Separation of data by Sample

[sylestiel]

In my analysis of 2 datasets representing two age groups via Arch R I'm unable to write to file Peak-2-Gene links that are specific to each sample (groupBy="Sample")as opposed to groupBy="Clusters".

Similarly, I was wondering how i could obtain separate plots Identification of Positive Transcription Factors (chapter 15.4 in full manual).

I started of
seGroupMotif <- getGroupSE(ArchRProj = proj5, useMatrix = "MotifMatrix", groupBy = "Sample")
However, proceeding from here I do not know what to change to get two plots: 1 for each sample.
Can you assist with the script needed to achieve this?

Thanks!

[rcorces]

Why not create two different project subsets using subsetArchRProject()?

[sylestiel]

Hmm interesting.
So follow the following script:
idxSample <- BiocGenerics::which(projHeme1$Sample %in% "scATAC_BMMC_R1")
cellsSample <- projHeme1$cellNames[idxSample]
projHeme1[cellsSample, ]

Would it not be the same as doing the analysis for each one separately?
How does normalization work in that case and also when the two are processed together?

[rcorces]

we recommend using the subsetArchRProject() function whenever subsetting a project because this is more stable.

How does normalization work in that case and also when the two are processed together?

I'm not sure what you mean by normalization.

[sylestiel]

How is subsetArchRProject() different from
idxSample <- BiocGenerics::which(projHeme1$Sample %in% "scATAC_BMMC_R1")
cellsSample <- projHeme1$cellNames[idxSample]
projHeme1[cellsSample, ]

Also would I need to repeat LSI_ATAC, LSI_RNA again or is it okay to proceed with the original LSI ?

Also how is this approach of combining different conditions in one analysis similar to merging or integration of data in Signac?

Thanks!

[rcorces]

Its preferable if you reply within the same conversation rather than adding a new comment.

subsetArchRProject() ensures that the data in your Arrow files matches exactly the expected data in your ArchRProject.
subsetting using [] does not alter the Arrow files at all, just the ArchRProject.
See the code if you need further clarification

Also would I need to repeat LSI_ATAC, LSI_RNA again or is it okay to proceed with the original LSI ?

This depends on your application. you dont need to.

Also how is this approach of combining different conditions in one analysis similar to merging or integration of data in Signac?

I dont know how to answer that question.

[sylestiel]

Thank you! On August 13, 2021 at 6:13 PM Ryan Corces ***@***.***> wrote:   Its preferable if you reply within the same conversation rather than adding a new comment. subsetArchRProject() ensures that the data in your Arrow files matches exactly the expected data in your ArchRProject. subsetting using [] does not alter the Arrow files at all, just the ArchRProject. See the code if you need further clarification Also would I need to repeat LSI_ATAC, LSI_RNA again or is it okay to proceed with the original LSI ? This depends on your application. you dont need to. Also how is this approach of combining different conditions in one analysis similar to merging or integration of data in Signac? I dont know how to answer that question. —You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.Triage notifications on the go with GitHub Mobile for iOS or Android.