Xin Zhou

OncoSpot benchmarks cancer spatial transcriptomics methods across tumor evolutionary modes by simulating realistic cancer data with known subclonal and copy-number features.

The paper develops an agent-based model to show that unresolved inflammation after radiotherapy creates a tissue state that promotes tumor recurrence.

CNVeil infers haplotype-specific copy number and subclonal architecture from single-cell cancer genomes, outperforming existing methods in accuracy and resolution.

DeepMalignant identifies malignant cells in scRNA-seq data by integrating gene expression and CNA, outperforming existing methods across multiple cancer types and platforms.

This study benchmarks 20 cell type annotation methods on spatial transcriptomics data to evaluate their performance in resolving cell types, subtypes, and biological organization across diverse datasets.

This study tracks the development of response inhibition in adolescent macaques, showing that behavioral improvements and neural maturation are linked to white matter pathway development.

CSsingle improves bulk and spatial transcriptomic data decomposition by integrating with scRNA-seq, showing better accuracy and revealing MCCs as potential biomarkers in EAC.

stDyer-image improves clustering of spatial transcriptomics and proteomics data by integrating morphological images directly into the clustering process.

This study uses spatial transcriptomics to map the organization of the mouse spinal cord and identifies changes in neuronal subtypes and communication patterns in neuropathic pain.

This study evaluates long single-end barcoded reads for structural variant detection, showing improved accuracy with longer read lengths compared to short-read and pangenome-based methods.

SCGclust integrates SNVs and CNAs using a graph autoencoder for cell clustering, achieving higher accuracy than existing methods on simulated and real cancer data.

scCotag improves single-cell multi-omics integration by using co-optimal transport and regularized barycentric mapping, outperforming existing methods in alignment and embedding accuracy.

FEAST generates synthetic spatial transcriptomics data with tunable spatial and transcriptional variation for benchmarking and 3D reconstruction.

The paper introduces SpaMV, a method for integrating spatial multi-omics data that preserves modality-specific information and improves clustering and dimensionality reduction.

stDyer improves spatial domain clustering in SRT data by combining GMVAE and GATs with dynamic graphs and mini-batch sampling, outperforming existing methods.

FocalSV improves structural variant detection by combining assembly-based precision with targeted analysis efficiency using long-read sequencing data.

This study shows that neural activity patterns in the prefrontal cortex predict cognitive improvements during adolescence, with long-distance white matter tracts playing a key role.

This paper benchmarks clustering, alignment, and integration methods for spatial transcriptomics using real and simulated datasets, providing recommendations and code for reproducibility.

MaskGraphene integrates multi-slice, multi-condition spatial transcriptomics data using a graph neural network to generate interpretable joint embeddings with preserved geometric information, improving downstream analyses.

CNVeil improves tumor subclone identification and copy number estimation in single-cell DNA sequencing by addressing noise and bias through novel computational methods.

scLongTree infers subclonal trees from longitudinal scDNA-seq data, outperforming existing tools in accuracy and scalability.

The paper examines how working memory-related neural activity changes with adolescent development in non-human primates, finding that the most significant changes occur in intermediate cortical layers rather than superficial layers.

This study compares statistical and machine learning methods for predicting complex diseases using genomic variants, finding statistical methods generally outperform machine learning and ensemble models like Super Learner perform best.

This paper expands genome benchmarking to include difficult-to-map regions using linked and long reads, identifying more false negatives in variant calling.

The paper examines gamma frequency oscillations in the adolescent prefrontal cortex of rhesus monkeys during a working memory task, finding increased gamma power in adolescents compared to adults.

AquilaDeepFilter improves the precision of large deletion SV detection from linked-reads by using a deep learning ensemble, while maintaining high recall and outperforming existing methods.

MARS detects haplotype-resolved structural variants from linked-read assemblies using Mendelian inheritance and HiFi reads, achieving 73%-87% validation rates for indels.

DeepDRIM is a deep neural network that reconstructs cell-type-specific gene regulatory networks from single-cell RNA-seq data, outperforming existing methods and distinguishing functional modules in COVID-19 patients.

The paper develops a text mining pipeline to extract gene-phenotype associations from autism-related literature, creating a database that can aid in understanding autism mechanisms and treatment.

Aquila uses linked-read data to assemble diploid genomes and detect genetic variants, achieving high coverage and accurate phasing of SNPs, indels, and structural variants.