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 representati...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2023-11-01
|
Series: | Genome Biology |
Subjects: | |
Online Access: | https://doi.org/10.1186/s13059-023-03104-7 |
_version_ | 1797559348579270656 |
---|---|
author | Iñigo Prada-Luengo Viktoria Schuster Yuhu Liang Thilde Terkelsen Valentina Sora Anders Krogh |
author_facet | Iñigo Prada-Luengo Viktoria Schuster Yuhu Liang Thilde Terkelsen Valentina Sora Anders Krogh |
author_sort | Iñigo Prada-Luengo |
collection | DOAJ |
description | 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. |
first_indexed | 2024-03-10T17:44:01Z |
format | Article |
id | doaj.art-a01c7581a83340df990a066ff0394ff0 |
institution | Directory Open Access Journal |
issn | 1474-760X |
language | English |
last_indexed | 2024-03-10T17:44:01Z |
publishDate | 2023-11-01 |
publisher | BMC |
record_format | Article |
series | Genome Biology |
spelling | doaj.art-a01c7581a83340df990a066ff0394ff02023-11-20T09:35:30ZengBMCGenome Biology1474-760X2023-11-0124111710.1186/s13059-023-03104-7N-of-one differential gene expression without control samples using a deep generative modelIñigo Prada-Luengo0Viktoria Schuster1Yuhu Liang2Thilde Terkelsen3Valentina Sora4Anders Krogh5Department of Computer Science, University of CopenhagenCenter for Health Data Science, University of CopenhagenDepartment of Computer Science, University of CopenhagenCenter for Health Data Science, University of CopenhagenDepartment of Computer Science, University of CopenhagenDepartment of Computer Science, University of CopenhagenAbstract 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.https://doi.org/10.1186/s13059-023-03104-7Deep generative modelsDeep learningDifferential expression analysisDEGDEseq2Transcriptomics |
spellingShingle | Iñigo Prada-Luengo Viktoria Schuster Yuhu Liang Thilde Terkelsen Valentina Sora Anders Krogh N-of-one differential gene expression without control samples using a deep generative model Genome Biology Deep generative models Deep learning Differential expression analysis DEG DEseq2 Transcriptomics |
title | N-of-one differential gene expression without control samples using a deep generative model |
title_full | N-of-one differential gene expression without control samples using a deep generative model |
title_fullStr | N-of-one differential gene expression without control samples using a deep generative model |
title_full_unstemmed | N-of-one differential gene expression without control samples using a deep generative model |
title_short | N-of-one differential gene expression without control samples using a deep generative model |
title_sort | n of one differential gene expression without control samples using a deep generative model |
topic | Deep generative models Deep learning Differential expression analysis DEG DEseq2 Transcriptomics |
url | https://doi.org/10.1186/s13059-023-03104-7 |
work_keys_str_mv | AT inigopradaluengo nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel AT viktoriaschuster nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel AT yuhuliang nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel AT thildeterkelsen nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel AT valentinasora nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel AT anderskrogh nofonedifferentialgeneexpressionwithoutcontrolsamplesusingadeepgenerativemodel |