Welcome to E-ovary!
This website contains the data from our publication on cell types in adult human ovaries, Wagner at al. 2020.
The publication is based on four different datasets. The data on cells from the outer layer of the ovary (cortex) was produced in our laboratory, and the data on cells from the inner part of the ovary (medulla) was retrieved from Fan et al. 2019 publication. The datasets are:
cortical cells
FACS sorted DDX4 antibody-positive and antibody-negative cortical cells
medulla cells
integrated cortex + medulla cells
E-ovary presents every dataset on a separate sheet. The sheets are called “cortex”, “DDX4 sorted cortex”, “medulla” and “cortex+medulla”.
The aim of E-ovary is to give you easy access to our data. Using this website, you can browse our data in a user-friendly way. We hope that this increases accessibility and interoperability, as well as interest in ovarian research. The raw data is also available in public repositories:
Cortex datasets: ArrayExpress E-MTAB-8381
Medulla datasets: GEO database GSE118127
Below you can find examples of how to study the expression of your favorite gene(s) in our datasets using UMAPs, violin/box plots, and bubble plots / heatmaps.
Visualization of your gene using UMAP
Do you want to see how your gene of interest is expressed across the individual cells in our data? You can do this by choosing “cell info vs gene expression” or “gene expression vs gene expression” under the relevant dataset.
Figure 1 shows the “cell info vs gene expression” view of the cortex data. Type in the gene symbol in the “Gene name” box (A) and select in “Cell information” (B) how you want to label the clusters (number of reads, number of detected genes, cluster identity, sample method if different patient types were analyzed, DDX4 sorting status…). Sample method refers to the tissue donor, which can be caesarean section (c-sec) or gender reassignment surgery patient (GRP). The UMAP maps will be colored and labelled accordingly. You can download the image as PDF or PNG by clicking the download buttons (C). In a similar manner, you can study the expression of two genes side by side by choosing “gene expression vs gene expression”.
If your gene cannot be found in the list, then it is not expressed at high enough level to be detected in our data. It does not necessarily mean that your gene is not expressed in ovaries, it may just be below the detection limit of the single-cell RNA-sequencing technology.
Figure 1. visualization of gene expression using color-coded UMAP. In this example, the cortex data is displayed using the “cell information vs expression” function. Type in your gene of interest to box (A) and type in the desired cell information to box (B). Here, expression of AMH is shown, and the cells have been annotated by cluster identity. You can download the images by clicking the buttons in the bottom (C). gran, granulosa cells; endo, endothelial cells; pv, perivascular cells
Visualization of gene co-expression using UMAP
Do you want to see how two genes are co-expressed by the individual cells in our data? You can do this by choosing “gene co-expression” in the pull-down menu under the relevant datasets.
Figure 2 shows the “gene co-expression” view of the cortex data. Type in the names of the two genes you wish to visualize to boxes (A) and (B). The color-coded map shows the expression of the two genes on a scale from blue to red. The number of cells expressing the genes is shown in panel (C). You can download the image as PDF or PNG by clicking the download buttons (D).
If your gene cannot be found in the list, then it is not expressed at high enough level to be detected in our data. It does not necessarily mean that your gene is not expressed in ovaries, it may just be below the detection limit of the single-cell RNA-sequencing technology.
Figure 2. Visualization of gene co-expression using color-coded UMAP. In this example, the cortex data is displayed. Select “gene co-expression” from the pull-down menu. Type in your two genes of interest in boxes (A) and (B). Here, co-expression of AMH and FOXL2 is shown. Cells only expressing FOXL2 are colored blue, those expression AMH in red, and cells co-expressing both in purple. The numbers of cells in each category are given in (C). You can download the images by clicking the download buttons (D).
Visualization of gene expression by violin/box plots
Do you want to get a better picture of the expression level of your gene in the different cell types? Then try the “violinplot/boxplot” function, which you can choose from the pull-down menu under each dataset.
Figure 3 shows the “violinplot/boxplot” view of the cortex data. Type in the name of the gene of interest in box (B) and choose in box (A) how you want to group the data (this could be cell cluster, sample method if different patient types were analyzed or DDX4 sorting status). Choose in (C) if you want a violin plot or a box plot. Click the download buttons (D) to get your violin plot as PDF or PNG file.
If your gene cannot be found in the list, then it is not expressed at high enough level to be detected in our data. It does not necessarily mean that your gene is not expressed in ovaries, it may just be below the detection limit of the single-cell RNA-sequencing technology.
Figure 3. Visualization of gene expression using violin plot. In this example, the cortex data is displayed. Select “Violinplot/Boxplot” from the pull-down menu. Type in your gene of interest in box (B) and select in box (A) how the data should be grouped. Here, expression of FOXL2 across the cell clusters is displayed as a violin plot. Download the image by clicking the button (D). gran, granulosa cells; endo, endothelial cells; pv, perivascular cells
Visualization of multiple genes using heatmap / bubble plot
Do you want to see the expression levels of multiple genes across the cell types in the datasets? Then try the “bubbleplot/heatmap” function, which you can choose from the pull-down menu under each dataset.
Figure 4 shows the “bubbleplot/heatmap” view of the cortex data. Type in the names of up to 50 genes to the box (A). Use official gene symbols, and separate the gene names by comma, semicolon, or line break. In box (B) you can choose how the results should be grouped on the x-axis (by cell cluster, sample method if different patient types were analyzed or DDX4 sorting status). You can decide to scale the gene expression (average=0, SD=1), cluster by rows, and/or cluster by columns using the tick boxes (C). The color of the dot depicts the gene expression level, and the size of the dot tells how big proportion of the cells express the gene. Click the download buttons (D) to get your plot as PDF or PNG file.
If your gene cannot be found in the list, then it is not expressed at high enough level to be detected in our data. It does not necessarily mean that your gene is not expressed in ovaries, it may just be below the detection limit of the single-cell RNA-sequencing technology.
Figure 4. Visualization of data using bubble plot. In this example, the cortex data is displayed. Select “bubbleplot/heatmap” from the pull-down menu. Type up to 50 genes of interest in box (A) and select in box (B) how the data should be grouped. By using the tick boxes (C), you can decide if you want to show the results as bubble plot or heatmap, and whether the expression data should be scaled and/or clustered. Here, expression of 42 genes across the cell clusters is displayed as a scaled bubble plot that is clustered by rows (genes). Forty genes were found in the dataset and used for the plot while three were not, as shown on the top of the plot. Download the image by clicking the button (D).
If you encounter bugs or problems with the website, please don’t hesitate to contact me.
Happy browsing!
Docent Pauliina Damdimopoulou
Division of Obstetrics and Gynecology
Department of Clinical Science, Intervention and Technology
Karolinska Institutet & Karolinska University Hospital
Stockholm, Sweden
References
Wagner, M, et al. Single-cell analysis of human ovarian cortex identifies distinct cell populations but no oogonial stem cells. Nature communications 11, 1-15 (2020) doi: 10.1038/s41598-020-74404-2. [Link]
Fan, X, et al. Single-cell reconstruction of follicular remodeling in the human adult ovary Nature communications 10, 1-13 (2019) doi: 10.1038/s41467-019-11036-9 [Link]