Xun Xu

BaSiCPy: Scalable and Robust Shading Correction for Optical Microscopy Images

BaSiCPy improves illumination correction in microscopy images by addressing foreground correlations and enabling automated processing.

ScienceRSI: Recursive Scientific Self-Improvement

ScienceRSI improves scientific AI through recursive self-improvement by transferring capabilities across model scales, achieving top performance on multiple benchmarks.

Spatiotemporal atlas resolves the mesenchymal-epithelial stem cell axis governing human hair follicle aging and pathology

This study creates a spatiotemporal atlas to understand mesenchymal-epithelial stem cell dynamics in human hair follicle aging and androgenetic alopecia, identifying PTN as a key regenerative factor.

Spatial Transcriptomics Reveal Developmental Dynamics of the Human Cerebral Cortex and Striatum

This study uses spatial transcriptomics and snRNA-seq to identify 25 forebrain regions and characterize dynamic changes in the human fetal cortex and striatum during late first and early second trimesters.

Single-Cell Multi-Omics Dissects Transcript Isoform and Immune Repertoire Dynamics in Human Immunosenescence

This study uses single-cell multi-omics to analyze transcript isoform and immune repertoire changes in aging human immune cells, identifying age-related molecular features in T cells and B cells.

Complete genome-derived metabolic interactions reveal impact of gut ecology on human health

This study uses complete genomes to model gut microbiome metabolic interactions, revealing ecological group dynamics linked to inflammatory bowel disease and improving disease diagnosis.

Scaling genome annotation across the eukaryotic tree of life with OrionGeno

OrionGeno annotates over 5,300 eukaryotic genomes using a phylogeny-aware deep learning model, outperforming existing methods in gene and protein structure prediction.

OmniCell: Unified Foundation Modeling of Single-Cell and Spatial Transcriptomics for Cellular and Molecular Insights

OmniCell integrates single-cell and spatial transcriptomics to model tissue context, revealing how gene programs operate in different anatomical environments.

The genetic repertoire of the deep sea: from sequence to structure and function

This study analyzes the genetic diversity of the deep sea, linking protein structures to genetic variants and identifying biotechnologically relevant features.

OneGenome-Rice (OGR): A genomic foundation model for rice

OneGenome-Rice (OGR) is a genomic foundation model for rice pre-trained on 422 genomes, outperforming existing models in 11 tasks and enabling downstream applications like trait analysis and gene expression prediction.

ExMODE: A Multi-Omics Repository for Extremophile Adaptation and Bioprospecting

ExMODE is a multi-omics database for extremophiles that integrates and standardizes data to enable comparative analysis and bioprospecting.

Single-cell transcriptomic atlas of mouse oocyte development from growth to ovulation

This study creates a single-cell transcriptomic atlas of mouse oocyte development from growth to ovulation using a new platform and integrates transcriptomic and morphological data to identify temporal shifts in gene expression.

Multimodal AI Decodes Extreme Environment Functional Dark Matter Beyond Homology

The paper develops ACCESS, a multimodal graph neural network to annotate functional dark matter proteins from extreme environments beyond homology-based methods.

Single cell resolved spatial immune repertoire unveils spatial heterogeneity of lymphoid aggregates in human immune disorders

This study introduces Stereo-XCR-seq, a method for in situ single-cell T and B cell repertoire sequencing, revealing spatial heterogeneity in lymphoid aggregates in immune disorders.

New genetic variants associated with major adverse cardiovascular events in patients with acute coronary syndromes and treated with clopidogrel and aspirin

This study identifies six new genetic variants associated with major adverse cardiovascular events in ACS patients on dual antiplatelet therapy and develops a predictive classifier for MACE.

Intra-host Variation and Evolutionary Dynamics of SARS-CoV-2 Population in COVID-19 Patients

This study characterizes intra-host SARS-CoV-2 genomic variants and their evolutionary dynamics in COVID-19 patients using deep sequencing data.

LFMD: detecting low-frequency mutations in high-depth genome sequencing data without molecular tags

LFMD improves low-frequency mutation detection in high-depth sequencing by combining duplex sequencing and bottleneck sequencing to enhance sensitivity and specificity.

Advanced Whole Genome Sequencing Using an Entirely PCR-free Massively Parallel Sequencing Workflow

This paper demonstrates that PCR-free whole genome sequencing using the DNBSEQ platform and machine learning variant calling significantly reduces InDel errors and improves variant calling performance.

PALM-Seq: simultaneous sequencing of cell-free long RNA and small RNA

PALM-Seq simultaneously sequences cell-free long and small RNA, identifying pregnancy-related RNAs and sample-specific signatures in plasma, amniotic fluid, leukocytes, and placentas.

Sequencing of the MHC region defines HLA-DQA1 as the major independent risk for anti-citrullinated protein antibodies (ACPA)-positive rheumatoid arthritis in Han population

This study identifies HLA-DQA1 as the major independent genetic risk for ACPA-positive RA in Han population through MHC region sequencing and HLA-typing.