Loading Seurat object into Cell Harmony

9 views
Skip to first unread message

beste...@gmail.com

unread,
Sep 29, 2020, 4:22:31 PM9/29/20
to Alternative Splicing and Functional Prediction
Dear Altanalyze team,

Trying to use two Seurat objects to make an ICGS reference. 

The R script example provided for Cell Harmony only shows how to work with one Seurat object . Thus,  query and reference are the same Seurat object. 

What to do when one is comparing to a reference object that is a different Seurat object such as a control sample and experimental sample? 
I tried, produces reference and query files but not the ICGS-reference file.  Script produced some errors, see screen shot. I then hit execute to proceed with the remainder of the function.

Nathan Salomonis

unread,
Oct 1, 2020, 10:07:09 AM10/1/20
to Alternative Splicing and Functional Prediction, Brian Estevez, Schnell, Daniel (Dan)
Hi Brian,

There may be a few things going on here, so I wanted to ask a few questions and get the input of others. 

1) It sounds like you have an experimental sample/group (scRNA-Seq) and a control sample/group (scRNA-Seq) as an integrated object in Seurat, correct?

2) You are using the Seurat code to export the necessary files:
    a) a full scaled expression matrix of all scRNA-Seq samples together (log2 of counts per 10,000 scaled)
    b) a "reference" file, in the form of an AltAnalyze heatmap (Seurat top 50 markers and Seurat clusters).

3) If using this protocol you will also need a labels file described in the link below.


Dan, can you look through the attached doc and this thread and add details to this above google doc necessary to replicate the cellHarmony workflow where you skip the alignment step to generate all downstream cellHarmony comparison results (comprehensive differential expression and cell frequency analyses). 

Best,
Nathan


--
You received this message because you are subscribed to the Google Groups "Alternative Splicing and Functional Prediction" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alt_predictio...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/alt_predictions/a7d342b1-be19-44c1-98ed-3852585f956bn%40googlegroups.com.

beste...@gmail.com

unread,
Oct 5, 2020, 2:14:40 PM10/5/20
to Schnell, Daniel (Dan), Nathan Salomonis, Alternative Splicing and Functional Prediction
Dan,

Thanks so much for this. I will have a look and let you know if I have any questions.

-Brian

On Mon, Oct 5, 2020 at 1:34 PM Schnell, Daniel (Dan) <Daniel....@cchmc.org> wrote:

Hi Brian,

 

The attached set of files provides an example of creating files to pass results from a Seurat integrated clustering run to cellHarmony (for Ref vs. Query differential expression analysis w/in each cluster).  It is based on one of the Seurat integration vignettes.  An intermediate dataset is also included as a check.

 

It does require the user to change some R code, e.g. file paths and name of Seurat clustering variable to use, but the code is commented so as to point out such instances. The readme.rtf file has an overview of the process and detailed instructions and suggestions.

 

I hope this is helpful.  Feel free to reconnect if there are difficulties adapting/using the programs to prepare the datasets or the cellHarmony run itself.

 

Best,

Dan

 

 

 

From: "Schnell, Daniel (Dan)" <Daniel....@cchmc.org>
Date: Thursday, October 1, 2020 at 10:28 AM
To: Nathan Salomonis <nsalo...@gmail.com>, Alternative Splicing and Functional Prediction <alt_pre...@googlegroups.com>, Brian Estevez <beste...@gmail.com>
Subject: Re: AltAnalyze User Group Loading Seurat object into Cell Harmony

 

Will do.

 

From: Nathan Salomonis <nsalo...@gmail.com>
Date: Thursday, October 1, 2020 at 10:12 AM
To: Alternative Splicing and Functional Prediction <alt_pre...@googlegroups.com>, Brian Estevez <beste...@gmail.com>, "Schnell, Daniel (Dan)" <Daniel....@cchmc.org>
Subject: Re: AltAnalyze User Group Loading Seurat object into Cell Harmony

 

Hi Brian,

 

There may be a few things going on here, so I wanted to ask a few questions and get the input of others. 

 

1) It sounds like you have an experimental sample/group (scRNA-Seq) and a control sample/group (scRNA-Seq) as an integrated object in Seurat, correct?

 

2) You are using the Seurat code to export the necessary files:

    a) a full scaled expression matrix of all scRNA-Seq samples together (log2 of counts per 10,000 scaled)

    b) a "reference" file, in the form of an AltAnalyze heatmap (Seurat top 50 markers and Seurat clusters).

 

3) If using this protocol you will also need a labels file described in the link below.

 

 

Dan, can you look through the attached doc and this thread and add details to this above google doc necessary to replicate the cellHarmony workflow where you skip the alignment step to generate all downstream cellHarmony comparison results (comprehensive differential expression and cell frequency analyses). 

 

Best,

Nathan

 

 

On Tue, Sep 29, 2020 at 4:22 PM beste...@gmail.com <beste...@gmail.com> wrote:

Dear Altanalyze team,

 

Trying to use two Seurat objects to make an ICGS reference. 

 

The R script example provided for Cell Harmony only shows how to work with one Seurat object . Thus,  query and reference are the same Seurat object. 

 

What to do when one is comparing to a reference object that is a different Seurat object such as a control sample and experimental sample? 

I tried, produces reference and query files but not the ICGS-reference file.  Script produced some errors, see screen shot. I then hit execute to proceed with the remainder of the function.

Error! Filename not specified.

Schnell, Daniel (Dan)

unread,
Oct 12, 2020, 1:03:06 PM10/12/20
to Nathan Salomonis, Alternative Splicing and Functional Prediction, Brian Estevez

Hi Brian,

 

The attached set of files provides an example of creating files to pass results from a Seurat integrated clustering run to cellHarmony (for Ref vs. Query differential expression analysis w/in each cluster).  It is based on one of the Seurat integration vignettes.  An intermediate dataset is also included as a check.

 

It does require the user to change some R code, e.g. file paths and name of Seurat clustering variable to use, but the code is commented so as to point out such instances. The readme.rtf file has an overview of the process and detailed instructions and suggestions.

 

I hope this is helpful.  Feel free to reconnect if there are difficulties adapting/using the programs to prepare the datasets or the cellHarmony run itself.

 

Best,

Dan

 

 

 

From: "Schnell, Daniel (Dan)" <Daniel....@cchmc.org>
Date: Thursday, October 1, 2020 at 10:28 AM
To: Nathan Salomonis <nsalo...@gmail.com>, Alternative Splicing and Functional Prediction <alt_pre...@googlegroups.com>, Brian Estevez <beste...@gmail.com>
Subject: Re: AltAnalyze User Group Loading Seurat object into Cell Harmony

 

Will do.

 

From: Nathan Salomonis <nsalo...@gmail.com>
Date: Thursday, October 1, 2020 at 10:12 AM
To: Alternative Splicing and Functional Prediction <alt_pre...@googlegroups.com>, Brian Estevez <beste...@gmail.com>, "Schnell, Daniel (Dan)" <Daniel....@cchmc.org>
Subject: Re: AltAnalyze User Group Loading Seurat object into Cell Harmony

 

Hi Brian,

 

There may be a few things going on here, so I wanted to ask a few questions and get the input of others. 

 

1) It sounds like you have an experimental sample/group (scRNA-Seq) and a control sample/group (scRNA-Seq) as an integrated object in Seurat, correct?

 

2) You are using the Seurat code to export the necessary files:

    a) a full scaled expression matrix of all scRNA-Seq samples together (log2 of counts per 10,000 scaled)

    b) a "reference" file, in the form of an AltAnalyze heatmap (Seurat top 50 markers and Seurat clusters).

 

3) If using this protocol you will also need a labels file described in the link below.

 

 

Dan, can you look through the attached doc and this thread and add details to this above google doc necessary to replicate the cellHarmony workflow where you skip the alignment step to generate all downstream cellHarmony comparison results (comprehensive differential expression and cell frequency analyses). 

 

Best,

Nathan

 

 

On Tue, Sep 29, 2020 at 4:22 PM beste...@gmail.com <beste...@gmail.com> wrote:

Dear Altanalyze team,

 

Trying to use two Seurat objects to make an ICGS reference. 

 

The R script example provided for Cell Harmony only shows how to work with one Seurat object . Thus,  query and reference are the same Seurat object. 

 

What to do when one is comparing to a reference object that is a different Seurat object such as a control sample and experimental sample? 

I tried, produces reference and query files but not the ICGS-reference file.  Script produced some errors, see screen shot. I then hit execute to proceed with the remainder of the function.

Error! Filename not specified.

cHXworkflow_forBE.zip
Reply all
Reply to author
Forward
0 new messages