EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

iGTP: learning interpretable cellular embedding for inferring biological mechanisms underlying single-cell transcriptomics

Kang-Lin Hsieh, Kai Zhang, Yan Chu, Lishan Yu, Xiaoyang Li, Nuo Hu, Isha Kawosa, Patrick G Pilié, Pratip K Bhattacharya, Degui Zhi, Xiaoqian Jiang, Zhongming Zhao und 1 weitere

Briefings in Bioinformatics · 2025

DOI: 10.1093/bib/bbaf296 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Kang-Lin Hsieh und 12 weiteren Autor:innen. Beteiligte Einrichtung: Division of Cancer Medicine, Department of Genitourinary Medical Oncology, UT MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, Unite…. Hauptthemen laut MEDLINE: Single-Cell Analysis, Transcriptome, Computational Biology, Gene Expression Profiling. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: NIA NIH HHS, Translational Cancer Research Training Program (BIG-TCR) funded by the Cancer Prevention & Research Institute of Texas, NLM NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep-learning models like Variational AutoEncoder have enabled low dimensional cellular embedding representation for large-scale single-cell transcriptomes and shown great flexibility in downstream tasks. However, biologically meaningful latent space is usually missing if no specific structure is designed. Here, we engineered a novel interpretable generative transcriptional program (iGTP) framework that could model the importance of transcriptional program (TP) space and protein–protein interactions (PPI) between different biological states. We demonstrated the performance of iGTP in a diverse biological context using gene ontology, canonical pathway, and different PPI curation. iGTP not only elucidated the ground truth of cellular responses but also surpassed other deep learning models and traditional bioinformatics methods in functional enrichment tasks. By integrating the latent layer with a graph neural network framework, iGTP could effectively infer cellular responses to perturbations. Lastly, we applied iGTP TP embeddings with a latent diffusion model to accurately generate cell embeddings for specific cell types and states. We anticipate that iGTP will offer insights at both PPI and TP levels and holds promise for predicting responses to novel perturbations.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Single-Cell AnalysisTranscriptomeComputational BiologyGene Expression ProfilingDeep LearningHumansProtein Interaction MapsNeural Networks, Computer

Stichwörter der Autor:innen

graph neural networkinterpretable deep learningsingle-cell transcriptomicstranscriptional programvariational AutoEncoder

Autor:innen und Institutionen

  • Division of Cancer Medicine, Department of Genitourinary Medical Oncology, UT MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, United States.
  • Center for Secure Artificial Intelligence for Healthcare, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin St, Houston, TX 77030, United States.
  • Division of Radiation Oncology, Department of Radiation Physics, UT MD Anderson Cancer Center, 1400 Pressler St, Houston, Houston, TX 77030, United States.
  • Center for Artificial Intelligence and Genome Informatics, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin St, Houston, Houston, TX 77030, United States.
  • Center for Design and Analysis, Amgen, One Amgen Center Drive, Thousand Oaks, CA 91320, United States.
  • Center for Precision Health, McWilliams School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin St, Houston, TX 77030, United States.

Förderung

  • NIA NIH HHS
  • Translational Cancer Research Training Program (BIG-TCR) funded by the Cancer Prevention & Research Institute of Texas
  • NLM NIH HHS
  • NIDCR NIH HHS
  • National Institutes of Health grants awarded to Y.D., L.Y., and Z.Z
  • Z.Z
  • National Cancer Institute's
  • Mulva Family Fund

Zitieren

APA

Hsieh, K. L., Zhang, K., Chu, Y., Yu, L., Li, X., Hu, N., Kawosa, I., Pilié, P. G., Bhattacharya, P. K., Zhi, D., Jiang, X., Zhao, Z., & Dai, Y. (2025). iGTP: learning interpretable cellular embedding for inferring biological mechanisms underlying single-cell transcriptomics. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf296

Vancouver

Hsieh KL, Zhang K, Chu Y, Yu L, Li X, Hu N, et al. iGTP: learning interpretable cellular embedding for inferring biological mechanisms underlying single-cell transcriptomics. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf296

BibTeX
@article{Hsieh2025,
  author = {Hsieh, Kang-Lin and Zhang, Kai and Chu, Yan and Yu, Lishan and Li, Xiaoyang and Hu, Nuo and Kawosa, Isha and Pilié, Patrick G and Bhattacharya, Pratip K and Zhi, Degui and Jiang, Xiaoqian and Zhao, Zhongming and Dai, Yulin},
  title = {iGTP: learning interpretable cellular embedding for inferring biological mechanisms underlying single-cell transcriptomics},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  doi = {10.1093/bib/bbaf296}
}
Kostenlos starten →