TACA

Sorry, we cannot retrieve data from the server at this time. Please refresh the page later. If the issue persists, please contact us.

Color the cells by

Select / unselect a metadata value to apply the filter to all samples.
Select "S" or "H" to always show or hide a specific sample.

Dataset Information

This page shows the basic information of the dataset.

Sample Metadata

This page shows the list of samples and their metadata.

Cell Viewer

This page shows the "Cell Viewer" visualization tool. The plot can be colored in several ways and filtered by samples and cell types.

Color scheme

The dimensional reduction plot can be colored by the following three methods.

Color by cell types. Legends are displayed in the "OVERVIEW" section.

Color by gene expression (normalized). Click this button to search for genes.

Color by sample metadata. Click this button then select a metadata field.

Filter by cell types

In the "OVERVIEW" section, click the corresponding button to toggle the visibilities of cell types.

Filter by samples

Click this button to open the sample selector.

Click the values in the metadata columns to filter the samples by this value.

Alternatively, manually select whether to always show (S) or hide (H) a sample.

Differential Expression

This page provides tools for exploring DE comparisons and serves as an entry point for their downstream analyses results.
When a DE comparison (and log2FC and FDR cutoffs) is selected, its downstream analyses results may become available to view depending on whether a sufficient number of DEGs is available (i.e., PPI Network, Enrichment Analysis, Drug Screen),
and the description of the selected DE comparison is shown in the navigation panel, under Differential Expression.
When DE comparison selection or cutoffs change, these downstream tool pages will reset (navigation buttons become dark), but may be re-accessed by clicking the corresponding buttons on this page.

Definitions

DE comparisons were defined by the following three metadata fields and corresponding values for each cell type:
SUBSET - Determines which metadata field to use to subset the entire dataset. "N/A" - no subset was done.
GROUP - Determines which metadata field to use to split the (subsetted) dataset into groups for comparison. Comparisons were performed between the groups.
ADJUST - Determines which metadata field(s) should be adjusted. "N/A" - no confounding adjustment was done.
DE comparisons were organized by the following steps:
Strategy - A set of metadata fields used in SUBSET, GROUP, and ADJUST.
Comparison - A set of metadata values correspond to the fields in selected Strategy.
Cell type - Comparison was done for this cell type.

DE selection

Select Strategy, Comparison, and Cell type in this order. This action will reset the downstream analysis result pages.

Cutoff selection

Select log2FC and FDR cutoffs by clicking the cells in the "Number of DEGs by cutoffs" table. This action will reset the downstream analysis result pages.

Volcano plot

DEGs (those met the cutoff criteria) are in red. Top 10 up-DEGs and 10 down-DEGs are shown with labels. Click the dots to hide/show the labels.

Result search

Search a gene to view its log2FC and FDR. Alternatively, the two tables below the volcano plot contain all the results.

Downstream analyses

Access the downstream analyses by the following buttons. The cutoffs of number (n) of DEGs for each analysis are as below. For mouse datasets, n is the number of mapped human genes.
PPI network - n > 0
Enrichment Analysis - either n(up) ≥ 10 or n(down) ≥ 10, and n(up) ≤ 1000, and n(down) ≤ 1000
Drug Screen - 15 ≤ n ≤ 500

PPI Network

This page shows the protein-protein interaction network of the DEGs. Up to 200 DEGs with the smallest FDR are included in the network.

Enrichment Analysis

This page shows the enrichment analysis results of the DEGs. Source: Enrichr.
The plots show the top 20 pathways or GO terms with the smallest FDR. Vertical lines indicate FDR = 0.05. Circle sizes indicate overlap.

Drug Screen

This page provides tools for exploring the drug screening results against the selected DE comparison and serves as an entry point for their downstream network analyses results.
When a perturbation is selected, its downstream analysis Drug Target Network may become available to view depending on the availability of the target information,
and the description of the selected perturbation against the DE comparison is shown in the navigation panel, under Drug Screen.
The Perturbation Network is always available to view. Refer to the help of these two pages for more information.
When perturbation selection changes, these downstream tool pages will reset (navigation buttons become dark), but may be re-accessed by clicking the corresponding buttons on this page.

Definitions

CMap defines "signatures" as the consensus gene expression changes from multiple replicates of a perturbagen (e.g., chemical treatment) in certain cell line, dose, and exposure time.
Here we generalize the term "perturbation" to indicate "signature". For clarity, we show the compound names (can appear more than once) in the result tables instead of the unique signature IDs.
An enrichment score (ES) was computed for each perturbation against the selected DE comparison. The results are split into two sets:
SIGNIFICANTLY INVERSELY RELATED - ES > 0, the perturbation leads to opposite gene expression pattern to that of the selected DE comparison. In other words, the perturbation "reverses" the effect of group 1 vs. group 2 in the DE comparison.
SIGNIFICANTLY POSITIVELY RELATED - ES < 0, the perturbation leads to similar gene expression pattern to that of the selected DE comparison. In other words, the perturbation "reinforces" the effect of group 1 vs. group 2 in the DE comparison.
Note that which one of the two relationships is desired depends on how the comparison was performed. For example, when the DEGs describe "AD" vs. "healthy control", SIGNIFICANTLY INVERSELY RELATED will be the desired relationship; conversely,if the DEGs describe "healthy control" vs. "AD", SIGNIFICANTLY POSITIVELY RELATED will be the desired relationship.

Perturbation selection

Click an entry from the top two tables that show the significantly inversely related (ES > 0) and positively related (ES < 0) perturbations respectively. This action will reset the downstream analysis result pages.

Signature plot

The signature plot visualizes the gene profiles of the perturbation and the DEGs simultaneously.
The blue continuous line shows the gene expression changes in Z scores (left y-axis) in ascending order of the selected perturbation. The vertical short lines show the log2FC of the DEGs (right y-axis and color scale).

Method

Drug screening was conducted using the gene set enrichment analysis algorithm. For each perturbation, we calculated the ES against the selected DE comparison as below:
formula_1 ESup and ESdown were calculated for the up-DEGs and down-DEGs from the DE comparison separately, allowing either of the two sets to be empty. To compute ESup and ESdown, we first computed aup/down and bup/down as: formula_2 where j = 1, 2, ..., s were the indexes of DEGs sorted by their ranks in the gene profiles of the perturbation in ascending order. The rank of gene j is denoted by V(j), where 1 ≤ V(j) ≤ r, with r being the total number of genes in the profile. Then, ESup/down was set to aup/down if aup/down > bup/down, and was set to -bup/down if bup/down > aup/down. Permutation tests repeated 10000 times using randomly generated gene lists with the same n(up) and n(down) as the DEGs were performed to measure the one-tailed p values of the ES scores. Depending on the sign of the ES, the one-tailed p value measures either the left-tail (ES < 0) or the right-tail (ES > 0). P values were FDR adjusted and FDR < 0.05 were considered significant.

Ref 1. Zhou Y, Hou Y, Shen J, Huang Y, Martin W, Cheng F. Network-based drug repurposing for novel coronavirus 2019-nCoV/SARS-CoV-2. Cell Discov. 2020;6:14.
Ref 2. Sirota, M. et al. Discovery and preclinical validation of drug indications using compendia of public gene expression data. Sci Transl Med. 2011;3(96):96ra77.

Drug Target Network

This page shows the drug-target network of the selected drug. Protein-protein interactions among the targets are shown as well. Gene nodes are colored and sized by their log2FC and FDR respectively from the selected DE comparison.

Perturbation Network

This page shows the perturbation network of the selected perturbation. From all the DEGs, at most 50 DEGs with the lowest Z scores and 50 with the highest Z scores in the perturbation profile are shown in the network.
Gene nodes are colored and sized by their log2FC and FDR respectively from the selected DE comparison, while their border colors and edge (to the compound) colors indicate the Z scores.
As a result, plot for inversely related perturbation and DEGs will have inverse node and border+edge color (e.g., ), while positively related perturbation and DEGs will have similar node and border+edge color (e.g., ).
Protein-protein interactions among the DEGs are shown as grey edges.

Cell Interactions

This page provides tools for exploring cell-cell interaction (CCI) analyses and serves as an entry point for their downstream analyses results.
When a CCI analysis is selected, its downstream analyses results will become available to view. (i.e., LRI Network and CCI Network), and the description of the selected CCI analysis is shown in the navigation panel, under Cell Interactions.
When CCI analysis selection changes, these downstream tool pages will reset (navigation buttons become dark), but may be re-accessed by clicking the corresponding buttons on this page.

Definitions

CCI analyses were defined by one or more metadata fields SUBSET and their associated values used in the subsetting of the entire dataset, and were organized by the following steps:
Strategy - One or more metadata fields used in SUBSET.
Analysis - Values(s) correspond to the field(s) in SUBSET.

CCI analysis selection

Select Strategy and then Analysis. This action will reset the downstream analysis result pages.

Heatmap

The heatmap shows the number of significant (p < 0.05) ligand-receptor interactions (LRI) in all the cell-cell pairs.
Click a cell to select a specific CCI. This action will reset the LRI Network page, which can be accessed by clicking View LRI network for selected CCI. Additionally, the bottom left table displays all the significant LRIs in the selected CCI.

LRI table

The bottom right table shows all the LRIs that are significant in at least one CCI. Click the to show the CCI(s) for the selected LRI.

LRI Network

This page shows the ligand-receptor interaction network for the selected cell-cell interaction.

CCI Network

This page shows the cell-cell interaction network for the selected ligand-receptor interaction.