Fitness Landscape for Antibodies 2: Benchmarking Reveals That Protein AI Models Cannot Yet Consistently Predict Developability Properties [new]

CRISPR-HAWK: Haplotype- and Variant-aware guide design toolkit for CRISPR-Cas [new] Incorporates genetic variation for improved gRNA design. Population diversity analysis reveals variants affecting on-target activity.

Automated Image-Based Cell Sorting by Targeted Photopolymerization [new] Enables high-throughput cell sorting using image analysis and targeted photopolymerization to selectively remove unwanted cells.

Iterative Epitope Expansion Enables Structure-Guided de novoDesign of IgE-Binding Miniproteins Targeting the FcεRI Interface][new] Epitope Expansion: de novo mini-proteins target IgE-FcεRI via sequential binding.

BCR repertoire analysis and cloning of antibody candidates targeting native and Asp7-isomerized beta-amyloid [new] Identified anti-amyloid antibody candidates (esp. Asp7-isomerized) via BCR-seq and B cell lineage analysis.

DEEPGEOSEARCH: LLM-POWERED SCHEMALESS RETRIEVALFOR BIOMEDICAL DATA DISCOVERY [new] LLM extracts study attributes from dataset text, harmonizes terminology, and ranks datasets by contextual alignment with queries.

Moving from Association to Causation: Instrumental factor models for causal inference in high-dimensional multi-omics data [new] Causal effects in hi-dim omics: supervised IV & low-dim instrumental factors.

SuperFocus enables whole-slide cell-level spatial omics from spot-based measurements and histology across modalities [new] AI-powered single-cell spatial omics imputes whole-slide features via spot data/histology.

Improving Cryo-EM Optimization Robustness with an Optimal Transport Loss Function for Noisy Images [new] Cryo-EM optimization via Sliced Wasserstein: wider basins, robust gradients.

Ensemble-DeepSets: an interpretable deep learning framework for single-cell resolution profiling of immunological aging [new] Immune age pred. fr. cell data; reveals aging, links to SLE.

OMNI: Optimized Multi-view Network Integration with Heterogeneous Graph Attention for Biomedical Interaction Prediction [new] OMNI: Predicts biomedical relationships in networks using a novel graph attention network.

Deep learning analyses of DNA sequences resolve the retention of the Duffy-null resistance to Plasmodium vivax malaria in Africa [new]

CoPhaser: generic modeling of biological cycles in scRNA-seq with context-dependent periodic manifolds [new] scRNA-seq models cycles, separates periodic/nonperiodic variation via VAE, revealing cycle dynamics.

Fully computational design of PAM-relaxed Staphylococcus aureus Cas9 with expanded targeting capability [new] New SaCas9 KRH variant broadens genome editing via relaxed PAM (NNNRRT).

Variational Inference with Node Embeddings (VINE) for Scalable Bayesian Phylogenetics [new] Fast Bayesian phylogenetics via VINE inference.

Clearing the whey: Efficient inhibition and removal of chymosin from whey by de novo designed protein-based inhibitors [new] New proteins inhibit/remove chymosin in whey, enabling sustainable food & protease control.

To Predict is to Design: Unlocking Generative Capabilities in All-Atom Structure Predictors via Geometric Score Distillation [new] Geom score distills all-atom prot design via frozen diffusion.

SpaceTrooper: a quality control framework for imaging-basedspatial omics data [new] Data-driven QC combines expression & morphology features to identify low-quality cells from spatial omics data, improving downstream analysis.

FlashAffinity: Bridging the Accuracy-Speed Gap in Protein-Ligand Binding Affinity Prediction [new] Fast, accurate virtual screening via lightweight docking & EGNN affinity predictor.

Genome-Level Hierarchical Attention Transformer with Multi-Head Attention Weighted Sum for Broad-Spectrum Antimicrobial Resistance Prediction and Discovery of Resistance-Related Genomic Contexts [new]

A Computational Pipeline for Glioblastoma Vaccine Development: Integrating Novel Omics-Driven OIP5 Target Discovery to Create a Deep Learning-Based Immunogenicity Framework for Personalized Immunother... [new]

Sampling Function-Related Metastable States of Proteins With DASH [new] Uses deep learning & MD to sample protein's metastable states from native, revealing new conformations, allosteric sites, & activators.

A Foundational Generative Model for Cross-platform Unified Enhancement of Spatial Transcriptomics [new] Gen. model fixes spatial transcriptomics' defects via multimodal, joint opt.

Self-Supervised Missing Wedge Correction in Soft X-Ray Tomography: Towards Accurate Cellular Morphology and Volume Quantification [new]

Highly Efficient Multiplexed Genome Engineering and Clone Selection to Enable Next Generation Induced Pluripotent Stem Cell (iPSC)-based Cell Therapies [new]

Biophysically Grounded Deep Learning Improves Protein-Protein ΔΔG Prediction](new] SKEMPI2 has data leakage. Adding biophysical priors to residue representations improves mutation-induced binding affinity change prediction.

TASC: A transcriptome-driven machine learning classifier to explore molecular heterogeneity and relapse-associated programs in T-cell Acute Lymphoblastic Leukemia [new]

B-PPI: A Cross-Attention Model for Large-Scale Bacterial Protein-Protein Interaction Prediction [new] Predicts bacterial PPIs using embeddings from a protein language model and cross-attention on a large, bacteria-specific dataset.

A Hybridization-Based Novel DNA Storage Technology with Rapid Read Performance [new] Hybridization-based DNA storage offers fast read speeds, high parallelism, and encryption, addressing limitations of synthesis/sequencing methods.

Engineered translation release factor 1 suppresses disease-causing nonsense mutations and enables multi-site nonstandard amino acids incorporation [new]

How medically important antimicrobials bind to the 30S ribosomal subunit in a bacterial pathogen [new] Predicts Borrelia ribosome-antimicrobial binding variability via computation, analyzing past structures.

MAGNET: an all-in-one foundation model for cross-modal and cross-dimensional microscopic image restoration [new] Univ. micro image restor w/ multi-task,& X modal/dim capab.

A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction [new] U-Net: Representation learning, prediction, & controllable seq. gen. using genomic data.

CancerSTFormer enables multi-scale analysis of spot-resolution spatial transcriptomes and dissects the gene and immune regulatory responses of targeted therapies [new]

MOTLAB: A Weighted Multi-Omics Transfer Learning Approach to Mitigate Breast Cancer Racial Disparities [new] MOTLAB: weighted multi-omics integ., data aug., boosts transfer learning; mitigates breast cancer racial disparities.

FlashDeconv enables atlas-scale, multi-resolution spatial deconvolution via structure-preserving sketching [new] Leverage-score sampling enables fast deconvolution, preserving rare cell signals and revealing spatial architecture.

A Comparative Review of Deep Learning Methods for RNA Tertiary Structure Prediction [updated] DL models eval RNA struct pred, acc, gen, incl complexes.

Transcriptome-Conditioned Personalized De Novo Drug Generation for AML Using Metaheuristic Assembly and Target-Driven Filtering [new] De novo drug design via RNA-seq IDs AML biomarkers, generates ligands targeting hotspots for personalized leads.

Clever Hans in Chemistry: Chemist Style Signals Confound Activity Prediction on Public Benchmarks [new] Chemist Style Signals Confound Activity Prediction Models trained on public data can predict bioactivity by inferring the submitting chemist's intent.

How Much Does Protein Structure Really Help? A Case Study in Mutation-Induced Stability Prediction [new] PLMs predict mut. stability, fold prior. Geom. needed 4 atypical variants.

Integrating NMR restraints into coarse-grained simulations: toward accurate conformational ensembles of complex protein systems [new] NN-enhanced CG sim. for accurate protein ensembles via NMR.

Large generative mRNA language foundation model for efficient coding sequence generation and design with mRNA-GPT [new] mRNA-GPT: LLM gen. mRNA seqs, trains on bact., euk., & archaeal data.

Queryome: Orchestrating Retrieval, Reasoning, and Synthesis across Biomedical Literature [new] Orchestrates LLM agents for iterative, evidence-based synthesis across biomedical literature, improving reasoning on complex queries.

Predictive design of tissue-specific mammalian enhancers that function in vivo in the mouse embryo [new] DL models trained on chromatin data design synthetic enhancers active in mouse embryo tissues (heart, limb, CNS).

Mechanistic Interpretability of Fine-Tuned Protein Language Models for Nanobody Thermostability Prediction [new] Sparse autoencoders analyze fine-tuned PLMs, revealing nanobody thermostability principles & design guides.

SurfDiff: protein surface profiling for selective or broadly reactive epitope prioritisation in binder and immunogen design [new] Epitope prioritz.: structure/seq. for surface comparison & binder design.

Drug-like antibodies with low immunogenicity in human panels designed with Latent-X2 [new] Generates developable, non-immunogenic antibodies de novo. It models bound complex, sequence, and structure, achieving high affinity & low immunogenicity.

Deep Learning Classification of EEG Responses to Multi-Dimensional Transcranial Electrical Stimulation [new] Classifies EEG responses to transcranial electrical stimulation with deep learning to objectively measure brain state, bypassing sensory input.

SynCraft: Guiding Large Language Models to Predict Edit Sequences for Molecular Synthesizability Optimization [new] LLM predicts atom edits to optimize molecular synthesizability, addressing generative AI limits.

Molecular interactions underlying selective partitioning of clients into MED1 and FUS condensates [new] Charge/aromaticity guides protein partitioning in MED1/FUS condensates, IDing binding sites.