FateDriver: Weakly supervised dual-view graph learning for transition-specific prioritization of candidate regulators in cellular reprogramming
- 1Ewha Womans University
Abstract
Cellular reprogramming supports regenerative medicine, yet the endogenous regulators of a specified source-to-target transition remain incompletely defined. Existing computational methods rank candidate regulators from network topology, trajectory dynamics, or cell-state labels, leaving unused the factors experimentally delivered to induce the transition. We present FateDriver, a weakly supervised dual-view graph neural network that jointly represents cells and genes from single-cell transcriptomes, treats the delivered factors as sparse gene-level labels, and ranks the remaining genes for a specified transition under a positive--unlabeled formulation. Across two OSKM datasets and one GMT dataset, FateDriver ranked first or second among six methods in recovering externally curated reference genes, exceeded a supervision-matched non-graph model despite weaker co-expression with the delivered factors, and achieved the highest recovery of protein targets of pluripotency-inducing chemical cocktails in both OSKM datasets. Holding either the source or the endpoint fixed yielded distinct candidate programs, including a DDX58--interferon axis associated with cells lacking endogenous pluripotency markers. Applying the same framework to candidate source--target pairs prioritized distinct candidate origins for two in vivo partially reprogrammed intestinal populations. FateDriver converts sparse experimental knowledge into transition-specific hypotheses for regulator prioritization from single-cell trajectories.
Related
- Two FT-like genes of the PEBP family constitute critical regulators of floral transition in Prunella vulgaris
- Transcriptional differences between repeatedly recultured and freshly induced embryogenic calli reveal candidate developmental regulators
- Mining the Pseudomonas aeruginosa genome for regulators of type III secretion system gene expression using FT-Tn-seq
- Editorial: Novel insights in insect immunity: regulators, signaling, and mechanisms
- Novel Regulators of Sugar-dependent Food Consumption in Drosophila melanogaster
BibTeX
@misc{sung2026fatedriver,
title = {{FateDriver: Weakly supervised dual-view graph learning for transition-specific prioritization of candidate regulators in cellular reprogramming}},
author = {J. Sung and K. Son and H.-J. Cha and W. Kim and H. Lee},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.27.762150},
url = {https://www.biorxiv.org/content/10.64898/2026.10.27.762150v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - FateDriver: Weakly supervised dual-view graph learning for transition-specific prioritization of candidate regulators in cellular reprogramming
AU - J. Sung
AU - K. Son
AU - H.-J. Cha
AU - W. Kim
AU - H. Lee
PY - 2026
DA - 2026/10/27
PB - bioRxiv
DO - 10.64898/2026.10.27.762150
UR - https://www.biorxiv.org/content/10.64898/2026.10.27.762150v1
AB - Cellular reprogramming supports regenerative medicine, yet the endogenous regulators of a specified source-to-target transition remain incompletely defined. Existing computational methods rank candidate regulators from network topology, trajectory dynamics, or cell-state labels, leaving unused the factors experimentally delivered to induce the transition. We present FateDriver, a weakly supervised dual-view graph neural network that jointly represents cells and genes from single-cell transcriptomes, treats the delivered factors as sparse gene-level labels, and ranks the remaining genes for a specified transition under a positive--unlabeled formulation. Across two OSKM datasets and one GMT dataset, FateDriver ranked first or second among six methods in recovering externally curated reference genes, exceeded a supervision-matched non-graph model despite weaker co-expression with the delivered factors, and achieved the highest recovery of protein targets of pluripotency-inducing chemical cocktails in both OSKM datasets. Holding either the source or the endpoint fixed yielded distinct candidate programs, including a DDX58--interferon axis associated with cells lacking endogenous pluripotency markers. Applying the same framework to candidate source--target pairs prioritized distinct candidate origins for two in vivo partially reprogrammed intestinal populations. FateDriver converts sparse experimental knowledge into transition-specific hypotheses for regulator prioritization from single-cell trajectories.
ER -