Daifeng Wang

This study performs a single-nucleus transcriptome-wide association study to identify gene-trait associations in brain disorders, revealing cell-type-specific genetic regulation and ancestry-specific patterns.

The paper constructs a single-nucleus atlas of genetic regulation in the human brain, identifying cell-type-specific regulatory effects and disease-associated genes.

SegJointGene improves cell segmentation and spatial gene prioritization by integrating spatial gene expression with imaging data using deep learning.

This study uses single-nucleus multiomic sequencing to analyze gene expression and chromatin accessibility in the Down syndrome prefrontal cortex, revealing widespread dysregulation in metabolic, synaptic, and neuroinflammatory pathways.

CellTRIP uses multi-agent reinforcement learning to infer dynamic virtual cell environments from single-cell gene expression data, enabling simulation and prediction of cell interactions and changes over time and space.

This study identifies a NR2F-regulated gene network in Schwann cells, revealing enrichment for nonmyelinating Schwann cell genes and novel transcription factor interactions.

COSIME integrates multi-omics data to predict Alzheimer's disease phenotypes, improving accuracy and interpretability by analyzing feature interactions across data views.

The paper constructs a gene regulatory network resource across species and brain regions to study neurogenesis dynamics and their links to neurodevelopmental disorders.

The paper identifies genes and regulatory networks driving electrophysiological maturation of neurons in primate prefrontal cortex, with implications for autism.

The paper uses single-nucleus sequencing to identify oxidative stress and metabolic changes in Down syndrome basal forebrain neurons at birth.

NetREm identifies cell-type-specific transcription factor coordination for gene regulation using single-cell data and validates findings with functional genomics.

NeuroTD infers time-varying neural activity delays using adaptive sliding windows, showing longer delays in hippocampal signals during working memory tasks and gene expression correlations in mouse motor cortex data.

VAPOR disentangles co-occurring biological processes in gene expression dynamics using a variational autoencoder with transport operators, applied to single-cell and spatial transcriptomics data.

The paper identifies cooperative transcription factors in oligodendrocytes using single-cell multi-omics data and validates their regulatory roles through experimental analysis.

The paper processes single-nuclei multi-omics data from 388 human brains to identify regulatory elements and build cell-type networks, revealing changes in aging and neuropsychiatric disorders.

DeepGAMI improves phenotype prediction from multimodal data by integrating biological knowledge and imputing missing modalities, outperforming existing methods and providing interpretable insights.

MANGEM is a web app that aligns and analyzes neuronal multi-modal data using machine learning, enabling visualization and clustering of cells from different modalities.

CMOT infers missing single-cell modalities using optimal transport, outperforming existing methods and producing biologically interpretable results.

Directly-reprogrammed human neurons show reduced neurite growth ability with age, mirroring rodent developmental patterns.

The paper identifies trans-eQTLs in the human prefrontal cortex and links them to schizophrenia risk genes, showing their role in disease mechanisms.

JAMIE imputes and embeds multimodal single-cell data with partial modality matches, outperforming existing methods and providing mechanistic insights.

The paper uses single-cell multi-omics data to predict gene regulatory networks in Alzheimer's disease and control, identifying cell-type-specific dysregulation and potential drug targets.

BOMA identifies conserved and specific developmental gene expression programs between human brains and organoids at cellular resolution.

This study identifies reduced WNT signaling in Down syndrome progenitors, leading to fewer COUP-TFII+ interneurons and impaired neurogenesis.

This study shows that a short Nutlin-3 treatment in young FMR1-deficient mice prevents long-term neurogenic and cognitive deficits in adulthood.

This paper introduces deepManReg, a deep learning model that improves phenotype prediction from multi-modal data by aligning features into a common latent space and using cross-modal manifolds for regularization.

The paper develops a chemical cocktail to induce senescence in iPSC-derived neurons to better model age-related neurodegenerative disorders.

The paper uses manifold learning to align single-cell transcriptomic and electrophysiological data, revealing gene expression-electrophysiology relationships and developmental trajectories in mouse neurons.

The paper predicts gene regulatory networks from multi-omics data to link genetic risk variants with neuroimmunology and Alzheimer's disease phenotypes, identifying potential biomarkers.