Biology subjects

Yan, Q.

Publications and source records attributed to Yan, Q..

At least 19 recordsLinked to original sources

bioRxiv2026

METTL3-mediated m6A modification alleviates skeletal muscle atrophy in ALS by regulating ACSL1-dependent triglyceride metabolism

Skeletal muscle-intrinsic defects are increasingly recognized as critical contributors to the pathogenesis of amyotrophic lateral sclerosis (ALS), yet the underlying molecular mechanisms remain poorly understood. N6-methyladenosine (m6A) is the most abundant post-transcriptional RNA modification in mammalian mRNAs that plays important regulatory roles in diverse physiological and pathological processes, however its role in ALS-associated skeletal muscle atrophy remains unknown. Here we found that global m6A methylation and the expression of its methyltransferase METTL3 are reduced in the skeletal muscle of hSOD1G93A transgenic ALS mice. Multi-omics analyses reveal that m6A dysregulation is closely associated with impaired triglyceride metabolism. Functionally, METTL3 promotes triglyceride accumulation in ALS skeletal muscle in an m6A-dependent manner; overexpression of METTL3 in ALS myotubes improves energy supply and alleviates muscle atrophy by promoting triglyceride accumulation. In vivo overexpression of Mettl3 in skeletal muscle attenuates muscle atrophy, improves motor performance, and prolongs survival of ALS mice. Mechanistically, we identify the fatty acid activation enzyme ACSL1 as a key downstream effector of METTL3-mediated regulation of triglyceride metabolism. METTL3-mediated m6A modification promotes the expression of Acsl1 by enhancing its mRNA stability. Further investigation revealed that the regulatory effect of m6A on Acsl1 expression depends on the m6A reader protein IGF2BP2. Collectively, our findings establish m6A modification as a critical regulator of skeletal muscle atrophy in ALS by maintaining triglyceride metabolic homeostasis, highlighting a promising therapeutic target for ALS.

neuroscience↗

bioRxiv2026

Sequence-specific and programmable catalytic degradation of extracellular microRNAs by membrane-anchored DNA nanomachines

Extracellular microRNAs (miRNAs) contribute to metastasis through communication between tumour cells and their microenvironment, yet their sustained, sequence-selective depletion remains challenging. Here we report Programmable RNA-Intercepting Molecular Eraser (PRIME), a membrane-anchored DNA nanomachine that organizes miRNA recognition and catalytic degradation at the cell surface. A double-tetrahedral DNA scaffold positions a programmable recognition hairpin beside RNase H. Target binding generates an RNA/DNA hybrid for RNase H-mediated cleavage, and product release resets the hairpin for subsequent degradation cycles. Membrane-anchored PRIME depleted matched extracellular miRNAs by 70.6-89.4% across lung, breast and cervical cancer cell models. Multi-target PRIME (M-PRIME) inhibited migration by 71.3-96.0% in Transwell assays across these models. To evaluate its therapeutic potential, we established an orthotopic lung-tumour model and an early-intervention breast-cancer lung-colonization model in mice. Aptamer-directed M-PRIME combined with gemcitabine reduced pulmonary tumour burden by 74.5% in established lung tumours and 93.2% in the lung-colonization model. These effects were consistent with PRIME-mediated extracellular miRNA degradation, with circulating target miRNA levels reduced by 77.5-98.7% relative to gemcitabine alone. Together, these findings establish a programmable material strategy for sustained extracellular RNA depletion and support its potential to enhance chemotherapy in preclinical tumour models.

biochemistry↗

bioRxiv2026

Conserved microglial programmes, non-transferable classifiers: boundaries of the single-cell TBI molecular clock

Background: Traumatic brain injury (TBI) pathology evolves over hours to months, and molecular staging could complement clinical and imaging-based assessment. Single-cell transcriptomics captures cell-type-resolved temporal programs, but whether a transcriptional "molecular clock" trained on one injury model can be transferred to others remains unknown. Methods: We compiled four public mouse TBI single-cell/single-nucleus datasets spanning two injury models - controlled cortical impact (GSE277487; CEREBRI) and fluid percussion injury (GSE160763) - plus an independent mild-FPI cohort reserved for validation (GSE247339), totalling 307,991 cells across four pathological stages (control, acute, subacute, chronic). Classifiers were evaluated at cell and sample levels (held-out libraries) against a permutation null, with per-cell-type, strict leave-one-dataset-out (LODO) generalization, glial-only transfer and gene-level SHAP analyses. Results: Within GSE277487, sample-level balanced accuracy reached 0.67 (permutation null 0.32 +/- 0.09; p = 0.001), and the signal was reproduced in the independent mild-FPI cohort (0.40 vs null 0.19 +/- 0.05; p = 0.0010). Cell-level cross-validation overestimated performance by ~0.16, quantifying within-sample leakage. Cross-dataset generalization was poor (strict LODO balanced accuracy 0.20-0.31); the acute immediate-early response was untransferable (recall 0.002) and 80% of chronic (6-month) cells were assigned to the subacute class. Glial-only models raised subacute recall across models (0.91 and 0.75), but precision was indistinguishable from a subacute-only classifier (0.51 vs 0.50; 0.58 vs 0.58) and one-vs-rest AUROC was at or below chance (0.38-0.54) across four classifier families - classifier collapse, not transferable discrimination. The reproducible cross-model component was instead a gene-expression programme: a microglial complement/lysosomal damage programme (Trem2-Apoe axis) was up-regulated across the severe CCI and FPI models (13/13 concordant genes), whereas the immediate-early module was model-specific. Conclusions: A single-cell molecular clock for TBI staging is feasible within datasets once within-sample leakage is controlled, but it does not transfer across datasets as a classifier: apparent cross-model transfer reflected classifier collapse onto the majority class, and the reproducible cross-model component was a microglial complement/lysosomal gene-expression programme. These findings define the boundaries of cross-dataset time inference in TBI and provide evaluation criteria for claims of transferable single-cell classifiers.

neuroscience↗

bioRxiv2026

AnnoAudit: a marker-based protocol for auditing single-cell atlas annotations reveals annotation-driven artifacts in a widely used traumatic brain injury resource

Single-cell atlas annotations are routinely treated as ground truth but rarely validated before use. We present AnnoAudit, a marker-based audit protocol that combines four convergent checks - marker scoring, unsupervised clustering, margin-gated module scoring, and applicability-gated pretrained models - into a composite Annotation Contamination Score (ACS), plus an independent-gene discrimination step that distinguishes genuine mis-assignment from ambient-RNA detection artifacts, and a trajectory-correlation fingerprint tracing suspicious signals to their cell type of origin. Applied to CEREBRI (GSE269748), a widely used single-cell TBI atlas (73 citations; 65 citing works; six re-analysis studies, none of which validated the annotations), the audit shows that contamination is systematic: the official "glutamatergic neuron" label is 97.8% non-excitatory by markers, with at least 67.4% of its cells independently confirmed as non-neuronal (microglia-dominated); the GABAergic (46.6%), OPC (66.1%), and astrocyte (37.7%) labels are likewise contaminated by marker-confirmed microglia, whereas the microglial, oligodendrocyte, endothelial, and pericyte labels are largely clean (85-96% self-confirmed). The contamination is state-dependent: astrocyte and OPC labels are largely correct in uninjured controls (76.0% OPC self-marker; 18.5% discrimination-confirmed contamination for the astrocyte label) but collapse in the acute 24-h window (14-16% self-marker; 57-71% microglia), recovering by 6 months - so the annotation failure concentrates precisely in the window of maximal injury response. This temporal structure generates coherent false signals - a biphasic trajectory for 40 of 307 ion-channel genes, a KCNC3-specific OXPHOS signature, and an inverted KCNC3 trajectory at 7 days - whereas the corrected response is a sustained acute KCNC3 up-regulation conserved across three independent datasets and three injury models. Simulation-calibrated ACS is 82.6% for CEREBRI. In a human ALS atlas (GSE330130), an ambient-aware re-analysis shows that the official neuron labels are largely correct (1.8-6.1% contamination after discrimination), confirming that the audit does not flag well-annotated resources, and that marker-only estimates on snRNA data with high ambient RNA must be interpreted with the discrimination step. AnnoAudit needs only the deposited count matrix and a canonical panel; we propose it as a routine quality step for atlas-based re-analysis.

bioinformatics↗

bioRxiv2026

Distillation enables scalable high-fidelity virtual screening across ultra-large chemical libraries

Accurate virtual screening of ultra-large chemical libraries remains challenging. Existing approaches rely on lower-fidelity scoring functions or sampling-based strategies that can limit predictive accuracy and bias the exploration of chemical space. Here, we present FastBindRank, a distillation-based framework that transfers the predictive power of the structure-based model Boltz-2 into an efficient sequence-based surrogate. Trained on [~]1% of the 122-million-compound PubChem library, FastBindRank enables high-fidelity screening at scale. Applied to histone deacetylase 11 (HDAC11), FastBindRank substantially enriched high-confidence binders relative to the background chemical space. The lightweight model captured structural patterns associated with predicted binding, revealing structural determinants of binding. Under a comparable computational budget, FastBindRank achieved a 74-fold increase in hit rate and over a 30-fold increase in discovery yield over direct subset-based screening. Experimental validation confirmed the activity of two novel compounds. These results establish distillation as a practical strategy for scalable, high-fidelity virtual screening of ultra-large chemical libraries.

bioinformatics↗

bioRxiv2026

Discovery of ARQ-501 as a potent PafA inhibitor that elevates KatG levels to potentiate isoniazid in mycobacteria

The prokaryotic ubiquitin-like protein (Pup) conjugation system (PPS), which is essential for Mycobacterium tuberculosis (Mtb) virulence but absent in humans, presents an attractive drug target. Here, we report the discovery of ARQ-501, a quinone-based, covalent, substrate-competitive inhibitor of the Pup ligase PafA. ARQ-501 exhibited potent anti-mycobacterial activity against Mtb under host-mimicking stress conditions and within macrophages. We further identified the catalase-peroxidase KatG, essential for activation of the frontline prodrug isoniazid (INH), as a pupylation substrate. ARQ-501 inhibits KatG pupylation, causing its accumulation and creating a selective synergy with INH. This quantity over quality mechanism successfully rescued INH activation by the clinically prevalent KatG S315T mutant in enzymatic assays and enhanced INH efficacy against clinical S315T isolates to variable degrees. This work identifies a novel class of PafA inhibitors and a previously unrecognized role of pupylation in regulating KatG, offering a potential therapeutic avenue to combat drug-resistant tuberculosis.

microbiology↗

bioRxiv2026

Assessing and Optimizing Low-Frequency Somatic Mutation Detection: A Multi-Platform High-Throughput Sequencing Perspective

The availability of multiple commercial short-read sequencing platforms necessitates systematic cross-platform performance comparisons, particularly for challenging applications such as low-frequency somatic mutation detection. Here, a large-scale targeted sequencing dataset from five Genome in a Bottle (GIAB) human genomic DNA reference standards, HG001 to HG005, alongside Twist Biosciences cfDNA reference standards featuring 1% variant allele frequency (VAF), was generated by six platforms (NovaSeq 6000, NovaSeq X, FASTASeq 300, GenoLab M, SURFSeq 5000, and MGISEQ-T7). To build a realistic benchmark while keeping authentic sequencing backgrounds, we developed PosMix, a simulating tool that generates position-specific VAFs. To overcome the limitations of conventional variant callers (high recall with poor precision for VarScan2, higher precision with lower recall for Strelka2/Mutect2), we developed SomaticXGB, a machine learning-based caller. In this study, SURFSeq 5000 consistently exhibited the lowest error rates and achieved superior accuracy for VAFs as low as 0.5%, outperforming all other sequencing platforms. On the other hand, SomaticXGB attained F1 scores of approximately 0.92 on simulated datasets with VAFs ranging from 0.5% to 1.5% and 0.89 on Twist 1% standards, substantially outperforming conventional methods. This work delivers a valuable rich multi-platform data resource, offering a standardized pipeline for performance benchmarking and a machine learning-based strategy for optimized somatic mutation detection.

bioinformatics↗

bioRxiv2026

CMS: Achieving Uniform and High-Quality Sequencing across Challenging Non-canonical Genomic Regions

High-throughput sequencing is essential in modern biological research, yet low-complexity sequences remain challenging as they form structurally complex, non-canonical (non-B) DNA conformations that impede sequencing enzyme read-through. This leads to a long-standing trade-off: maximizing coverage introduces false positives (FP), while stringent filtering causes coverage loss and false negatives (FN). To address this, we developed CMS (Cross Mountains and Seas) on GeneMind sequencing platforms by optimizing its chemistry and enzymatic systems to traverse these secondary structures with high fidelity. Benchmarking across whole-genome (WGS) and whole-exome (WES) sequencing demonstrates that CMS addresses the trade-off by simultaneously enhancing both coverage uniformity and accuracy, notably achieving an approximately 100-fold reduction in low-coverage bins for WGS and a 70% reduction in FN insertions/deletions (INDELs) within complex non-B regions. Specifically, a synthetic G-quadruplex (G4) motif sequencing experiment demonstrates that CMS maintains a 1:1 strand ratio, effectively handling G4-induced biases where benchmarked platforms exhibit extensive depletion. These findings establish CMS as a reliable technology for the precise characterization of structural-challenging but functional-essential genome regions.

genomics↗

bioRxiv2026

Rhizoxin and 2,4-diacetylphloroglucinol Contribute to Biocontrol of Pseudomonas protegens Pf-5 Against Pea Ascochyta Blight Pathogen Didymella pinodes

Biological control using beneficial bacteria is a promising strategy for managing pea Ascochyta blight (AB), yet the underlying mechanisms remain poorly understood. In this study, we identified ten bacterial strains from four genera, including Bacillus, Paenibacillus, Peribacillus, and Pseudomonas, that significantly reduced the severity of AB caused by Didymella pinodes under greenhouse conditions. Most strains inhibited D. pinodes in vitro, suggesting antibiosis as a primary mode of action. To further elucidate the biocontrol mechanisms, we used Pseudomonas protegens Pf-5, which produces eight known antimicrobial compounds, as a model. While wild-type Pf-5 strongly inhibited D. pinodes in cultures and controlled AB in planta, a derivative ({Delta}8-fold mutant) lacking all eight compounds showed significantly compromised biocontrol efficacy. Individual complementation of biosynthetic genes for rhizoxin, 2,4-diacetylphloroglucinol (DAPG), pyrrolnitrin, or hydrogen cyanide partially restored inhibitory activity, confirming their roles in inhibition of D. pinodes. Notably, restoring rhizoxin and DAPG biosynthesis recovered the disease control capability of the {Delta}8-fold mutant in greenhouse trials. These results demonstrate that rhizoxin and DAPG are key metabolites driving the biocontrol activity of P. protegens against D. pinodes. SIGNIFICANCEAn advanced understanding of how beneficial bacteria control plant diseases can help us better use these microorganisms in agriculture. In this study, beneficial bacteria isolated from pea roots and soils effectively mitigated damages of pea Ascochyta blight caused by the fungal pathogen Didymella pinodes. Most of the identified beneficial bacteria inhibited the fungal pathogen in cultures, indicating antimicrobial compounds were likely produced by the bacteria to control the disease. Using the soil beneficial bacterium Pseudomonas protegens Pf-5 as a model, we demonstrated that four bacteria-derived antimicrobial compounds, rhizoxin and 2,4-diacetylphloroglucinol (DAPG), pyrrolnitrin, and hydrogen cyanide play important roles in inhibiting D. pinodes growth. This study also showed that rhizoxin and DAPG produced by Pf-5 contribute to the suppression of AB development. These findings provided new insights into the molecular basis of beneficial bacteria-mediated disease suppression of pea Ascochyta blight.

microbiology↗

bioRxiv2026

A light-labile signal molecule acts as a photoregulator of secondary metabolite biosynthesis in a heterotrophic bacterium

While light is essential for many phototrophic organisms, our knowledge of the light-dependent regulation in heterotrophic organisms is scarce. Here, we found that the heterotrophic soil bacterium Pseudomonas protegens differentially accumulated 234 secreted metabolites depending on the light conditions. These metabolites included important antimicrobials such as pyoluteorin, 2,4-diacetylphloroglucinol, pyrrolnitrin, and rhizoxin analogs. Pyoluteorin, for instance, was 140-fold more abundant in the dark, due to a strong upregulation of the transcript of the pyoluteorin biosynthesis gene pltL. We discovered that 2,4-dichlorophloroglucinol (PG-Cl2), an activator of pyoluteorin biosynthesis, acts as a photoregulator that degrades much faster in the presence of light than in the dark. The resulting higher PG-Cl2 levels activate the pltL promoter in the dark. These findings reveal a novel mode of regulation in which light instability of a signal molecule allows the producing organism to sense and transmit light information to regulate the biosynthesis of a secondary metabolite.

microbiology↗

bioRxiv2026

A cross-species single-cell epigenome kidney atlas identifies epithelial cells as a driver of epigenetic aging

Epigenetic aging is a hallmark of chronic diseases, arising from sustained injuries and unresolved repairs. To investigate cell-type-specific epigenetic alterations, we built a cross-species single-cell multi-omics atlas of DNA methylomes, chromatin accessibilities, and transcriptomes on healthy, injured (human) and aging (mouse) kidneys. We identified accelerated epigenetic aging dominated by tubular epithelia in diseased kidneys. The pathological state mirrors transcriptional trajectories observed in normal aging, driven by the preferential dysregulation of lineage-specific genes lacking CpG islands. Spatially, these epigenetic changes mapped to pathological niches of failed repair. Co-profiling of single-cell DNA methylation and 3D genome architecture revealed that epithelial repair states in disease undergo significant higher-order genome reorganizations, activating genes associated with renal decline. Our findings demonstrate that epithelial aging is driven by a collapse of 3D chromatin structure and local methylome integrity, which silences cell identity and promotes a non-resolving repair state.

genomics↗

bioRxiv2026

Tumor Cell Clustering Enhances Metastatic Competence by Regulating the H3K36 Histone Demethylase KDM2A

Disseminated tumor cells can form clusters via cell-cell adhesion, which increases their capacity to initiate metastasis. Metastatic clusters are characterized by distinct changes in transcription, suggesting that epigenetic mechanisms underlie their unique phenotypic state. By performing functional epigenomic studies in models of non-small cell lung cancer, we identified the histone H3 lysine 36 (H3K36) demethylase KDM2A as being differentially required for the fitness of metastatic cell clusters. This contextual dependency on KDM2A is predicated by tumor cell-cell aggregation, which specifically induces KDM2A binding to CpG island enriched promoters. At these defined genomic loci, KDM2A maintains H3K36 monomethylation, which preferentially correlates with transcriptional activation. KDM2A directly targets oxidative phosphorylation genes and KDM2A activity is required for optimal mitochondrial respiration and apical cell junction integrity in cell clusters. Consequently, suppressing KDM2A reduces metastatic seeding and colonization in multiple organs, including in the brain. These findings reveal a chromatin regulatory mechanism by which homotypic cell communication instructs the epigenome of disseminated tumor cells to potentiate their metastatic competence.

cancer biology↗

bioRxiv2026

Optimizing broadly neutralizing antibodies via all-atom interaction modeling and pre-trained language models

Antibody optimization is a fundamental challenge, and the identification of antibody-antigen interactions is crucial in the optimization process. However, current methods cannot accurately predict antibody-antigen interactions, providing limited functional guidance to improve the time-consuming and costly traditional optimization techniques. Here, we present InterAb and InterAb-Opt, a unified computational framework that integrates all-atom modeling with antibody language models to predict antibody-antigen interactions and enable antibody optimization. Leveraging the proposed all-atom modeling approach, AtomInter, and pre-trained antibody language models, InterAb outperforms existing methods in predicting antibody specificity and antibody-antigen binding affinity. InterAb successfully identified influenza A virus-binding antibodies from an antibody library and accurately detected high-affinity antibodies in the AIntibody competition. Empowered by the robust functional insights from InterAb, InterAb-Opt was developed to optimize broadly neutralizing antibodies. For R1-32 antibody, biolayer interferometry results reveal that 85%, 80%, 90%, and 67.5% of the 40 InterAb-Opt-optimized antibodies exhibit enhanced binding affinities to wild-type SARS-CoV-2, Lambda, BQ.1.1, and EG.5.1, respectively, with a maximum improvement of up to 96-fold. For the newly emerging BA.2.86 and KP.3, 55% and 52.5% of the optimized antibodies notably transition from non-binding to binding. Neutralization assays demonstrated that the optimized antibodies exhibited enhanced neutralization activity across multiple targets, highlighting the capability of InterAb-Opt in engineering broadly neutralizing antibodies. This technology enables precise analysis of antibody-antigen interactions and optimization of broadly neutralizing antibodies, holding promise for addressing challenges in immune evasion and vaccine design.

bioinformatics↗

bioRxiv2025

AI Identifies Broadly Neutralizing Antibodies from an IGHV1-69 Public Antibody Class Exerting Continued Selection Over SARS-CoV-2

R1-32-like public antibodies, characterized by shared IGHV1-69/IGLV1-40 usage, are elicited in more than 50% of individuals with COVID-19 and have been implicated in driving recurrent mutations at L452SARS2 and F490SARS2 within their convergent epitope in the SARS-CoV-2 spike receptor-binding domain. These mutations effectively mediate escape from non-affinity-matured R1-32-like antibodies with germline-like sequences. Here, we characterize four affinity-matured R1-32-like antibodies, C092, C807, BD56-104, and BD56-597, that tolerate L452SARS2 and F490SARS2 mutations. We show that this tolerance arises from residues introduced by somatic hypermutation at convergent positions across multiple CDR loops and surrounding regions, thereby creating additional contacts that reinforce epitope binding. An unusual N354SARS2 glycosylation site, which emerged in BA.2.86 and became fixed in its descendants, is linked to escape from affinity-matured R1-32-like antibodies, implying ongoing selection by this public antibody class. Using an AI model trained on integrated structural, neutralization, and binding data, we further identified ZL525, an ultrapotent R1-32-like antibody with pan-SARS-CoV-2 variant activity, including against the highly evasive KP.3 variant carrying the N354SARS2 glycosylation, and broad sarbecovirus cross-reactivity extending to SARS-CoV-1. Together, these findings show how affinity maturation enables public antibodies to adapt to viral antigenic drift, reveal their role in shaping SARS-CoV-2 antigenic evolution, and demonstrate the potential of AI-empowered strategies for discovering broadly neutralizing antibodies.

immunology↗

bioRxiv2025

Closing the Sim-to-Real Gap: An End-to-End Robotic Ultrasound System Leveraging In Vivo Reinforcement Learning and 3D-Prior Guided Hybrid Control

Abdominal ultrasound is a crucial first-line diagnostic tool, yet its efficacy is inherently constrained by a strong dependency on operator skill, leading to significant inter-operator variability and limiting its widespread adoption. While robotic ultrasound systems with artificial intelligence have emerged to mitigate this, current approaches face critical limitations: they often rely on simulated or phantom data for training, which hampers generalizability, and employ control strategies that lack the flexibility to adapt to complex human anatomy. To address these challenges, this paper introduces a novel, end-to-end robotic system for autonomous abdominal scanning. Our methodology is systematic: it begins with coarse scan planning via RGB and 3D point cloud-based localization of key anatomical landmarks (e.g., xiphoid process). The core of our approach involves training a reinforcement learning scanning policy directly on live human volunteers, enabling the development of a strategy that is robust to anatomical diversity. This high-level policy is executed by a sophisticated hybrid force-position controller, enhanced with real-time normal vector calibration and 3D point cloud-based respiratory phase detection to dynamically adjust contact force, particularly during inspiration. This innovation proves critical for improving scan quality in subjects with higher Body Mass Index (BMI). Extensive validation against expert sonographers demonstrates that our system achieves high precision in standard plane recognition and superior scanning efficiency. Furthermore, to holistically assess performance where standard planes are difficult to acquire, we introduce a 3D reconstruction-based coverage metric. Results confirm that our system delivers strong adaptability and significantly improved consistency, marking a substantial step toward clinically viable, operator-independent ultrasound robotics.

bioengineering↗

bioRxiv2025

AI recognizes convergent somatic hypermutation signatures to allow the discovery of variant-resilient broadly neutralizing antibodies

Elucidating the mechanisms by which broadly neutralizing antibodies (bnAbs) develop to confer durable immunity in humans is pivotal for the rational design of next-generation vaccines and therapeutics. VH3-53/3-66-encoded public antibodies are widely elicited in the population after the COVID-19 pandemic. Here, we isolated 15 VH3-53/3-66-encoded bnAbs from elite neutralizers who experienced sequential SARS-CoV-2 Omicron breakthrough infections. These bnAbs exhibit exceptional potency and breadth against circulating variants and, as members of a long-lived public antibody lineage, likely contribute to durable humoral protection. Genetic analysis of VH3-53/3-66-encoded bnAbs reveals a convergent somatic hypermutation pattern at seven positions in and around the CDR loops that emerges with repeated viral exposures and distinguishes them from non-bnAbs. Grafting these combined mutations onto clinically escaped antibodies broadens their breadth and restores neutralization activity. Structural analysis shows that these convergent mutations cooperatively remodel CDRs to bind and tolerate virus receptor-binding domain mutations. Using an AI-based antibody language model trained on data linking somatic hypermutations to binding affinity and neutralization, the model is able to identify mutational patterns predictive of breadth, enabling the discovery of a rare bnAb with protective activity against the latest variants from early-pandemic antibody repertoires. This demonstrates the feasibility of establishing an AI-empowered pipeline to identify mutation-tolerant bnAbs from early-pandemic antibody repertoires to fight fast-evolving newly introduced viruses.

immunology↗

bioRxiv2025

The Janus face of Selinexor (KPT-330) in bacterial infection: short-term protection versus long-term lethality

The clinical application of the anti-cancer drug KPT-330 (Selinexor) has been associated with increased infection-related mortality, yet its direct impact on infection pathogenesis remains unclear. Here, we found that KPT-330 treatment reduced host survival from 66% to 22% in a mouse model of bacterial pneumonia. Interestingly, KPT-330 conferred transient lung protection at 24 h post infection by blocking the nuclear export of I{kappa}B, suppressing NF{kappa}B signaling and immune cell recruitment. Meanwhile, KPT-330 induced apoptosis of lung epithelial cells, enhancing bacterial adhesions and invasion, and leading to dramatically increased bacterial load. These effects culminated in a compensatory immune rebound and more severe lung injury at later stages. Adjunctive therapy with the antibiotic polymyxin E (colistin) rescued survival, whereas the immunosuppressant dexamethasone did not, underscoring that timely bacterial clearance is critical for managing this adverse effect. Our findings provide direct evidence that KPT-330 exacerbates bacterial infection and advocate for an adjunctive antimicrobial prophylaxis strategy to ensure its safer use, especially in high-risk patients.

microbiology↗

bioRxiv2025

A Generative Foundation Model for Antibody Design

Antibodies are indispensable components of the immune system, yet the design of high-affinity antibodies remains a time-consuming and experimentally intensive process. To address this challenge, we present IgGM, a novel generative foundation model designed to accelerate high-affinity antibody engineering. IgGM learns the complex relationships underlying the binding interactions between antigens and antibodies, as well as the mapping between antibody sequences and structures. By conditioning on different inputs, IgGM supports a wide range of antibody design tasks, including complex structure prediction, inverse design, affinity maturation, framework optimization, humanization, and de novo antibody design. It is compatible with both conventional antibodies and nanobodies, and allows user-defined CDR loop lengths for flexible design. To prioritize candidates, we introduce a frequency-based computational screening strategy that enhances design efficiency. Extensive evaluation through both in silico benchmarks and in vitro experiments across diverse antigens such as PD-L1, Protein A, TNF-, IL-33, SARS-CoV-2 RBD and its variants demonstrates that IgGM consistently generates antibodies or nanobodies with high measured affinity. These results underscore IgGMs versatility and effectiveness as a powerful tool for next-generation antibody discovery and optimization.

bioinformatics↗