Wing Hung Wong

scMTG reconstructs single-cell temporal dynamics from unpaired time-series data by learning cell representations and transitions, outperforming existing methods in interpolation and transition matrix inference.

ProVADA adapts proteins to non-native contexts by steering a pretrained generator while preserving function, successfully converting renin and deimmunizing antibodies.

This study introduces epiBrainLLM, a framework using genomic LLMs to improve causal analysis of genotype-brain-clinical pathways in Alzheimer's disease.

This paper introduces a statistical framework to prioritize disease-related rare variants by integrating gene expression data, showing improved sensitivity in small sample sizes and identifying 16 rare variants in Alzheimer's disease.

The paper identifies sequence-conserved enhancer-like elements across tissues and links them to complex traits using GWAS data, finding tissue-specific heritability and novel candidate genes for BMI and schizophrenia.

scTIE integrates temporal multimodal data and infers gene regulatory relationships predictive of cellular state changes, outperforming existing methods in data integration and biological signal preservation.

EpiGePT is a transformer-based model that predicts epigenomic signals and chromatin contacts by incorporating transcription factor activities and 3D genome interactions, outperforming existing methods in cross-cell-type prediction.

The paper develops a method to detect complex structural variants across human genomes and finds they are associated with phenotypic hotspots and evolutionary changes.

SpecVar improves heritability enrichment by integrating regulatory networks with GWAS data to identify relevant tissues and shared heritability.

This paper introduces a new method to analyze free energy landscapes from MD data, showing improved performance on benchmark datasets.

DeepCAGE predicts chromatin accessible regions using deep learning, integrating sequence data and transcription factor activity, and identifies disease-associated variants.

scJoint integrates atlas-scale scRNA-seq and scATAC-seq data using transfer learning, achieving higher cell type accuracy and joint visualizations than existing methods.

The paper identifies ruminant-specific conserved non-coding elements and develops a method to interpret their regulatory roles in rumen development using gene regulatory networks.

This paper develops a method to generate neuronal cells from adult fibroblasts and uses it to study gene expression and epigenetic changes in 22q11.2 deletion syndrome.

The paper examines dynamic chromatin changes in human CAR T cells during exhaustion, identifying regulatory networks and enhancer roles in exhaustion-associated genes.

This paper introduces cfSNV, a somatic SNV caller for cfDNA that outperforms existing tools in detecting low-frequency mutations and applies it to immunotherapy prognosis in non-small-cell lung cancer.

This paper develops a Bayesian framework integrating GWAS with regulatory networks to improve the identification of trait-associated genes by modeling network topology.

This study examines the transcriptome and epigenome effects of the 15q13.3 microdeletion in human induced neurons, identifying gene expression and pathway changes linked to neuropsychiatric disorders.

NeuronMotif decodes cis-regulatory motifs from DCNNs by analyzing neuron activations, revealing motif combinations and improving model performance.

The paper introduces CTLR-Seq, a method to resolve complex genomic regions with high accuracy, and applies it to study CNVs associated with schizophrenia and autism.

scDEC uses deep generative networks to simultaneously model and cluster scATAC-seq data, showing better performance than existing tools and enabling trajectory inference and cell type discovery.

This study shows that tyrosine kinase inhibitors cause mitochondrial dysfunction in cardiomyocytes by altering GATA4-mediated networks, and GATA4 overexpression can rescue these effects.

The paper provides a comprehensive whole-genome analysis of K562, identifying genomic structural features and their impact on functional genomics data.

This paper provides a haplotype-resolved genome analysis of the HepG2 cell line, identifying genomic features and allele-specific regulation to support functional genomics studies.

DeepTACT predicts high-resolution chromatin contacts using deep learning, improving resolution over existing methods and identifying hub promoters related to disease.

The paper generates deep re-sequencing datasets for the Venter/HuRef genome to develop and benchmark genome analysis tools using Illumina sequencing.

The paper introduces coupled nonnegative matrix factorizations to integrate single cell RNA-seq and ATAC-seq data by coupling clustering across samples.

This paper introduces scABC, an R package for unsupervised clustering of scATAC-seq data to classify cell types based on open chromatin regions.

PAC-MAN enables scalable analysis of multi-sample CyTOF data by clustering and aligning subpopulations across samples to identify dataset-level cellular states.

AC-PCA reduces dimensionality and adjusts for confounding variation in biological datasets, recovering anatomical structure and shared variation across species.