Yi‐Juan Hu

This study benchmarks 12 methods for differential composition analysis in scRNA-seq data, finding LOCOM performs best in controlling false discovery rate and maintaining power.

This paper introduces gemGAT, a graph attention network for gene expression imputation across tissues, showing improved performance over regression-based methods on GTEx and ADNI data.

SMS improves high-dimensional mediation analysis by controlling FDR and increasing sensitivity, identifying mediators missed by existing methods in metabolomics and DNA methylation data.

TestNet infers microbial networks with FDR control, handling clustered and unclustered samples by generating null replicates that account for compositional effects and clustering.

LOCOM2 improves differential abundance analysis in microbiome studies by controlling error rates and handling challenges like relative abundance data and unbalanced designs.

This study examines how azithromycin affects microbial composition, function, and resistance genes in endocervical, vaginal, and rectal microbiomes of women with Chlamydia trachomatis infection.

The paper presents the first reference genome of the Black Redstart, assembled using PacBio HiFi sequencing and annotated with 18,609 genes, providing a resource for studying high-altitude adaptation in passerines.

MERIT improves the statistical efficiency of Monte-Carlo hypothesis testing by controlling the Monte-Carlo error rate, outperforming the GH method in sensitivity and specificity.

The paper introduces LDM-clr, an extension of the LDM method for analyzing microbiome data using centered-log-ratio transformed counts, enabling compositional differential abundance analysis with covariates.

This paper introduces Com-2seq, a method that integrates 16S and shotgun metagenomic data to improve statistical efficiency in testing differential abundance.

The paper introduces a joint model with matching and regularization to detect OTU-specific compositional trajectories predictive of disease onset in metagenomic data.

LOCOM is a logistic regression method for differential abundance testing in microbiome data that controls FDR and is robust to experimental biases.

The paper introduces a non-stochastic PERMANOVA extension for microbiome presence-absence analyses to address library size confounding without rarefaction, showing improved performance over resampling-based methods.

The paper extends PERMANOVA to test microbiome mediation effects by incorporating exposure and outcome as covariates, showing improved power and error control over existing methods.

The paper introduces an omnibus test based on LDM to analyze microbiome data across different scales, improving power for various association mechanisms.

The paper develops new statistical methods to test microbiome associations with censored survival outcomes at both community and individual taxon levels, showing improved performance over existing methods.

The paper introduces LDM-med, a new mediation analysis method for microbiome data that controls type I error and FDR, outperforming existing methods in power and flexibility.

The paper develops TReCASE, a statistical method for eQTL mapping using ASE data, showing it improves power and efficiency compared to existing methods.

The paper extends the LDM to test presence-absence associations in microbiome data while accounting for library size confounding without using rarefaction.

The paper introduces the linear decomposition model (LDM) to unify global and individual OTU hypothesis testing in microbiome studies while controlling FDR and offering flexibility for various analyses.

The paper introduces a strategy for analyzing matched-set microbiome data using PERMANOVA and LDM, showing improved performance in simulations and real-world applications.

The paper designs a genotyping array (ADPC) for African-specific genetic variation using whole genome sequences from the CAAPA consortium, improving GWAS for African ancestry populations.

This study identifies novel asthma susceptibility loci in individuals of African ancestry through genome-wide association analysis and admixture mapping.