Skip to content

Interpretation of CellphoneDB results from the output files #221

@ankitasamanta95

Description

@ankitasamanta95

Hi,

I am using CellphoneDB for my research project and have several questions about the output files and their potential interpretations.

In the p-values.txt file, I notice that many cell type interactions have a p-value of 0 or 1. For those with a value of zero, should I interpret them as interactions with higher significance since they are less than 0.05? For those with p-values of 1, should I consider them as not significant because they are greater than 0.05?
According to the online documentation, I should focus on specific cell type pairs and manually review interactions, prioritizing those with a lower p-value and/or higher mean expression. Do you recommend using any specific lower threshold for the mean expression value?
Regarding the significant_means.txt file, can you please explain how to interpret the results? The documentation states that if the p-value < 0.05, the value will be the mean; otherwise, the value is set to 0. (Only in significant_means.txt). However, in my output, many cells are blank and contain no values. I am unsure how to interpret this, and any advice would be greatly appreciated.

Thank you very much for your help!

Best regards,
Ankita Samanta

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions