SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types
Single-cell RNA sequencing is increasing our understanding of the behavior of complex tissues or organs, by providing unprecedented details on the complex cell type landscape at the level of individual cells. Cell type definition and functional annotation are key steps to understanding the molecular...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Biology |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-7737/12/4/579 |
| _version_ | 1827745848193187840 |
|---|---|
| author | Pelin Gundogdu Inmaculada Alamo Isabel A. Nepomuceno-Chamorro Joaquin Dopazo Carlos Loucera |
| author_facet | Pelin Gundogdu Inmaculada Alamo Isabel A. Nepomuceno-Chamorro Joaquin Dopazo Carlos Loucera |
| author_sort | Pelin Gundogdu |
| collection | DOAJ |
| description | Single-cell RNA sequencing is increasing our understanding of the behavior of complex tissues or organs, by providing unprecedented details on the complex cell type landscape at the level of individual cells. Cell type definition and functional annotation are key steps to understanding the molecular processes behind the underlying cellular communication machinery. However, the exponential growth of scRNA-seq data has made the task of manually annotating cells unfeasible, due not only to an unparalleled resolution of the technology but to an ever-increasing heterogeneity of the data. Many supervised and unsupervised methods have been proposed to automatically annotate cells. Supervised approaches for cell-type annotation outperform unsupervised methods except when new (unknown) cell types are present. Here, we introduce SigPrimedNet an artificial neural network approach that leverages (i) efficient training by means of a sparsity-inducing signaling circuits-informed layer, (ii) feature representation learning through supervised training, and (iii) unknown cell-type identification by fitting an anomaly detection method on the learned representation. We show that SigPrimedNet can efficiently annotate known cell types while keeping a low false-positive rate for unseen cells across a set of publicly available datasets. In addition, the learned representation acts as a proxy for signaling circuit activity measurements, which provide useful estimations of the cell functionalities. |
| first_indexed | 2024-03-11T05:13:16Z |
| format | Article |
| id | doaj.art-a12437f7fea94bb98ed71796197a7647 |
| institution | Directory Open Access Journal |
| issn | 2079-7737 |
| language | English |
| last_indexed | 2024-03-11T05:13:16Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biology |
| spelling | doaj.art-a12437f7fea94bb98ed71796197a76472023-11-17T18:24:13ZengMDPI AGBiology2079-77372023-04-0112457910.3390/biology12040579SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell TypesPelin Gundogdu0Inmaculada Alamo1Isabel A. Nepomuceno-Chamorro2Joaquin Dopazo3Carlos Loucera4Computational Medicine Platform, Andalusian Public Foundation Progress and Health-FPS, 41013 Sevilla, SpainComputational Medicine Platform, Andalusian Public Foundation Progress and Health-FPS, 41013 Sevilla, SpainDpto. de Lenguajes y Sistemas Informáticos, Universidad de Sevilla, 41013 Seville, SpainComputational Medicine Platform, Andalusian Public Foundation Progress and Health-FPS, 41013 Sevilla, SpainComputational Medicine Platform, Andalusian Public Foundation Progress and Health-FPS, 41013 Sevilla, SpainSingle-cell RNA sequencing is increasing our understanding of the behavior of complex tissues or organs, by providing unprecedented details on the complex cell type landscape at the level of individual cells. Cell type definition and functional annotation are key steps to understanding the molecular processes behind the underlying cellular communication machinery. However, the exponential growth of scRNA-seq data has made the task of manually annotating cells unfeasible, due not only to an unparalleled resolution of the technology but to an ever-increasing heterogeneity of the data. Many supervised and unsupervised methods have been proposed to automatically annotate cells. Supervised approaches for cell-type annotation outperform unsupervised methods except when new (unknown) cell types are present. Here, we introduce SigPrimedNet an artificial neural network approach that leverages (i) efficient training by means of a sparsity-inducing signaling circuits-informed layer, (ii) feature representation learning through supervised training, and (iii) unknown cell-type identification by fitting an anomaly detection method on the learned representation. We show that SigPrimedNet can efficiently annotate known cell types while keeping a low false-positive rate for unseen cells across a set of publicly available datasets. In addition, the learned representation acts as a proxy for signaling circuit activity measurements, which provide useful estimations of the cell functionalities.https://www.mdpi.com/2079-7737/12/4/579scRNA-seqdeep learningexplainable artificial intelligencecell signalingcell-type identification |
| spellingShingle | Pelin Gundogdu Inmaculada Alamo Isabel A. Nepomuceno-Chamorro Joaquin Dopazo Carlos Loucera SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types Biology scRNA-seq deep learning explainable artificial intelligence cell signaling cell-type identification |
| title | SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types |
| title_full | SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types |
| title_fullStr | SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types |
| title_full_unstemmed | SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types |
| title_short | SigPrimedNet: A Signaling-Informed Neural Network for scRNA-seq Annotation of Known and Unknown Cell Types |
| title_sort | sigprimednet a signaling informed neural network for scrna seq annotation of known and unknown cell types |
| topic | scRNA-seq deep learning explainable artificial intelligence cell signaling cell-type identification |
| url | https://www.mdpi.com/2079-7737/12/4/579 |
| work_keys_str_mv | AT pelingundogdu sigprimednetasignalinginformedneuralnetworkforscrnaseqannotationofknownandunknowncelltypes AT inmaculadaalamo sigprimednetasignalinginformedneuralnetworkforscrnaseqannotationofknownandunknowncelltypes AT isabelanepomucenochamorro sigprimednetasignalinginformedneuralnetworkforscrnaseqannotationofknownandunknowncelltypes AT joaquindopazo sigprimednetasignalinginformedneuralnetworkforscrnaseqannotationofknownandunknowncelltypes AT carlosloucera sigprimednetasignalinginformedneuralnetworkforscrnaseqannotationofknownandunknowncelltypes |