relationships such as membership boundaries, functional groupings, or local family structures.
WitChi is a computationally efficient tool that uses empirical chi-squared testing to detect and iteratively prune compositionally biased sites from large-scale phylogenomic alignments, thereby restoring accurate phylogenetic topologies without the high computational cost of complex composition-aware models.