Wei Sun

mFrag: Integrating External DNA Methylation References Improves Lung Cancer Early Detection by Fragmentomics

mFrag improves lung cancer early detection using fragmentomics by integrating DNA methylation data and combining two models to enhance prediction accuracy.

Translating Histopathology Foundation Model Embeddings into Cellular and Molecular Features for Clinical Studies

STpath translates histopathology image embeddings into cell-type composition and gene expression estimates for clinical applications.

Immunotherapy with Pro-regenerative Macrophages from Embryonic Stem Cells Ameliorate Osteoarthritis via TNFAIP3-Mediated Chondroprotection

This study generates pro-regenerative macrophages from hESCs to treat OA, showing they reduce inflammation and restore cartilage via TNFAIP3 signaling.

Improved Deep Learning Prediction of TCR-HLA Association

The paper evaluates and improves a deep learning model for predicting TCR-HLA associations by addressing confounding factors and combining datasets.

The expression of IL-35 in the prophase of liver failure and its preliminary exploration for the mechanism of immunoregulation by IL-35 and glucocorticoids

The paper determines the optimal dose of concanavalin for a liver failure mouse model and investigates the role of IL-35 and glucocorticoids in immunoregulation and disease progression.

Exploit Spatially Resolved Transcriptomic Data to Infer Cellular Features from Pathology Imaging Data}

This paper introduces STpath, a transfer-learning model that uses spatially resolved transcriptomic data to annotate pathology images and classify tumor microenvironments, showing promising performance with limited training data.

Supervised Deep Learning with Gene Annotation for Cell Classification

The paper introduces SDAN, a deep learning method that integrates gene annotations with scRNA-seq data to improve cell classification and gene set identification, outperforming existing methods in three applications.

Neural Network Models for Sequence-Based TCR and HLA Association Prediction

The paper introduces a neural network model, DePTH, to predict TCR-HLA associations from amino acid sequences and shows its association with cancer patient survival outcomes.

ROKET: Associating Somatic Mutation with Clinical Outcomes through Kernel Regression and Optimal Transport

ROKET uses kernel regression and optimal transport to associate somatic mutations with clinical outcomes, identifying associations in three cancer types.

Individual Level Differential Expression Analysis for Single Cell RNA-seq data

The paper introduces IDEAS, a method for differential expression analysis in scRNA-seq data at the individual level, finding neurogranin-expressing neurons and ERBB signals in microglia are associated with autism.

Prioritizing candidate peptides for cancer vaccines by PEPPRMINT: a statistical model to predict peptide presentation by HLA-I proteins

The paper develops PEPPRMINT, a statistical model to predict HLA-I peptide presentation, and applies it to identify immunogenic neoantigens in melanoma patients.

Cell type-aware analysis of RNA-seq data

The paper introduces CARseq, a framework for RNA-seq differential expression analysis that accounts for cell type composition, showing higher power and accuracy in identifying differentially expressed genes in schizophrenia and autism studies.

ICeD-T Provides Accurate Estimates of Immune Cell Abundance in Tumor Samples by Allowing For Aberrant Gene Expression Patterns

ICeD-T improves immune cell abundance estimation in tumor samples by handling aberrant gene expression patterns, outperforming existing methods in simulations and real data.

Mapping Tumor-Specific Expression QTLs In Impure Tumor Samples

The paper develops a method to map tumor-specific expression QTLs in impure tumor samples using RNA-seq data from The Cancer Genome Atlas, showing it controls type I error and has higher power than alternatives.