diffcyt is a tool used to identify statistically different cell populations. diffcyt allows for the analysis of differential abundances (DA) and differential states (DS).
Weber, L.M., Nowicka, M., Soneson, C. et al. diffcyt: Differential discovery in high-dimensional cytometry via high-resolution clustering. Commun Biol 2, 183 (2019). https://doi.org/10.1038/s42003-019-0415-5
1. Add the diffcyt Task to your Workflow
Click Add new child task and select diffcyt from the task selector.
2. Perform Statistical Test: Differential Abundances (DA)
Differential abundances compares the different cell populations identified through filters. These can be identified through a manual gating task or clustering algorithm.
2.1 Select Files
Select the Files you want to perform the statistical testing on.
2.2 Enter diffcyt Settings
Method: Select a statistical analysis package for DA. Included packages for DA are edgeR, GLMM, and voom.
To learn more about the differences of the statistical packages, you can read the original diffcyt paper cited above. Further references about the statistical tests can be found in the References for Statistics section of the article.
Grouping Column: Select how your files will be grouped based on a metadata column.
To learn more about adding metadata to your dataset, please view Setting your Metadata.
Configure contrasts: Configure the groups to be tested against each other. Click on Configure contrasts. Select a Baseline group then select a Comparison group. Click Submit.
The groups available to select are based on the metadata column. You can add multiple comparisons for testing by using Add Row.
Pairing Column (optional): Data pairing can be selected here (e.g. before and after treatment of same patient) through a pairing metadata.
Select Filters: Select the filters you want to test for differential abundances.
2.3 Run
Click Run diffcyt. This will take you to the Status tab and you can watch the progress. However, you are free to go back to your workflow or do whatever you please while this runs in the cloud. The status can also be seen in the Workflow itself, or you can have an email sent to you when it is completed.
2.4 Review your Results
Go to the Results tab to review your Results. You can choose to download the volcano plot using the Export SVG function or download the Generated CSV file containing the values in the volcano plot. A snippet of the CSV file is shown.
3. Perform Statistical Test: Differential States (DS)
Differential states compare the median intensity of features that can describe the cellular state such as signaling proteins (e.g. phosphorylated proteins) or functional proteins (e.g. cytokines). This helps to understand the state of cells within given populations.
3.1 Select Files
Select the Files you want to perform the statistical testing on.
3.2 Enter diffcyt settings
Method: Select a statistical analysis package for DS. Included packages for DS are limma and lmm.
To learn more about the differences of the statistical packages, you can read the original diffcyt paper cited above. Further references about the statistical tests can be found in the References for Statistics section of the article.
Grouping Column: Select how your files will be grouped based on a metadata column.
To learn more about adding metadata to your dataset, please view Setting your Metadata.
Configure contrasts: Configure the groups to be tested against each other. Click on Configure contrasts. Select a Baseline group then select a Comparison group. Click Submit.
The groups available to select are based on the metadata column. You can add multiple ways of testing by using Add Row.
Pairing Column (optional): Data pairing can be selected here (e.g. before and after treatment of same patient) through a pairing metadata.
Select Filters: Select the filters you want to test for differential states.
3.3 Select Features
Select the Features you want to include in the statistical testing. Select Features panel will only be available if a DS statistical test package is selected.
3.3 Run
Click Run diffcyt. This will take you to the Status tab and you can watch the progress. However, you are free to go back to your workflow or do whatever you please while this runs in the cloud. The status can also be seen in the Workflow itself, or you can have an email sent to you when it is completed.
3.4 Review your Results
Go to the Results tab to review your Results. You can choose to download the volcano plot using the Export SVG function or download the Generated CSV file containing the values in the volcano plot. You can also download the plots created by the limma algorithm. An example of the limma plot is shown.
Only the voom (DA) and limma (DS) algorithm generates diagnostic plots.
4. References for Statistics Cited by the diffcyt authors:
edgeR: Robinson, M.D., McCarthy, D.J., Smyth, G.K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data, Bioinformatics, Volume 26, Issue 1, January 2010, Pages 139–140, https://doi.org/10.1093/bioinformatics/btp616
voom: Law, C.W., Chen, Y., Shi, W. et al. voom: precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol 15, R29 (2014). https://doi.org/10.1186/gb-2014-15-2-r29
limma: Ritchie, M.E., Phipson, B., Wu, D., et al. limma powers differential expression analyses for RNA-sequencing and microarray studies, Nucleic Acids Research, Volume 43, Issue 7, 20 April 2015, Page e47, https://doi.org/10.1093/nar/gkv007
GLMM, lmm: Nowicka, M., Krieg, C., Crowell, H.L. et al. CyTOF workflow: differential discovery in high-throughput high-dimensional cytometry datasets. F1000Research 2019, 6:748, https://doi.org/10.12688/f1000research.11622.4