Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 29, 2026Communications BiologyOpen Access

HarveST uses a heterogeneous graph learning framework to reveal spatial transcriptomics patterns

View Full Paper
Ask AI
Bookmark
Share

Authors

JFJunning FengHong Kong University of Science and TechnologyTYTianwei YuChinese University of Hong Kong, ShenzhenYZYanlin ZhangGCI Science & Technology (China)

Discussion

Loading...

Member takes

Implication

HarveST demonstrates improved spatial domain identification and marker gene detection in complex tissues, suggesting advanced analysis capabilities.

Key Points

  • The aim is to enhance spatial domain identification and marker gene detection in spatial transcriptomics using a novel computational approach.
  • Developed a heterogeneous graph-based framework integrating spatial and transcriptomic data.
  • Employed self-supervised learning for feature extraction and partially supervised refinement for domain identification.
  • Used Random Walk with Restart algorithm to identify spatially variable genes across different tissues.
  • HarveST successfully identified biologically meaningful spatial domains and associated marker genes.
  • The framework outperformed conventional clustering methods.
  • Facilitated joint analysis across consecutive spatial transcriptomics sections, ensuring consistent functional domain reconstruction.

Cite This Study

Feng et al. (2026) studied this question.

synapsesocial.com/papers/69c8c2d1de0f0f753b39d399https://doi.org/10.1038/s42003-026-09841-2
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Single-cell resolution analysis of the human pancreatic ductal progenitor cell niche2020 · 198 citations
  2. 2AEBP1 is a Novel Oncogene: Mechanisms of Action and Signaling Pathways2020 · 80 citations
  3. 3Differences in potential key genes and pathways between primary and radiation-associated angiosarcoma of the breast2022 · 14 citations
  4. 4Single-cell transcriptomics reveals the landscape of intra-tumoral heterogeneity and transcriptional activities of ECs in CC2021 · 146 citations