Publication | Haron Xu

My work develops machine-learning methods that respect biological structure, from support-preserving generative models for spatial transcriptomics to interpretable integration of spatial multi-omics data.

Figures advance automatically. Use the arrows or dots to explore each paper, and click a figure to inspect it.

01 Generative biology

BioFlow

BioFlow: A Biologically Valid Support-Preserving Flow for Histology-Conditioned Spatial Transcriptomics Prediction

Haoran Xu, Yang Liu, Wei Yuan, Xiao Han^*

MICCAI 2026

A support-preserving flow-matching framework that predicts spatial gene expression from histology while keeping generated values biologically valid. BioFlow improves spatial correlation, training stability, and computational efficiency without relying on post-hoc correction.

Code

BioFlow architecture and support-preserving flow trajectories

Architecture & support-preserving flow

Comparison of generative trajectories produced by spatial transcriptomics prediction models

Generative trajectory comparison

Spatial gene-expression prediction

GPU-hour efficiency and prediction performance comparison

Performance & efficiency

02 Spatial multi-omics

SpaMV

Interpretable Spatial Multi-Omics Data Integration and Dimensionality Reduction with SpaMV

Yang Liu, Kexin Ma, Haoran Xu, et al.

Nature Communications · 2026

An interpretable multi-view framework that separates shared biological signals from modality-specific information. SpaMV supports spatial-domain discovery, multi-omics topic modeling, and biologically meaningful interpretation across diverse tissue datasets.

Paper Code

SpaMV assumptions, model architecture, and applications

Framework overview

SpaMV simulation benchmarks and recovered shared and private topics

Simulation study

Spatial transcriptome and epigenome integration benchmarks

Transcriptome + epigenome

Spatial epigenomics

Transcriptome and metabolome integration in tissue

Transcriptome + metabolome

Transcriptome and proteome integration and pathway enrichment

Transcriptome + proteome

SpaMV analysis of spatial breast cancer multi-omics data

Breast-cancer tissue analysis

Bibliographic records

2026

MICCAI
BioFlow: A Biologically Valid Support-Preserving Flow for Histology-Conditioned Spatial Transcriptomics Prediction

Haoran Xu, Yang Liu, Wei Yuan, and 1 more author

In International Conference on Medical Image Computing and Computer-Assisted Intervention, 2026

Abs Bib Code

BioFlow is a support-preserving flow-matching framework for histology-conditioned spatial transcriptomics prediction. It preserves the non-negative support of gene-expression data while improving spatial correlation, training stability, and computational efficiency.
@inproceedings{xu2026bioflow, title = {{BioFlow}: A Biologically Valid Support-Preserving Flow for Histology-Conditioned Spatial Transcriptomics Prediction}, author = {Xu, Haoran and Liu, Yang and Yuan, Wei and Han, Xiao}, booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention}, year = {2026}, }

Nat Commun

Interpretable spatial multi-omics data integration and dimensionality reduction with SpaMV

Yang Liu, Kexin Ma, Haoran Xu, and 9 more authors

Nature Communications, 2026

Abs DOI Bib HTML Code

@article{liu2026spamv,
  title = {Interpretable spatial multi-omics data integration and dimensionality reduction with {SpaMV}},
  author = {Liu, Yang and Ma, Kexin and Xu, Haoran and Xu, Ke and Hu, Yunfei and Lin, Zhenhan and Lin, Jiangli and Han, Bo and Li, Shuaicheng and Lin, Zhixiang and Zhou, Xin Maizie and Zhang, Lu},
  journal = {Nature Communications},
  volume = {17},
  pages = {8213},
  year = {2026},
  doi = {10.1038/s41467-026-74718-1},
}