N-of-one differential gene expression without control samples using a deep generative model

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

N-of-one differential gene expression without control samples using a deep generative model. / Prada-Luengo, Iñigo; Schuster, Viktoria; Liang, Yuhu; Terkelsen, Thilde; Sora, Valentina; Krogh, Anders.

I: Genome Biology, Bind 24, Nr. 1, 263, 2023.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Prada-Luengo, I, Schuster, V, Liang, Y, Terkelsen, T, Sora, V & Krogh, A 2023, 'N-of-one differential gene expression without control samples using a deep generative model', Genome Biology, bind 24, nr. 1, 263. https://doi.org/10.1186/s13059-023-03104-7

APA

Prada-Luengo, I., Schuster, V., Liang, Y., Terkelsen, T., Sora, V., & Krogh, A. (2023). N-of-one differential gene expression without control samples using a deep generative model. Genome Biology, 24(1), [263]. https://doi.org/10.1186/s13059-023-03104-7

Vancouver

Prada-Luengo I, Schuster V, Liang Y, Terkelsen T, Sora V, Krogh A. N-of-one differential gene expression without control samples using a deep generative model. Genome Biology. 2023;24(1). 263. https://doi.org/10.1186/s13059-023-03104-7

Author

Prada-Luengo, Iñigo ; Schuster, Viktoria ; Liang, Yuhu ; Terkelsen, Thilde ; Sora, Valentina ; Krogh, Anders. / N-of-one differential gene expression without control samples using a deep generative model. I: Genome Biology. 2023 ; Bind 24, Nr. 1.

Bibtex

@article{c87472916d4349ea9a7d24d70e31a635,
title = "N-of-one differential gene expression without control samples using a deep generative model",
abstract = "Differential analysis of bulk RNA-seq data often suffers from lack of good controls. Here, we present a generative model that replaces controls, trained solely on healthy tissues. The unsupervised model learns a low-dimensional representation and can identify the closest normal representation for a given disease sample. This enables control-free, single-sample differential expression analysis. In breast cancer, we demonstrate how our approach selects marker genes and outperforms a state-of-the-art method. Furthermore, significant genes identified by the model are enriched in driver genes across cancers. Our results show that the in silico closest normal provides a more favorable comparison than control samples.",
keywords = "Deep generative models, Deep learning, DEG, DEseq2, Differential expression analysis, Transcriptomics",
author = "I{\~n}igo Prada-Luengo and Viktoria Schuster and Yuhu Liang and Thilde Terkelsen and Valentina Sora and Anders Krogh",
note = "Publisher Copyright: {\textcopyright} 2023, The Author(s).",
year = "2023",
doi = "10.1186/s13059-023-03104-7",
language = "English",
volume = "24",
journal = "Genome Biology (Online Edition)",
issn = "1474-7596",
publisher = "BioMed Central Ltd.",
number = "1",

}

RIS

TY - JOUR

T1 - N-of-one differential gene expression without control samples using a deep generative model

AU - Prada-Luengo, Iñigo

AU - Schuster, Viktoria

AU - Liang, Yuhu

AU - Terkelsen, Thilde

AU - Sora, Valentina

AU - Krogh, Anders

N1 - Publisher Copyright: © 2023, The Author(s).

PY - 2023

Y1 - 2023

N2 - Differential analysis of bulk RNA-seq data often suffers from lack of good controls. Here, we present a generative model that replaces controls, trained solely on healthy tissues. The unsupervised model learns a low-dimensional representation and can identify the closest normal representation for a given disease sample. This enables control-free, single-sample differential expression analysis. In breast cancer, we demonstrate how our approach selects marker genes and outperforms a state-of-the-art method. Furthermore, significant genes identified by the model are enriched in driver genes across cancers. Our results show that the in silico closest normal provides a more favorable comparison than control samples.

AB - Differential analysis of bulk RNA-seq data often suffers from lack of good controls. Here, we present a generative model that replaces controls, trained solely on healthy tissues. The unsupervised model learns a low-dimensional representation and can identify the closest normal representation for a given disease sample. This enables control-free, single-sample differential expression analysis. In breast cancer, we demonstrate how our approach selects marker genes and outperforms a state-of-the-art method. Furthermore, significant genes identified by the model are enriched in driver genes across cancers. Our results show that the in silico closest normal provides a more favorable comparison than control samples.

KW - Deep generative models

KW - Deep learning

KW - DEG

KW - DEseq2

KW - Differential expression analysis

KW - Transcriptomics

UR - http://www.scopus.com/inward/record.url?scp=85177077909&partnerID=8YFLogxK

U2 - 10.1186/s13059-023-03104-7

DO - 10.1186/s13059-023-03104-7

M3 - Journal article

C2 - 37974217

AN - SCOPUS:85177077909

VL - 24

JO - Genome Biology (Online Edition)

JF - Genome Biology (Online Edition)

SN - 1474-7596

IS - 1

M1 - 263

ER -

ID: 374645010