Yidong Chen

This paper introduces DECENT, a deep learning method to reconstruct embryonic copy number variations and reduce maternal DNA contamination in cfDNA from SECM, improving the accuracy of noninvasive preimplantation genetic testing.

Spatial-Live provides a lightweight tool for visualizing single-cell spatial-omics data in 3D with interactive features and data integration.

The paper performs transcriptomic analysis of human dental biopsies to identify gene expression differences between symptomatic and asymptomatic patients and between sexes.

DepLink links genetic and pharmacologic dependencies in cancer by integrating CRISPR, pharmacologic, and expression data through an R Shiny app.

CancerSiamese improves cancer type prediction using one-shot learning with gene expression data, achieving 8% and 4% accuracy gains over 1-NN for primary and metastatic tumors.

The paper introduces MR-GAN, a GAN-based model for predicting epitranscriptome modification sites using unsupervised learning, showing significant improvement over one-hot encoding and SRAMP.

This paper designs an ancestry informative marker panel using whole exome sequencing data to estimate genetic ancestry in Hispanic populations, achieving high accuracy in admixture proportion estimation.

This study shows that deleting Lats1 and Lats2 in adult mouse pancreatic acinar cells causes fibrosis and inflammation through YAP1/TAZ activation.

This paper introduces Deep-m6A, a deep learning model for predicting m6A sites and identifies 709 functionally significant m6A-regulated genes and nine subnetworks associated with cancer and other diseases.