FFPERescuer: deep unsupervised domain adaptation for the reconstruction of gene expression profiles derived from formalin-fixed paraffin-embedded samples
- 1Department of Surgery, The Chinese University of Hong Kong, Shatin, Hong Kong SAR, China
Abstract
Formalin-fixed paraffin-embedded (FFPE) tumor tissues often suffer from RNA degradation, posing a long-standing challenge for reliable transcriptomic profiling. Here, we propose FFPERescuer, a deep learning framework employing unsupervised domain adaptation, to rectify distorted gene expression data. FFPERescuer comprises a partial encoder that maps a small subset of genes to high-level representations and a decoder to reconstruct full gene expression profiles. On simulated data with varying noise levels, FFPERescuer faithfully recovered gene expression profiles, achieving high Pearson correlation coefficients (PCCs > 0.85) with the ground truth. In FF-FFPE-matched cohorts, FFPERescuer significantly enhanced expression profile concordance, with average PCCs increased by 23% (P < 0.05). Applying to cancer subtyping, FFPERescuer improved classification accuracy from 67% to 92%, recapitulated subtype-specific biological properties lost in the FFPE-derived data, and enhanced survival associations. Our studies provide a powerful framework for reliable transcriptomic profiling from FFPE-archived tumor samples that are widely available in the clinic.
Related
- Top2All: Benchmarking sparse-to-transcriptome reconstruction from high-expression profiles across sex- and sex-steroid-defined conditions
- Unsupervised decomposition of massively parallel TraDIS-Xpress fitness data from Escherichia coli identifies distinct functional and expression modules of antimicrobial resistance determinants
- Biopsy-aware phylogenetic reconstruction from longitudinal single-cell copy number profiles
- A deep generative decoder predicts microRNA expression from bulk and single-cell mRNA profiles
- Gene expression dynamics through time reveal low pleiotropy genes underlying local adaptation in silver birch (Betula pendula)
BibTeX
@misc{he2026ffperescuer,
title = {{FFPERescuer: deep unsupervised domain adaptation for the reconstruction of gene expression profiles derived from formalin-fixed paraffin-embedded samples}},
author = {l. He and K. Song and Y. Li and Y. Dong and C. Y. N. Wong and L. Qi and X. Zhang and K. Lenos and T. d. Back and C. Elbers and C. Xu and R. M. H. Leung and R. Deng and Y. Zhang and S. Qiao and F. Gao and Y. Chen and S. S.-M. Ng and S. Zhou and L. Vermeulen and X. Wang},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.02.748790},
url = {https://www.biorxiv.org/content/10.64898/2026.09.02.748790v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - FFPERescuer: deep unsupervised domain adaptation for the reconstruction of gene expression profiles derived from formalin-fixed paraffin-embedded samples
AU - l. He
AU - K. Song
AU - Y. Li
AU - Y. Dong
AU - C. Y. N. Wong
AU - L. Qi
AU - X. Zhang
AU - K. Lenos
AU - T. d. Back
AU - C. Elbers
AU - C. Xu
AU - R. M. H. Leung
AU - R. Deng
AU - Y. Zhang
AU - S. Qiao
AU - F. Gao
AU - Y. Chen
AU - S. S.-M. Ng
AU - S. Zhou
AU - L. Vermeulen
AU - X. Wang
PY - 2026
DA - 2026/09/02
PB - bioRxiv
DO - 10.64898/2026.09.02.748790
UR - https://www.biorxiv.org/content/10.64898/2026.09.02.748790v1
AB - Formalin-fixed paraffin-embedded (FFPE) tumor tissues often suffer from RNA degradation, posing a long-standing challenge for reliable transcriptomic profiling. Here, we propose FFPERescuer, a deep learning framework employing unsupervised domain adaptation, to rectify distorted gene expression data. FFPERescuer comprises a partial encoder that maps a small subset of genes to high-level representations and a decoder to reconstruct full gene expression profiles. On simulated data with varying noise levels, FFPERescuer faithfully recovered gene expression profiles, achieving high Pearson correlation coefficients (PCCs > 0.85) with the ground truth. In FF-FFPE-matched cohorts, FFPERescuer significantly enhanced expression profile concordance, with average PCCs increased by 23% (P < 0.05). Applying to cancer subtyping, FFPERescuer improved classification accuracy from 67% to 92%, recapitulated subtype-specific biological properties lost in the FFPE-derived data, and enhanced survival associations. Our studies provide a powerful framework for reliable transcriptomic profiling from FFPE-archived tumor samples that are widely available in the clinic.
ER -