pre-processing problem

[Roumenina]

Hello, I am trying to run your test dataset as well as my data, but in both cases the pre-processing doesn't launch. There is a small "running man" that appears but nothing happens on the screen. When I try to pass on the next steps, it tells me that I haven't done the pre-processing. What am I doing wrong? From the paper and manual I cannot figure it out... Thanks in advance for your help. Best regards Lubka Roumenina

[kuanrongchan]

Hi Lubka, thanks for trying out STAGEs! Mind uploading your file here? We will troubleshoot and get back to you.

[Roumenina]

Thanks dear Kuan, It makes it with the demo file as well. So the problem is not necessarily my file. Apparently, my file is too large to pass via this email (4 MB). Thanks a lot for your help. I really think that your tool is great and I want to implement it for my research. Best regards Lubka

Prof. Lubka Roumenina

work mail: @.***

Researcher INSERM, HDR, DR2

Cordeliers Research Center, INSERM UMRS 1138 InC2 lab: Inflammation, complement and cancer, deputy director

15 rue de l'Ecole de Médecine Escalier E, 3ème étage 75006 Paris France

On Sat, May 13, 2023 at 6:02 AM Kuan Rong Chan @.***> wrote:

Hi Lubka, thanks for trying out STAGEs! Mind uploading your file here? We will troubleshoot and get back to you.

— Reply to this email directly, view it on GitHub https://github.com/kuanrongchan/STAGES/issues/9#issuecomment-1546518234, or unsubscribe https://github.com/notifications/unsubscribe-auth/A72KTIGI2IBMU75RNXQ6QODXF4BWFANCNFSM6AAAAAAYAHRM7U . You are receiving this because you authored the thread.Message ID: @.***>

--

Prof. Lubka Roumenina

work mail: @.***

Researcher INSERM, HDR, DR2

Cordeliers Research Center, INSERM UMRS 1138 InC2 lab: Inflammation, complement and cancer, deputy director

15 rue de l'Ecole de Médecine Escalier E, 3ème étage 75006 Paris France

[kuanrongchan]

Thanks for the feedback. Mind sending to kuanrong.chan@duke-nus.edu.sg or kuanrongchan@gmail.com?

[kohcl17]

Hi Prof Lubka,

As of now, there are no issues with the test datasets on STAGEs.

Regarding your dataset, there are some formatting issues:

1. First column is ENSEMBLE gene ID which the downstream analyses (enrichr, gsea, string?) wouldn’t detect
2. The excel sheet has both log2FC and pvalue data together with the count data
3. No metadata attached

As a side note, Enrichr and GSEA may not be as good for mouse data since we don’t have any mouse gene libraries at the moment.

I've attached a formatted version of your dataset. PBS vs PHZ lung.xlsx

Thank you for your interest in the app. :)

Best regards, Clara

[Roumenina]

Thanks a lot dear Clara, It still doesn't work... Maybe I am doing something wrong... Could we set a rapid zoom call so that you can show me how to upload and analyse the data? We have human cells datasets as well. For the mouse indeed, we should use other software for the GO but at least for the correlations and the volcano plots it will be very useful. Sorry for bothering you so much Yours Lubka

On Mon, May 15, 2023 at 7:50 AM kohcl17 @.***> wrote:

Hi Prof Lubka,

As of now, there are no issues with the test datasets on STAGEs.

Regarding your dataset, there are some formatting issues:

1. First column is ENSEMBLE gene ID which the downstream analyses (enrichr, gsea, string?) wouldn’t detect
2. The excel sheet has both log2FC and pvalue data together with the count data
3. No metadata attached

As a side note, Enrichr and GSEA may not be as good for mouse data since we don’t have any mouse gene libraries at the moment.

I've attached a formatted version of your dataset. PBS vs PHZ lung.xlsx https://github.com/kuanrongchan/STAGES/files/11474502/PBS.vs.PHZ.lung.xlsx

Thank you for your interest in the app. :)

Best regards, Clara

— Reply to this email directly, view it on GitHub https://github.com/kuanrongchan/STAGES/issues/9#issuecomment-1547228051, or unsubscribe https://github.com/notifications/unsubscribe-auth/A72KTIA67ZXANHNWTPOXUCLXGG72FANCNFSM6AAAAAAYAHRM7U . You are receiving this because you authored the thread.Message ID: @.***>

--

Prof. Lubka Roumenina

work mail: @.***

Researcher INSERM, HDR, DR2

Cordeliers Research Center, INSERM UMRS 1138 InC2 lab: Inflammation, complement and cancer, deputy director

15 rue de l'Ecole de Médecine Escalier E, 3ème étage 75006 Paris France

[kohcl17]

Hi Prof Lubka,

You may be unable to see a correlation matrix as there is a single comparison in your dataset (PHZ vs PBS). The correlation matrix would only work with more than 1 comparison as it correlates the log2 fold-change values of each comparison with one another. Likewise, this would also result in the clustergram function being unable to plot the data.

An example of how the correlation matrix and clustergram with the example dataset would look like:

As for pathway analysis, if you have mouse gene set libraries, you may also upload them in the Enrichr or GSEA step for your analysis.

I've attached some screenshots of what to select for some of the steps but you may customise them based on your preferences downstream of the file upload step. :)

File upload

Pre-processing

Differential Expression Analysis

Enrichr

GSEA Prerank

STRING

Hope this helps!

Cheers, Clara