FFixR: a machine learning framework for accurate somatic mutation calling from FFPE RNA-seq data in cancer
Bioinformatics · 02.07.2026 · 42 (7)
DOI: 10.1093/bioinformatics/btag405 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 24.06.2026. Verfasst von Or Livne und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Cell Biology and Cancer Science, The Ruth and Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, 3525…. Hauptthemen laut MEDLINE: Machine Learning, Mutation, Neoplasms, RNA-Seq. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Israel Science Foundation, Israel Cancer Research Fund, Eric and Wendy Schmidt Fund for Strategic Innovation.
Abstract (Originalsprache)
Motivation Formalin-fixed paraffin-embedded (FFPE) tissues are widely used in clinical and research settings, yet their use for detecting somatic mutations from RNA sequencing (RNA-seq) is hindered by artefactual mutations introduced by cytosine deamination and strand-specific damage. Existing FFPE noise-filtering tools are tailored to DNA sequencing (DNA-seq) and rely on strand bias, rendering them unsuitable for RNA-seq. Here, we present FFixR, a machine learning–based framework that filters FFPE-induced artefacts from RNA-seq data without requiring matched-normal samples. Results Trained on FFPE melanoma samples with matched DNA, FFixR leverages allele-specific read counts, variant features, and mutational signature probabilities. FFixR removed up to 98% of artefactual mutations while maintaining ∼92% recall of true variants. SHAP analysis revealed key feature interactions guiding model decisions. When applied to independent cohorts, FFixR restored the correlation between RNA- and DNA-derived tumor mutational burden (R2 = 0.881) and recovered biologically meaningful mutational signatures. FFixR enables accurate somatic variant calling from FFPE RNA-seq data, expanding the utility of archival samples for research and clinical applications. Availability and implementation FFixR tool is freely available on the web at https://github.com/yizhak-lab-ccg/FFixR and https://doi.org/10.6084/m9.figshare.31998315. The repository also includes a readme file describing the inputs, outputs and the entire pipeline. The results presented here were produced using v1.0.0.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Cell Biology and Cancer Science, The Ruth and Bruce Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, 3525422, Israel.
- The Taub Faculty of Computer Science, Technion-Israel Institute of Technology, Haifa, 3200003, Israel.
ORCID-Profile: Or Livne, Keren Yizhak
Förderung
- Israel Science Foundation
- Israel Cancer Research Fund
- Eric and Wendy Schmidt Fund for Strategic Innovation
- Ruth and Bruce Rappaport Technion Integrated Cancer Center (RTICC)
Zitieren
APA
Livne, O., & Yizhak, K. (2026). FFixR: a machine learning framework for accurate somatic mutation calling from FFPE RNA-seq data in cancer. Bioinformatics, 42(7). https://doi.org/10.1093/bioinformatics/btag405
Vancouver
Livne O, Yizhak K. FFixR: a machine learning framework for accurate somatic mutation calling from FFPE RNA-seq data in cancer. Bioinformatics. 2026;42(7). doi:10.1093/bioinformatics/btag405
BibTeX
@article{Livne2026,
author = {Livne, Or and Yizhak, Keren},
title = {FFixR: a machine learning framework for accurate somatic mutation calling from FFPE RNA-seq data in cancer},
journal = {Bioinformatics},
year = {2026},
volume = {42},
number = {7},
doi = {10.1093/bioinformatics/btag405}
}
EUVIMED