Question about SPOTlight_deconvolution_analysis

[hoon2577]

Hi,
I was trying to follow SPOTlight_deconvolution_analysis with PDAC (https://github.com/MarcElosua/SPOTlight_deconvolution_analysis) and i faced an error in '1a-PDAC_paired_deconvolution.Rmd'.

I used data GEO accession number GSE111672 and also was aware of the discrepancy between raw data (GSE111672) and the data in that code as you mentioned.

The question is when i use
st_b_list <- sapply(sample_pdac_b, USE.NAMES = TRUE, function(geo) ......

the Seurat object in st_b_list (such as st_b_list$GSM3036911 ) Seurat object doesn't recognize the image

> st_a_list$GSM3036911
An object of class Seurat 
25753 features across 956 samples within 1 assay 
Active assay: RNA (25753 features, 0 variable features)

> st_a_list$GSM3036911@images
[[1]]
rastergrob[GRID.rastergrob.11]

> Seurat::Images(st_a_list$GSM3036911)
NULL

Is there any idea that i can make proper Seurat Object with image?

Also, i think not raw data but processed data that you used (such as single cell annotation or filtered spatial data) is proper for trying SPOTlight_deconvolution_analysis. So, is it possible to offer the real data that you used for me?
Thanks.

[MarcElosua]

Hi @hoon2577,
You should be able to download the processed data from this link. PDAC-*_ST_list.RDS contains a list with the spatial data for each slide and PDAC-*_itai_joint.RDS contains the single cell data.
Let me know if this works.

[13198910213]

Hello Author, First of all, congratulations on the successful completion of this outstanding work. I have the same trouble as the previous commenter, the ST data of PDAC is not available for Seurat related operations. I got the data from the URL you provided at https://crgcnag-my.sharepoint.com/:f:/g/personal/melosua_crg_es/EgqkdTRqeDFNkecpKmyQNOEB4WOjVerwccTL28KsPvGnww?e=JC6zJD
But in the subsequent operation, it still reports an error: Non-S4 category objects (category "rastergrob") do not have a slot for "coordinates".
Can you please provide me with the ST data of PDAC that is compatible with Seurat operation?

[MarcElosua]

hi @13198910213

I never transitioned the ST data for the PDAC samples, I used the Spaniel format.
See the discussion at the time here and the tips people had to convert it.

Hope this helps!

[13198910213]

hi @13198910213

I never transitioned the ST data for the PDAC samples, I used the Spaniel format. See the discussion at the time here and the tips people had to convert it.

Hope this helps!

Dear Professor.
Hello, thank you very much for your reply, but I am sorry that I still don't know what to do with the ST data of PDAC. From the address you provided https://crgcnag-my.sharepoint.com/:f:/g/personal/melosua_crg_es/EgqkdTRqeDFNkecpKmyQNOEB4WOjVerwccTL28KsPvGnww?e=JC6zJD下载的 ST data (PDAC-_ST_list.RDSPDAC-_itai_joint.RDS) is also not available, its data format is not Seurat object, but rasterized data. How can I convert it to Seurat object? I look forward to your reply!

PDAC-A_ST_list.RDS：

I need the ST data in the following format:

[13198910213]

Hi @hoon2577, You should be able to download the processed data from this link. PDAC-*_ST_list.RDS contains a list with the spatial data for each slide and PDAC-*_itai_joint.RDS contains the single cell data. Let me know if this works.

The ST data downloaded here is rasterized image data and does not apply to Seurat objects, need your further help, thanks!

[MarcElosua]

Hi @13198910213,

Thank you for your interest in the data we used in SPOTlight's paper.
We did not generate this data, it was already publicly available after being originally published here. We never converted the data to Seurat, instead we worked with it using the Spaniel package so I cannot help you in the conversion process. For the image, however, it is fairily straight forward to convert the raster format to others, I believe to convert the image to a Seurat-friendly format you can try img <- as.matrix(img)

Hope this helps!
Marc

[13198910213]

Dear Professor. Thank you very much for your prompt reply. I still need your help. Can you please share the processed ST data (filtered_feature_bc_matrix.tsv, tissue_positions_list_spatial_object.tsv, lowres_rot.jpg) with me? My email address is ***@***.*** I sincerely look forward to hearing from you. Sincerely,  Dr. Lee

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "MarcElosua/SPOTlight" ***@***.***&gt;; 发送时间:&nbsp;2023年5月12日(星期五) 下午5:31 ***@***.***&gt;; ***@***.******@***.***&gt;; 主题:&nbsp;Re: [MarcElosua/SPOTlight] Question about SPOTlight_deconvolution_analysis (Issue #66) Hi @13198910213, Thank you for your interest in the data we used in SPOTlight's paper. We did not generate this data, it was already publicly available after being originally published here. We never converted the data to Seurat, instead we worked with it using the Spaniel package so I cannot help you in the conversion process. For the image, however, it is fairily straight forward to convert the raster format to others, I believe to convert the image to a Seurat-friendly format you can try img <- as.matrix(img) Hope this helps! Marc — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***&gt;