Search articles by title, topic or section.
Khabre desk
Recent reporting, analysis and explainers from the bioinformatics desk.
science
A bioRxiv preprint introduces LangAAI, a model that uses antibody and antigen sequences to model antibody hypervariable regions without requiring 3D structures. The authors report that changing the antigen input to a non-binding decoy particularly affects sequence recovery in the CDR3 loop.
science
A bioRxiv preprint describes Carbon-A, a 1.2-billion-parameter model for predicting protein-coding genes across eukaryotes. Its authors used the model to create a database covering 48,167 genome assemblies and report strong performance across 42 benchmarks.
science
A bioRxiv preprint introduces hEVO, a genomic model trained on 4.784 billion haplotype sequences reconstructed from whole-genome data. Its authors report strong results across variant-interpretation benchmarks, including tests that required no task-specific training.
science
A bioRxiv preprint introduces Chaperone, a framework for coordinating protein-design tools. Its authors report reduced token use and GPU costs across nine design tasks, alongside results from several protein-design campaigns.
science
A bioRxiv preprint reports that simulated mixtures of E. coli strains made sequence-type and serotype prediction less reliable than detection of individual resistance and virulence genes. Increasing the dominant strain’s abundance improved strain-level predictions, while longer reads alone did not resolve the challenge.
science
A bioRxiv preprint reports that correcting ambient RNA before detecting doublets can leave more doublet-containing droplets in single-cell RNA-sequencing data. The authors recommend detecting doublets first, except when using CellBender.
health
A bioRxiv preprint describes NeoOmni Atlas, a database built by reanalysing mass-spectrometry data from 26 cancer types. It catalogues HLA-presented peptides and computationally prioritises potential tumour-associated and tumour-specific targets.
science
A bioRxiv preprint describes DeepMM, a method that detects suspected misassembly junctions and splits contigs before microbial genomes are grouped into bins. In benchmark datasets, this reduced gene annotations that disagreed with sample-matched reference genomes.
science
A bioRxiv preprint benchmarked 16 tools using 30× Illumina whole-genome sequencing and found that only five reliably produced exact deletion, insertion and duplication calls against truth sets. It also reported accuracy losses in ensemble pipelines compared with their strongest component tools.
science
A bioRxiv preprint combines data from eight healthy Indian cohorts into a reference atlas and proposes a four-part Indian Gut Health Index. The authors report geographic patterns in the data, while noting that cohort and geography effects are strongly intertwined.
science
A bioRxiv preprint reports that oversampling closely related prokaryotic strains makes genome collections less informative than their raw counts suggest. Applying the authors’ PhyloThin method to 402 species, the study estimates that the effective sample count is 41% below the actual count.
science
A bioRxiv preprint introduces AuditPPI, a framework for examining the signals behind protein-interaction predictions. Its authors report that benchmark performance can reflect cellular context and other cues, with limited evidence that the predictive signals specifically capture interaction interfaces.
health
A bioRxiv preprint reports a ClinVar-derived AlphaGenome Variant Impact quantile cutoff of 0.9938 for prioritising single-nucleotide variants. Nearly half of the uncertain variants assessed exceeded that threshold, but the result is a prioritisation signal, not a clinical reclassification.
science
A bioRxiv preprint describes ProxiNet, a framework that learns cellular proximity from spatial transcriptomics and applies those relationships to dissociated single-cell RNA sequencing. The authors report spatial validation and use the approach to identify inferred neighborhoods in an Alzheimer dataset.
science
A bioRxiv preprint describes BEACON, a method that learns background count patterns to assign single cells to their samples in multiplexed sequencing. The authors report improved classification across human datasets and applications in cancer-cell experiments and protein–transcriptome profiling.
science
A bioRxiv preprint describes PRIORITI, a framework that combines language-model synthesis of human genetic evidence with supervised learning to forecast whether a proposed drug target–disease pairing will succeed. The authors report results on historical and post-cutoff benchmark sets.
science
A bioRxiv preprint describes DS3dRNA, a method that scores interactions among groups of RNA structural elements to design sequences for target structures. The authors report improved benchmark performance and experimental activity in designed Mango II and twister RNAs.
science
A bioRxiv preprint describes MCseg, a no-code tool whose segmentation workflow was selected through AI-agent-guided searches. In a lung adenocarcinoma dataset, its fixed-parameter workflow scored higher on panoptic quality than an Optuna-tuned Cellpose baseline.
science
A bioRxiv preprint reports that RELAX produced different likelihood estimates in repeated runs, with some changes reversing statistical conclusions. The authors link the instability to site-class models and show that a seed alone did not ensure repeatable results in their tests.
science
A bioRxiv preprint introduces ProFinder, a computational tool that ranks candidate constitutive promoters in bacterial genomes. The authors report generating candidates for 99.76% of 56,742 genomes spanning 149 phyla.
technology
Anthropic says WHO Africa teams are using Claude to compile Ebola situation reports in under an hour and to support outbreak modelling, vaccine-development coordination and viral-genome analysis in the Democratic Republic of the Congo. The company describes these tools as support for work led by public-health and scientific teams.
science
A bioRxiv preprint presents stEDGE, a computational framework that uses local tissue boundaries to reconstruct nested spatial domains and transition interfaces. The authors report results across 15 tissue sections and an application to cell-resolved profiling of fibrotic human lung.
science
A bioRxiv preprint introduces SPECTRA, a graph-based model that predicts gene-expression responses to CRISPR perturbations by propagating signals through gene regulatory networks. The authors report improved recovery of sparse perturbation-related changes in a large-scale single-cell CRISPRi benchmark.
science
A bioRxiv preprint introduces ARCHER-LD, a multi-GPU tool for calculating linkage disequilibrium across very large sequencing datasets. The authors report speedups of about eightfold to 28-fold over PLINK in benchmark tests and complete a 10-quadrillion-element LD matrix in under six hours on 512 GPUs.
science
A bioRxiv preprint finds that the reproducibility of functional assays can cap measured performance for variant-effect predictors. Correcting for assay reliability shifts the apparent prediction gap from canonical splice sites toward intronic regions 11–50 base pairs from the splice boundary.
science
A bioRxiv preprint describes miRstring, a biogenesis-aware RNA language model trained on 77,708 miRNA precursors from 414 species. The authors report cross-species mature-miRNA boundary decoding and artificial miRNA scaffold designs that repressed target mRNAs in validation experiments.
science
A bioRxiv preprint presents Z-Hunt-DP, a dynamic-programming reformulation of a thermodynamic Z-DNA prediction method. The authors report that it preserved the original minimum-energy result while producing a 3.30 × 10^5 speedup at 20 dinucleotide positions.
science
A bioRxiv preprint presents GAVURD, a trio whole-genome analysis system for finding potentially pathogenic variants outside protein-coding regions. In a proof-of-concept involving 10 unresolved rare-disease cases, it prioritised six potentially causal variants for further study.
science
A bioRxiv preprint presents Sma3s v3, a scalable workflow for assigning functions to proteins in proteomes, pangenomes and metagenomes. In tests on a Vibrio cholerae pangenome and a metagenomic catalogue, it reported higher annotation coverage than the cited comparator tools.
science
A bioRxiv preprint presents LongPhase-TO, a long-read method for reconstructing somatic haplotypes from tumour samples without matched normal DNA. In tests on eight datasets from six cancer cell lines, the method lengthened haplotype blocks and improved somatic variant-calling scores.
science
A bioRxiv preprint presents Rsearch, an R package that integrates VSEARCH with visualisation, parameter optimisation and conversion tools for metabarcoding analysis. In a mock-community comparison with DADA2, the two pipelines produced abundance profiles that closely matched the expected composition, with differences in OTU counts and computational performance.
science
A bioRxiv preprint reports that incomplete reference profiles can leave bulk RNA deconvolution targets non-identifiable even when an algorithm returns precise estimates. The authors propose measuring coverage and false-certification risk, and estimate the measurement precision needed for a decisive result in peripheral blood mononuclear cells.
science
A UCSF-led bioRxiv preprint introduces CELL-FM, a bidirectional generative model that connects protein sequences and cellular context with fluorescence microscopy images. The authors use it for virtual localization, motif analysis and mutagenesis of intrinsically disordered peptides.
science
A University of Michigan team has introduced AmbientMapper, a computational method that estimates ambient chromatin contamination in combinatorial single-cell experiments. In a bioRxiv preprint, the authors report a 0.19% wrong-genome rate on a 26-genome reference panel and recovery of low-depth nuclei that standard pipelines discarded.
science
A bioRxiv preprint presents POME, a self-supervised model that represents incomplete mixed-type biomedical data as a graph. The authors report improved missing-data imputation and useful sample representations across three biomedical datasets.
science
A University of Gothenburg preprint models glycan motifs as a containment hierarchy rather than independent features. Across 45 glycomics datasets, the authors report more estimated true positives and identify motif redistributions that conventional marginal analysis misses.
science
Researchers at Shandong University and BGI Research describe DiffGSP, a computational framework that models mRNA diffusion to reconstruct spatial gene-expression patterns. The work is a bioRxiv preprint evaluated across diverse spatial-transcriptomics datasets.
science
A bioRxiv preprint finds that randomly dividing single-cell data can leak information between training and test sets when related cells share labels. The authors propose eakcheck, a tool that estimates this leakage before a predictive model is trained.
health
A medRxiv preprint introduces EMERGENe, a phylogenetic framework that estimates how transmissible bacterial traits are acquired and expand. Tests using simulations and 3,745 Shigella sonnei genomes detected known resistance trends and additional emerging traits.
science
A bioRxiv preprint presents PHACTn, a training-free method that uses mammalian evolutionary history to infer how tolerable single-nucleotide variants may be across the human genome. The authors report strong performance on benchmark sets of non-coding variants using four interpretable parameters and no GPU.
science
A bioRxiv preprint introduces LucaCell, a foundation model that represents genes through mRNA sequence embeddings instead of fixed gene identifiers. The authors report applications spanning cross-species cell annotation, microbial analysis, chromatin accessibility and influenza infection states.
science
A bioRxiv preprint presents ForceFlowAb, a mixture-of-experts flow-matching model that designs antibody CDR sequences and structures with force-field guidance. The authors report more favourable antibody–antigen interaction-energy results than two comparison methods in CDR-H3 and six-CDR design tasks.
science
A bioRxiv preprint reports that the protein language model ESM-2 retains meaningful biological information in intrinsically disordered protein regions, despite assigning them less attention. The authors also derive structural features of these proteins from the model’s internal signals.
science
A bioRxiv preprint comparing matched full-UDG and non-UDG libraries from two medieval Mongolian individuals found trade-offs between site retention, residual disagreement and kinship inference. Corrected mixed-library data supported a first-degree, likely mother–son relationship, while uncorrected data shifted one estimate towards second degree.
science
A bioRxiv preprint from Queen Mary University of London presents a genome-wide framework that uses native Oxford Nanopore reads to measure molecule-level DNA methylation variation and co-methylation. In human lymphoblastoid cell data, the analysis found patterns hidden by site-level averages.
science
A bioRxiv preprint describes Isocall, a deterministic method for jointly identifying known and novel transcripts across large PacBio long-read RNA-sequencing datasets. The authors report a joint call across 206 samples in 25 minutes after alignment and profiling.
science
A bioRxiv preprint introduces RADIS, a chemical-probing strategy that maps RNA secondary structure across long, intron-rich genomic loci. The study applies it to splice sites, branchpoints and Alu elements, including deep intronic regions that are difficult to analyse experimentally.