AI summary · may contain errors
Spatial transcriptomics enables mechanistically grounded inference of cell-cell communication by preserving physical cell organization, improving ligand-receptor modeling despite persistent analytical caveats.
- 2 cites
Oxford University Press · Genomics & Bioinformatics · ISSN 1467-5463, 1477-4054
Rebuilt
AI summary · may contain errors
Spatial transcriptomics enables mechanistically grounded inference of cell-cell communication by preserving physical cell organization, improving ligand-receptor modeling despite persistent analytical caveats.
AI summary · may contain errors
DeOPUS, a reference-based bulk RNA-seq deconvolution method across 122 tissues and 12 organ systems, achieves mean Pearson r = 0.82 and Spearman ρ = 0.77.
Newest first · last 60 days
AI summary · may contain errors
For antigen-specific nearest-neighbor retrieval using CDR3 sequences, classical sequence comparison (BLOSUM62, Levenshtein) beat general ESM2 protein language models by 7–12 percentage points across four B- and T-cell receptor databases.
AI summary · may contain errors
Enzymatic repair during DNA preparation and the computational tool FFPErase most effectively reduce formalin-induced cytosine deamination artifacts in FFPE whole-genome/exome sequencing.
AI summary · may contain errors
Koinetic, a phylogeny-aware dissimilarity for microbiome communities, is Euclidean-valid and integrates with kernel methods, improving power for phylogenetically localized prevalence shifts in simulations.
AI summary · may contain errors
ESGI demultiplexes and processes single-cell sequencing data with arbitrary barcode architectures, matching barcodes with insertions and deletions via Levenshtein distance.
AI summary · may contain errors
ASAOT improves spatial deconvolution in mouse brain spatial transcriptomics by using structure-enhanced asymmetric optimal transport with a differentiated sharpening term to reduce diffusion and boundary distortion.
AI summary · may contain errors
HGMetaADR, a heterogeneous graph meta-learning model for ADR prediction, consistently outperforms state-of-the-art baselines across few-shot, zero-shot, and majority-to scenarios.
AI summary · may contain errors
GLAlign integrates heterogeneous single-cell multi-omics by learning a shared latent space that jointly uses global manifold constraints and local marker-gene overlap, enabling strong modality mixing across PBMC, mouse brain, CITE-seq, and NCI-N87 datasets.
AI summary · may contain errors
scGSL, a graph neural network for tumor scRNA-seq, achieves 84.83% average accuracy (F1 80.92%) for cross-patient cell type annotation across 49,024 cells from 19 patients.
AI summary · may contain errors
CRAFT enables pMHC-conditioned de novo design of antigen-specific TCR CDR3β sequences under physicochemical constraints, achieving competitive naturalness with favorable binding and predicted structural plausibility.
AI summary · may contain errors
scYOU jointly learns cell and protein embeddings from proteomic data, using supercell-constrained alignment and GO-regularized protein priors to improve clustering and embedding separability.
AI summary · may contain errors
GPU-accelerated quantum simulation enables quantum algorithms to be developed, validated, and benchmarked on classical hardware at 30–40 noiseless qubits, bridging toward quantum-accurate molecular modeling.
AI summary · may contain errors
MulDR, a structured multi-view relational learning framework for drug response prediction, reduces root mean square error by 6.07% versus the strongest baseline under the same evaluation protocol.
AI summary · may contain errors
CalDyn, a MATLAB-based automated workflow, detects calcium and electrophysiological events with 94.4% sensitivity, 89.9% precision, and an F1 score of 0.92 across patient-derived recordings.
AI summary · may contain errors
SpatialDG achieves the higher three-seed mean ARI on all 12 slices in a matched DLPFC reevaluation using a reference-label-independent checkpoint-selection rule.
AI summary · may contain errors
MRGN infers causal networks between molecular phenotypes in genomic trios under PMR while accounting for many confounders, improving inference accuracy and robustness.
AI summary · may contain errors
MAGE identifies statistically significant maximal Age Index gene sets at genome scale by using a time-varying, network-based Age Index metric that captures gene influence changes.
AI summary · may contain errors
HiCP2GAN enhances sparse, low-resolution Hi-C contact maps using a pretrained Vision Transformer discriminator trained on 118 million Hi-C patches across species and cell types.
AI summary · may contain errors
DMG-PoE delivers the highest average C-index for multi-omics cancer survival prediction on the SurvBoard benchmark across 10 TCGA cancer types, even with modality incompleteness.
AI summary · may contain errors
Transformer-based large language models are advancing genomic analysis by improving sequence tokenization, self-attention, and pre-training/fine-tuning, with impact spanning sequence, population genetics, and biodiversity research.
AI summary · may contain errors
ADInteractome identifies cell-type-specific, AD-associated gene interaction circuits from single-nucleus RNA-seq by extracting neural interaction states from a classifier trained on AD versus non-AD donors.
AI summary · may contain errors
CausalMut improves prediction of enzyme mutational effects on catalytic kinetics across three datasets, reducing root mean square error by 16.24% and increasing R² by 22.19% versus five baselines.
AI summary · may contain errors
Transformer-based models in NLP and bioinformatics are difficult to interpret because of their nonlinear complexity, motivating explainability methods to clarify performance and decisions.
AI summary · may contain errors
MTDKF predicts miRNA–drug–disease triplet resistance (MRDR) and sensitivity (MRDS) associations by fusing metapath topology with domain knowledge, consistently outperforming prior methods.
AI summary · may contain errors
spOT-NMF uses optimal transport-based non-negative matrix factorization to rank first among reference-free deconvolution methods across five gold-standard spatial transcriptomics datasets, especially under xenograft signal imbalance.
AI summary · may contain errors
IsoPrism predicts within-gene isoform proportions from short-read single-cell and spatial transcriptomes, achieving the highest mean dominant isoform accuracy and compositional similarity across five transfer scenarios.
AI summary · may contain errors
Cross-species Gene Ontology transfer enrichment across six non-model eukaryotes shows set-level fidelity declines monotonically with phylogenetic distance, reaching chance at clade boundaries in some panels (ID50 ~40%-59%).
AI summary · may contain errors
Hierarchical interactome representation learning linked HELLP syndrome to both late-onset and early-onset preeclampsia, while chorioamnionitis was linked only to late-onset disease.
AI summary · may contain errors
Interaction-finder recovers up to 4.3× as many known biological entity associations across 60 literature topics, while verifying every extracted quote against source text.
AI summary · may contain errors
Surrogate-guided self-play Monte Carlo tree search produced 16 antimicrobial peptide candidates, with AMP2, AMP7, and AMP9 achieving complete inhibition of at least one strain (18.75%) in a 37-strain panel.
AI summary · may contain errors
Training data composition strongly shapes cross-dataset generalization of deep learning RNA 3D structure quality predictors, with RNA3DQA giving the best overall accuracy and ranking stability for ARES.
Journals this month
Topics moving, week to 18 Nov 2026
By field