PlantscRNAdb: A database for plant single-cell RNA analysis

Abstract

Many functional characteristics of plant tissue attribute to highly specialized cells in the tissue. Dissecting cell types and profiling the transcriptome and proteome of individual cell types of plant tissues are important for understanding cellular activities underlying plant development and stress adaptation (Shaw et al., 2021Shaw R. Tian X. Xu J. Single-cell transcriptome analysis in plants: advances and challenges.Mol. Plant. 2021; 14: 115-126Abstract Full Text Full Text PDF PubMed Scopus (33) Google Scholar). Single-cell RNA sequencing (scRNA-seq) provides such a tool for studying the transcriptional activities of individual cell types at single-cell resolution in any given organism (Rich-Griffin et al., 2020Rich-Griffin C. Stechemesser A. Finch J. Lucas E. Ott S. Schäfer P. Single-cell transcriptomics: a high-resolution avenue for plant functional genomics.Trends Plant Sci. 2020; 25: 186-197Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar). scRNA-seq has been used to study the transcriptional heterogeneity of several plant tissues, including roots (Denyer et al., 2019Denyer T. Ma X. Klesen S. Scacchi E. Nieselt K. Timmermans M.C.P. Spatiotemporal developmental trajectories in the Arabidopsis root revealed using high-throughput single-cell RNA sequencing.Dev. Cell. 2019; 48: 840-852Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar; Jean-Baptiste et al., 2019Jean-Baptiste K. McFaline-Figueroa J.L. Alexandre C.M. Dorrity M.W. Saunders L. Bubb K.L. Trapnell C. Fields S. Queitsch C. Cuperus J.T. Dynamics of gene expression in single root cells of Arabidopsis thaliana.Plant Cell. 2019; 31: 993-1011Crossref PubMed Scopus (126) Google Scholar; Ryu et al., 2019Ryu K.H. Huang L. Kang H.M. Schiefelbein J. Single-cell RNA sequencing resolves molecular relationships among individual plant cells.Plant Physiol. 2019; 179: 1444-1456Crossref PubMed Scopus (156) Google Scholar; Shulse et al., 2019Shulse C.N. Cole B.J. Ciobanu D. Lin J. Yoshinaga Y. Gouran M. Turco G.M. Zhu Y. O'Malley R.C. Brady S.M. et al.High-throughput single-cell transcriptome profiling of plant cell types.Cell Rep. 2019; 27: 2241-2247Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar; Zhang et al., 2019Zhang T.Q. Xu Z.G. Shang G.D. Wang J.W. A single-cell RNA sequencing profiles the developmental landscape of Arabidopsis root.Mol. Plant. 2019; 12: 648-660Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar; Liu et al., 2020aLiu Q. Liang Z. Feng D. Jiang S. Wang Y. Du Z. Li R. Hu G. Zhang P. Ma Y. et al.Transcriptional landscape of rice roots at the single cell resolution.Mol. Plant. 2020; 14: 384-394Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar; Farmer et al., 2021Farmer A. Thibivilliers S. Ryu K.H. Schiefelbein J. Libault M. Single-nucleus RNA and ATAC sequencing reveals the impact of chromatin accessibility on gene expression in Arabidopsis roots at the single-cell level.Mol. Plant. 2021; 14: 372-383Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar), shoot apical meristem (Satterlee et al., 2020Satterlee J.W. Strable J. Scanlon M.J. Plant stem-cell organization and differentiation at single-cell resolution.Proc. Natl. Acad. Sci. U S A. 2020; 117: 33689-33699Crossref PubMed Google Scholar), germinal cells from the anther (Nelms and Walbot, 2019Nelms B. Walbot V. Defining the developmental program leading to meiosis in maize.Science. 2019; 364: 52-56Crossref PubMed Scopus (68) Google Scholar), and leaves (Liu et al., 2020bLiu Z. Zhou Y. Guo J. Li J. Tian Z. Zhu Z. Wang J. Wu R. Zhang B. Hu Y. et al.Global dynamic molecular profiling of stomatal lineage cell development by single-cell RNA sequencing.Mol. Plant. 2020; 13: 1178-1193Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). It has also been used to study the response of different cell types of plant tissue to environmental conditions, such as heat or sugar treatment (Jean-Baptiste et al., 2019Jean-Baptiste K. McFaline-Figueroa J.L. Alexandre C.M. Dorrity M.W. Saunders L. Bubb K.L. Trapnell C. Fields S. Queitsch C. Cuperus J.T. Dynamics of gene expression in single root cells of Arabidopsis thaliana.Plant Cell. 2019; 31: 993-1011Crossref PubMed Scopus (126) Google Scholar; Shulse et al., 2019Shulse C.N. Cole B.J. Ciobanu D. Lin J. Yoshinaga Y. Gouran M. Turco G.M. Zhu Y. O'Malley R.C. Brady S.M. et al.High-throughput single-cell transcriptome profiling of plant cell types.Cell Rep. 2019; 27: 2241-2247Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar), and to study the evolutionary trajectory of plant tissue, such as the root tip, in mono- and di-cotyledonous plants (Jean-Baptiste et al., 2019Jean-Baptiste K. McFaline-Figueroa J.L. Alexandre C.M. Dorrity M.W. Saunders L. Bubb K.L. Trapnell C. Fields S. Queitsch C. Cuperus J.T. Dynamics of gene expression in single root cells of Arabidopsis thaliana.Plant Cell. 2019; 31: 993-1011Crossref PubMed Scopus (126) Google Scholar; Shulse et al., 2019Shulse C.N. Cole B.J. Ciobanu D. Lin J. Yoshinaga Y. Gouran M. Turco G.M. Zhu Y. O'Malley R.C. Brady S.M. et al.High-throughput single-cell transcriptome profiling of plant cell types.Cell Rep. 2019; 27: 2241-2247Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar; Zhang et al., 2019Zhang T.Q. Xu Z.G. Shang G.D. Wang J.W. A single-cell RNA sequencing profiles the developmental landscape of Arabidopsis root.Mol. Plant. 2019; 12: 648-660Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar; Liu et al., 2020aLiu Q. Liang Z. Feng D. Jiang S. Wang Y. Du Z. Li R. Hu G. Zhang P. Ma Y. et al.Transcriptional landscape of rice roots at the single cell resolution.Mol. Plant. 2020; 14: 384-394Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Rapid increase of single-cell transcriptomic data demands establishment of databases and tools to accommodate and explore such data, so they can be easily accessed and utilized by everyone who is interested in the expression pattern of specific genes in different cell types, and in identifying reference and marker genes of specific cell types for further investigation. Many scRNA-seq databases have been developed using data from human and model animals. However, no scRNA-seq-specific database, particularly one including comprehensive analysis results, such as marker genes of specific cell types and their expression profiles, from multiple plant species is available up to now. All current available plant scRNA-seq databases are based on a single scRNA-seq study in a single plant species (Zhang et al., 2019Zhang T.Q. Xu Z.G. Shang G.D. Wang J.W. A single-cell RNA sequencing profiles the developmental landscape of Arabidopsis root.Mol. Plant. 2019; 12: 648-660Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar; Liu et al., 2020aLiu Q. Liang Z. Feng D. Jiang S. Wang Y. Du Z. Li R. Hu G. Zhang P. Ma Y. et al.Transcriptional landscape of rice roots at the single cell resolution.Mol. Plant. 2020; 14: 384-394Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar; Ma et al., 2020Ma X. Denyer T. Timmermans M. PscB: a browser to explore plant single cell RNA-sequencing data sets.Plant Physiol. 2020; 183: 464-467Crossref PubMed Scopus (11) Google Scholar). Therefore, it is imperative to build a comprehensive plant scRNA database for the plant community. We established a plant single-cell transcriptome database (termed PlantscRNAdb; http://ibi.zju.edu.cn/plantscrnadb/). PlantscRNAdb includes 26 326 marker genes of 128 different cell types from four plant species (Arabidopsis thaliana, Oryza sativa, Solanum lycopersicum, and Zea mays). The database can also be used to explore gene expression at both single-cell-type and genome-wide scale. We collected publicly available marker genes of diverse plant cell types in the database. The initial set of marker genes were classified into four types based on supporting evidence and data source. (1) Marker genes with experimental verification evidence, such as those identified based on expression of GFP reporter. We collected and manually annotated these cell-type marker genes from about 3000 reports in the literature that were obtained from PubMed by searching keywords, including “marker(s)” or “specific expression,” the names of tissue and cell types, etc. (2) Marker genes previously identified based on specific cell-type bulk RNA-seq studies. (3) Marker genes identified by the published scRNA-seq studies, i.e., those identified as significantly differentially expressed in different cell types based on scRNA-seq. (4) Marker genes we identified by re-analyzing the published scRNA-seq data using a unified method and a set of common parameters (see Supplemental file 1 for details), as we believed that some marker genes could be wrongly identified or might have been missed due to inconsistent methods and parameters used in different studies. The cell-type marker genes obtained based on the above strategies were then merged for each plant species to prevent redundancy. Finally, a total of 14 922 (representing 79 cell types from 10 tissues), 5428 (representing 35 cell types from 5 tissues), 5901 (representing 42 cell types from 9 tissues), and 75 (representing 25 cell types from 5 tissues) marker genes were collected for A. thaliana, O. sativa, Z. mays, and S. lycopersicum, respectively (Figure 1A and 1B ). Among the above marker genes, most (>90%) were mainly derived from recent scRNA data (Supplemental Figures 1 and 2). Furthermore, we formulated a simple rule to classify the marker genes identified by re-analyzing our scRNA data: when the number of reads of a gene in a specific cell type accounts for more than 80% of the total number of reads from the gene, i.e., the expression of the gene is mainly contributed by cell type, we classify the gene as class I, labeled “Marker#1.” Otherwise, the gene is classified as class II, labeled “Marker#2.” All marker genes can be downloaded from PlantscRNAdb individually for each organism or collectively as a whole set. Meanwhile, the database provides two search functions, i.e., via cell type or marker gene. After selecting the species, tissue, and cell type, information of marker genes corresponding to the cell type can be displayed in detail. By selecting the species of interest and inputting the gene ID of the marker gene of interest, the expression information of the marker gene in different scRNA-seq studies can be displayed through the TSNE or UMAP format (Figure 1C). For samples with treatments, the effects of treatments on the expression of the gene can also be compared. Being able to compare the gene expression profile in different cell types within and among the tissue(s) of interest is the biggest advantage of single-cell transcriptomic data. PlantscRNAdb contains detailed and comprehensive information of genes (especially cell-type marker genes) in diverse cell types. It uses a genome browser to provide more intuitive information about gene expression in different cell types (Figure 1D). The expression data displayed (the bam file of the corresponding cell type) were normalized by taking into account the different cell numbers sequenced in different scRNA studies (see Supplemental file 1 for details). To meet the high demand of personalized analysis of scRNA-seq data, PlantscRNAdb also provides the expression matrix corresponding to each scRNA-seq dataset. At the same time, detailed information (e.g., the sequencing protocol used and the number of cell sequenced) of each published article is provided in the database to provide guidance for users to design their own single-cell experiments. Most plant species, especially the non-model species, have fewer cell-type marker genes available than the model plant species, which makes it difficult to properly analyze the scRNA-seq data generated from these species. This drawback can be partially overcome by using homologous marker genes identified in their closely related species. To this end, PlantscRNAdb provides an online BLASTP tool to search for homologous marker genes in the four species collected in the database using sequences from plant species of interest as a query (Figure 1E). In short, we created a scRNA-seq database with the aim to track and analyze all available plant single-cell transcriptome-related data and to provide researchers a user-friendly platform for their single-cell studies. PlantscRNAdb will be regularly updated when new plant scRNA datasets are available and new functions will be developed and adopted in the database to meet the increasing demands for analyzing scRNA data. We hope PlantscRNAdb will become a single-stop database for plant scRNA-seq data and facilitate plant single-cell studies.