Xin Jin

This paper introduces a droplet-based organoid culture method for high-throughput screening of cancer vulnerabilities, identifying novel therapeutic targets in stomach cancer.

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

Contrasting a pretrained genomic language model's attention between case and control haplotypes ranked HBB first on chromosome 11 and recovered more known β-thalassemia variants than GWAS, without per-variant tests.

The paper creates a spatiotemporal transcriptome atlas of human embryos after gastrulation, revealing organ-specific regulons and regulatory networks during development.

Genos-m is a foundation model for human-associated microbial genomes that outperforms existing models in gene-fitness prediction and community structure analysis.

The paper presents a comprehensive compendium of fully circularized human gut bacterial genomes and reveals how mobile genetic elements drive functional diversity with therapeutic implications.

The paper introduces CIMA, a multi-omics atlas of human immune cells, identifying regulatory mechanisms and genetic associations across sex, age, and genetic factors.

The paper introduces CycSim, a context-aware simulator for long-read mapping that improves parameter optimization, achieving faster mapping and better accuracy across different sequencing platforms.

ZIPcnv improves CNV detection from shallow whole-genome sequencing by addressing zero-inflation and varying CNV sizes, outperforming existing tools.

SpaSEG is a convolutional neural network model for analyzing spatially resolved transcriptomics data, showing superior reliability and efficiency in detecting spatial patterns and intercellular interactions.

The paper evaluates CycloneSEQ long-read sequencing for improving bacterial genome assemblies, showing it can produce complete circular genomes when combined with short-read data.

This study identifies 138,859 structural variants in 2,002 individuals, finding 30 associations with gut microbes and 48 with blood metabolites, including causal relationships between microbiome and metabolites.

This study uses spatial transcriptomics to show that cerebral vasculatures, particularly meninges, drive inflammation and tissue damage in Japanese encephalitis by forming a feedback loop involving immune cells and cell death mechanisms.

The paper develops VID, an ensemble machine learning model to predict virus-infected cells in scRNA-seq data, revealing biological differences between infected and uninfected cells in virus-induced cancers.

spaTrack infers spatial cell trajectories using optimal transport, revealing spatiotemporal dynamics in regeneration and tumor development.

This study uses multi-omics data to analyze immune and metabolic changes in mountain climbers during acclimatization and extreme altitude exposure.

scPAFA efficiently computes pathway activity scores and identifies multicellular disease-related pathway modules in large-scale scRNA-seq data.

This study characterizes the urinary microbiome in 1579 Chinese individuals, identifying host factors and genetic associations that influence microbial composition and urotype classification.

The paper develops methods for analyzing low-depth NIPT sequencing data to accurately estimate allele frequencies and genetic effect sizes, enabling population genetic studies.

This study identifies 13 novel genetic associations and confirms 17 existing ones between SNPs and neonatal metabolites using genome-wide association analysis of maternal low-pass genomes.

This study performs a large-scale genetic analysis of 104 pregnancy phenotypes in 39,194 Chinese women, identifying 407 trait-locus associations and uncovering novel genetic insights into pregnancy-related traits.

The paper introduces two LD estimation algorithms to create high-resolution LD grids across species, revealing evolutionary patterns and population structure.

This study conducts the first GWAS on cell-free DNA (cfDNA) features in pregnant women, identifying 84 genetic loci and revealing causal relationships between hematological indicators and cfDNA biology.

This paper introduces an improved nanopore sequencing platform called CycloneSEQ, demonstrating long-duration sequencing and high yield with applications in genomics.

This paper introduces Genetic Transformer (GeneT), an LLM-driven method that significantly reduces candidate variant numbers and improves recall in rare disease diagnosis.

This paper develops a machine-learning model using cfDNA fragmentation patterns to predict early-onset severe preeclampsia with high accuracy in early pregnancies and NIPT samples.

The paper reports genotyping and population structure findings from the China Kadoorie Biobank, revealing relatedness, consanguinity, and population movement patterns.

The paper compares single-cell transcriptomic data of lung tissues from bats, cats, tigers, and pangolins to analyze viral receptor distribution and immune responses to respiratory viruses.

This study analyzes cell-free DNA in COVID-19 patients to show its potential for disease monitoring and tissue injury assessment.

This study identifies genetic loci associated with the oral microbiome using metagenome-genome-wide association studies in over 1,915 individuals.