Shaoheng Liang

A meta-interaction basis for cell-cell communication in tissues

SpiderNet identifies cell-cell communication programs from spatial transcriptomics data, revealing key interactions and their associations with phenotypes and therapeutic responses.

TissueNarrator: Generative Modeling of Spatial Transcriptomics with Large Language Models

This paper introduces a framework that uses large language models to generate and analyze spatial transcriptomics data, improving the simulation of cell behavior and prediction of interactions.

HEIMDALL: Disentangling tokenizer design for robust transfer in single-cell foundation models

This paper introduces HO_SCPLOWEIMDALLC_SCPLOW, a framework to analyze tokenizer design in scFMs, showing that specific tokenization choices improve generalization under distribution shift.

STEAMBOAT: Attention-based multiscale delineation of cellular interactions in tissues

The paper introduces a machine learning framework to identify multiscale cellular interactions in spatial-omics data, revealing both known and novel communication patterns across tissues.

An interpretable ML model to characterize patient-specific HLA-I antigen presentation

The paper introduces epiNB, an interpretable machine learning model for predicting patient-specific HLA-I antigen presentation using Mass Spectrometry data, achieving state-of-the-art accuracy with fewer parameters and no hyperparameter tuning.

novoRNABreak: local assembly for novel splice junction and fusion transcript detection from RNA-seq data

novoRNABreak detects novel splice junctions and fusion transcripts in cancer RNA-seq data with high sensitivity and efficiency.

Characterizing transition cells in developmental processes from scRNA-seq data

The paper models scRNA-seq data from developmental processes using SDEs to identify transition cells that explain heterogeneity and predict cell fate in early tumorigenesis.

Sensei: How many samples to tell evolution in single-cell studies?

Sensei determines required sample sizes for detecting cell type abundance changes in single-cell studies using a beta-binomial model and t-test expansion.

Stratified Test Accurately Identifies Differentially Expressed Genes Under Batch Effects in Single-Cell Data

The paper introduces the Van Elteren test to improve identification of differentially expressed genes in single-cell data by addressing batch effects, showing higher sensitivity and specificity than existing methods.

Single-Cell Manifold Preserving Feature Selection (SCMER)

SCMER identifies non-redundant molecular features that delineate both common cell lineages and rare cellular states in single-cell data.

Single-cell copy number lineage tracing enabling gene discovery

The paper introduces MEDALT and LSA to reconstruct copy number lineage trees from single-cell data and identify fitness-associated genes, showing improved accuracy over phylogenetics methods in cancer datasets.

Unbiased integration of single cell multi-omics data

bindSC integrates single-cell multi-omics data accurately by aligning rows and columns without approximations, outperforming existing methods in various biological contexts.

Batch-Corrected Distance Mitigates Temporal and Spatial Variability for Clustering and Visualization of Single-Cell Gene Expression Data

This paper introduces Batch-Corrected Distance (BCD) to improve clustering and visualization of single-cell gene expression data by mitigating batch effects, showing better performance than existing methods on longitudinal datasets.

Latent periodic process inference from single-cell RNA-seq data

Cyclum improves cell-cycle characterization in scRNA-seq data, enhancing subpopulation delineation and cell atlas construction.

Single-nuclei RNA-seq on human retinal tissue provides improved transcriptome profiling.

This study uses single-nuclei RNA-seq to profile human retinal tissue, identifying cell-type-specific gene expression and improving disease gene prioritization compared to bulk RNA-seq and mouse models.

SCMarker: ab initio marker selection for single cell transcriptome profiling

SCMarker identifies cell-type-discriminative markers in scRNA-seq data through unsupervised analysis of gene expression patterns without prior knowledge.