Cargando…

Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture

Spatial transcriptomics is an emerging technology widely applied to the analyses of tissue architecture and corresponding biological functions. Substantial computational methods have been developed for analyzing spatial transcriptomics data. These methods generate embeddings from gene expression and...

Descripción completa

Detalles Bibliográficos
Autores principales: Lin, Yu, Wang, Yan, Liang, Yanchun, Yu, Yang, Li, Jingyi, Ma, Qin, He, Fei, Xu, Dong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9411666/
https://www.ncbi.nlm.nih.gov/pubmed/36035139
http://dx.doi.org/10.3389/fgene.2022.912813
_version_ 1784775320048500736
author Lin, Yu
Wang, Yan
Liang, Yanchun
Yu, Yang
Li, Jingyi
Ma, Qin
He, Fei
Xu, Dong
author_facet Lin, Yu
Wang, Yan
Liang, Yanchun
Yu, Yang
Li, Jingyi
Ma, Qin
He, Fei
Xu, Dong
author_sort Lin, Yu
collection PubMed
description Spatial transcriptomics is an emerging technology widely applied to the analyses of tissue architecture and corresponding biological functions. Substantial computational methods have been developed for analyzing spatial transcriptomics data. These methods generate embeddings from gene expression and spatial locations for spot clustering or tissue architecture segmentation. Although the hyperparameters used to produce an embedding can be tuned for a given training set, a fixed embedding has variable performance from case to case due to data distributions. Therefore, selecting an effective embedding for new data in advance would be useful. For this purpose, we developed an embedding evaluation method named message passing-Moran’s I with maximum filtering (MP-MIM), which combines message passing-based embedding transformation with spatial autocorrelation analysis. We applied a graph convolution to aggregate spatial transcriptomics data and employed global Moran’s I to measure spatial autocorrelation and select the most effective embedding to infer tissue architecture. Sixteen spatial transcriptomics samples generated from the human brain were used to validate our method. The results show that MP-MIM can accurately identify high-quality embeddings that produce a high correlation between the predicted tissue architecture and the ground truth. Overall, our study provides a novel method to select embeddings for new test data and enhance the usability of deep learning tools for spatial transcriptome analyses.
format Online
Article
Text
id pubmed-9411666
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-94116662022-08-27 Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture Lin, Yu Wang, Yan Liang, Yanchun Yu, Yang Li, Jingyi Ma, Qin He, Fei Xu, Dong Front Genet Genetics Spatial transcriptomics is an emerging technology widely applied to the analyses of tissue architecture and corresponding biological functions. Substantial computational methods have been developed for analyzing spatial transcriptomics data. These methods generate embeddings from gene expression and spatial locations for spot clustering or tissue architecture segmentation. Although the hyperparameters used to produce an embedding can be tuned for a given training set, a fixed embedding has variable performance from case to case due to data distributions. Therefore, selecting an effective embedding for new data in advance would be useful. For this purpose, we developed an embedding evaluation method named message passing-Moran’s I with maximum filtering (MP-MIM), which combines message passing-based embedding transformation with spatial autocorrelation analysis. We applied a graph convolution to aggregate spatial transcriptomics data and employed global Moran’s I to measure spatial autocorrelation and select the most effective embedding to infer tissue architecture. Sixteen spatial transcriptomics samples generated from the human brain were used to validate our method. The results show that MP-MIM can accurately identify high-quality embeddings that produce a high correlation between the predicted tissue architecture and the ground truth. Overall, our study provides a novel method to select embeddings for new test data and enhance the usability of deep learning tools for spatial transcriptome analyses. Frontiers Media S.A. 2022-08-12 /pmc/articles/PMC9411666/ /pubmed/36035139 http://dx.doi.org/10.3389/fgene.2022.912813 Text en Copyright © 2022 Lin, Wang, Liang, Yu, Li, Ma, He and Xu. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Genetics
Lin, Yu
Wang, Yan
Liang, Yanchun
Yu, Yang
Li, Jingyi
Ma, Qin
He, Fei
Xu, Dong
Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title_full Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title_fullStr Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title_full_unstemmed Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title_short Sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
title_sort sampling and ranking spatial transcriptomics data embeddings to identify tissue architecture
topic Genetics
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9411666/
https://www.ncbi.nlm.nih.gov/pubmed/36035139
http://dx.doi.org/10.3389/fgene.2022.912813
work_keys_str_mv AT linyu samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT wangyan samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT liangyanchun samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT yuyang samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT lijingyi samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT maqin samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT hefei samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture
AT xudong samplingandrankingspatialtranscriptomicsdataembeddingstoidentifytissuearchitecture