Evaluation of a deep generative computational framework under constrained conditions for multi-regional placental single-cell transcriptomic inference
Placenta · 2026-08 · 182 · 140-149
DOI: 10.1016/j.placenta.2026.06.006 ↗Kurz gefasst
Fachartikel in der Zeitschrift Placenta, 2026-08. Verfasst von Zhenjie Tang und 10 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Antibody Engineering, School of Laboratory Medicine and Biotechnology, Southern Medical University, 1838 N. Guangzhou Ave, Guangzhou, 51…. Hauptthemen laut MEDLINE: Placenta, Single-Cell Analysis, Transcriptome, Pre-Eclampsia. Die Arbeit nennt 29 Literaturangaben. Gefördert von: National Natural Science Foundation of China, FSBSI Scientific Research Institute of Neurosciences and Medicine.
Abstract (Originalsprache)
INTRODUCTION The placenta is a highly heterogeneous organ, and its dysfunction is central to preeclampsia (PE). Although single-cell RNA sequencing (scRNA-seq) enables high-resolution characterization, its high cost limits large-scale clinical use. scSemiProfiler integrates deep generative modeling with bulk RNA-seq to infer single-cell transcriptomes, but the original framework relies on an active learning module to select informative samples. In this study, we evaluate a simplified workflow that omits active learning, assessing its performance under these constrained conditions in complex, multi-regional placental tissues. METHODS We generated 14 scRNA-seq libraries from three placental regions-basal plate (BP), placental villi (PV), and chorioamniotic membranes (CAM)-obtained from 5 pregnant women (1 normotensive controls and 4 PE cases), and integrated these with 118 bulk RNA-seq datasets (detailed in Table S1). scSemiProfiler was applied without active learning, using existing scRNA-seq data as references. Semi-profiled and real-profiled data were compared in cell type composition, marker genes, and Gene Ontology (GO) enrichment. The impact of gene-count filtering was also assessed. RESULTS Semi-profiled data recapitulated major tissue-specific cell composition patterns across regions, despite deviations in some populations (especially in some rare populations). Gene expression intensity and detection rates were reduced, but canonical markers remained identifiable. Filtering cells with fewer than 10 detected genes improved compositional concordance and marker visualization, with minimal impact on GO consistency. scSemiProfiler captured pathway-level functional variation and disease-associated GO changes, but was less reliable for rare cell types and subtle gene-level differences, particularly in case-control contrasts across regions. CONCLUSIONS The simplified scSemiProfiler workflow, applied without active learning, largely preserved broad tissue-level transcriptional patterns but showed limited performance for rare cell types and disease contrasts.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Antibody Engineering, School of Laboratory Medicine and Biotechnology, Southern Medical University, 1838 N. Guangzhou Ave, Guangzhou, 510515, PR China.
- Department of Obstetrics and Gynecology, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, PR China.
- Department of Fetal Medicine and Prenatal Diagnosis, Zhujiang Hospital, Southern Medical University, Guangzhou, 510280, PR China.
- Department of Obstetrics and Gynecology, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, PR China. Electronic address: shengchao3@126.com.
- Department of Obstetrics and Gynecology, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, PR China. Electronic address: lifenxia123@qq.com.
- Institute of Antibody Engineering, School of Laboratory Medicine and Biotechnology, Southern Medical University, 1838 N. Guangzhou Ave, Guangzhou, 510515, PR China. Electronic address: yangxuexi2017@126.com.
Förderung
- National Natural Science Foundation of China
- FSBSI Scientific Research Institute of Neurosciences and Medicine
Zitieren
APA
Tang, Z., Li, K., Liu, Y., Lu, S., Wei, X., Yang, Y., Yin, H., Guo, Z., Sheng, C., Li, F., & Yang, X. (2026). Evaluation of a deep generative computational framework under constrained conditions for multi-regional placental single-cell transcriptomic inference. Placenta, 182, 140-149. https://doi.org/10.1016/j.placenta.2026.06.006
Vancouver
Tang Z, Li K, Liu Y, Lu S, Wei X, Yang Y, et al. Evaluation of a deep generative computational framework under constrained conditions for multi-regional placental single-cell transcriptomic inference. Placenta. 2026;182:140-149. doi:10.1016/j.placenta.2026.06.006
BibTeX
@article{Tang2026,
author = {Tang, Zhenjie and Li, Kun and Liu, Yuming and Lu, Shijing and Wei, Xingyu and Yang, Yuxuan and Yin, Hongtao and Guo, Zhiwei and Sheng, Chao and Li, Fenxia and Yang, Xuexi},
title = {Evaluation of a deep generative computational framework under constrained conditions for multi-regional placental single-cell transcriptomic inference},
journal = {Placenta},
year = {2026},
volume = {182},
pages = {140--149},
doi = {10.1016/j.placenta.2026.06.006}
}
EUVIMED