Introduction
Welcome
Welcome to MORLscRNASeq.org. This database contains single-cell RNA-Seq datasets generated by the Molecular Otolaryngology and Renal Research Laboratories (MORL) at the University of Iowa. We hope you find this resource valuable.
About
This website was developed to query gene expression and transcript structure from cochlear single-cell RNA-Seq datasets. Get started by selecting a tool from the menu to the left. A summary of the tools availible is shown below.
Tools
Clustering
Violin Plots
Cluster Defining Genes
Transcript Structure Browser
Details
These datasets are actively maintained and updated to include additional cells, time points, and treatment conditions. The current dataset contains 140 murine single-cell samples from three cochlear cell types: Inner Hair Cells (n=35), Outer Hair Cells (n=68), and Deiters' cells (n=37). These samples are all from the p15 time point.
This app utilizes Seurat, a single-cell RNA-Seq analysis package for R, and JBrowse, an embeddable genome browser.
Created and maintained by Paul Ranum. Last updated 6-23-2018.
Instructions
Under the green Clustering heading below you will find the Seurat generated unbiased clusters depicted by tSNE plot. Each cluster is labelled with its corresponding cell type. To query the expression level of a gene of interest, input a gene ID into the input bar below the orange P15 Clustering by Gene heading and click the submit button. You may input multiple genes of interest at once.
Clusters
P15 Clustering by Gene
Click submit to query selected genes of interest
Instructions
Under the green Violin Plot heading below you will find an example of a Seurat generated violin plot depicting the log scaled expression level of Slc26a5. Cells are grouped by cell-type identity and individual cell expression levels are depicted as black dots. To query the expression level of a gene of interest, input a gene ID into the input bar below the orange P15 Violin Plots by Gene heading and click the submit button. You may input multiple genes of interest at once.
Violin Plots
P15 Violin Plots by Gene
Click submit to query selected genes of interest
Instructions
Below is a table of cell type defining genes extracted from the indicated cell type using Seurat's FindMarkers command. The table is interactive, sortable, and searchable. The gene list shown is ordered by Receiver Operating Characteristic (ROC) Area Under Curve (AUC) classifier. AUC classifiers values range between 0 and 1 with a value of 1 representing a 100% true positive sampling rate. When applied to scRNA-Seq an AUC value of 1 would indicate that every single-cell sample within a cluster strongly differentially expressed the indicated gene. An AUC value of .5 would indicate no discriminating power between cell types. Average differential expression values (avg_diff) are also listed and represent the log scaled, averaged, level of differential expression of the indicated gene between the indicated cell type and all other cells in the dataset. The avg_diff value is a more classical representation of gene expression data. Note that genes that are the best markers of a cell type by AUC do not always have the highest differential expression values. This is because some genes are strong markers even at lower expression levels.
OHC Cluster Defining Genes
Instructions
Below is a table of cell type defining genes extracted from the indicated cell type using Seurat's FindMarkers command. The table is interactive, sortable, and searchable. The gene list shown is ordered by Receiver Operating Characteristic (ROC) Area Under Curve (AUC) classifier. AUC classifiers values range between 0 and 1 with a value of 1 representing a 100% true positive sampling rate. When applied to scRNA-Seq an AUC value of 1 would indicate that every single-cell sample within a cluster strongly differentially expressed the indicated gene. An AUC value of .5 would indicate no discriminating power between cell types. Average differential expression values (avg_diff) are also listed and represent the log scaled, averaged, level of differential expression of the indicated gene between the indicated cell type and all other cells in the dataset. The avg_diff value is a more classical representation of gene expression data. Note that genes that are the best markers of a cell type by AUC do not always have the highest differential expression values. This is because some genes are strong markers even at lower expression levels.
IHC Cluster Defining Genes
Instructions
Below is a table of cell type defining genes extracted from the indicated cell type using Seurat's FindMarkers command. The table is interactive, sortable, and searchable. The gene list shown is ordered by Receiver Operating Characteristic (ROC) Area Under Curve (AUC) classifier. AUC classifiers values range between 0 and 1 with a value of 1 representing a 100% true positive sampling rate. When applied to scRNA-Seq an AUC value of 1 would indicate that every single-cell sample within a cluster strongly differentially expressed the indicated gene. An AUC value of .5 would indicate no discriminating power between cell types. Average differential expression values (avg_diff) are also listed and represent the log scaled, averaged, level of differential expression of the indicated gene between the indicated cell type and all other cells in the dataset. The avg_diff value is a more classical representation of gene expression data. Note that genes that are the best markers of a cell type by AUC do not always have the highest differential expression values. This is because some genes are strong markers even at lower expression levels.