Making a subset object in Xenium analysis

[NA0004]

Hi,

I would like to know how to subset Xenium object.

Following the tutorial "Analysis of Image-based Spatial Data in Seurat", I cropped xenium.obj using the tutorial codes.

cropped.coords <- Crop(xenium.obj[["fov"]], x = c(1200, 2900), y = c(3750, 4550), coords = "plot")
xenium.obj[["zoom"]] <- cropped.coords

Now I would like to make a new Xenium object of this area by using subset function so that I can analyze specific brain area and DEG between two areas using FindMarker function.

I have tried to use subset command mimicking visium tutorial found on the web, but I have been failing.

It would be very helpful if I could get help for subsetting Xenium object.

Thank you.

[gretajuu]

Hello,

Were you able to subset the data? I would be interested of this feature too.

[NA0004]

Hi, yes, I could subset data! It would be great if Seurat could provide tools to make susets! Best, NA 2024年3月11日(月) 9:44 gretajuu ***@***.***>:

…

Hello, Were you able to subset the data? I would be interested of this feature too. — Reply to this email directly, view it on GitHub <#7947 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BDTAPAYBHXGRORICM5HYSYTYXVVGDAVCNFSM6AAAAAA6UVLX6WVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4DONBSGQ4DQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[kelvinmo0513]

Hi @NA0004,

Could you please let me know how you subset the xenium data?

Thanks,
Kelvin

[ilovegreentea]

Hey I used subset_opt function https://github.com/alikhuseynov/add-on_R/blob/develop/R/subset_obj_seurat.R

…

On Mon, May 20, 2024, 00:00 Kelvin Mo ***@***.***> wrote: Hi @NA0004 <https://github.com/NA0004>, Could you please let me know how you subset the xenium data? Thanks, Kelvin — Reply to this email directly, view it on GitHub <#7947 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXOTFKLXRPY7M33XSQ5E3ULZDFYM5AVCNFSM6AAAAAA6UVLX6WVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TIOJSGEZDE> . You are receiving this because you commented.Message ID: ***@***.***>

[kelvinmo0513]

Thanks so much for sharing the function with me @ilovegreentea !!
This is my code and does this function work to subset the xenium object based on coordinates of cropped image?

crop <- Crop(xenium.obj[["fov"]], x = c(-96.42, 3370.79), y = c(1162.64, 3130.46))
xenium.obj[["cropped"]] <- crop

[ilovegreentea]

Hi Kelvin, I'm not sure, I understand your question well. If you added fov to a xenium object before subset, It also works in cropped images. 2024년 5월 20일 (월) 오전 3:24, Kelvin Mo ***@***.***>님이 작성:

…

Thanks so much for sharing the function with me @ilovegreentea <https://github.com/ilovegreentea> !! This is my code and does this function work to subset the xenium object based on coordinates of cropped image? crop <- Crop(xenium.obj[["fov"]], x = c(-96.42, 3370.79), y = c(1162.64, 3130.46)) xenium.obj[["cropped"]] <- crop — Reply to this email directly, view it on GitHub <#7947 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXOTFKL6GQKW6K5HUVANEADZDGQMJAVCNFSM6AAAAAA6UVLX6WVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TIOJTGIYDA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[kelvinmo0513]

Hi @ilovegreentea

Sorry for the confusion. So what I did was to crop the image from the xenium data and stored it in xenium.obj[["cropped"]]. However, I was not able to access the metadata of the cropped subset. Can I know how would you use the subset_opt function to subset the xenium.obj based on the cropped image?

Thanks,
Kelvin

[ilovegreentea]

Oh I got it. I'm going to send my code. I used barcode information in ***@***.******@***.*** xenium_obj<- Crop( xenium_obj [["fov"]], x = c(300, 2500), y = c(000, 2300), coords = "plot") xenium_obj[[" xenium_obj _1"]] <- xenium_obj _1 ***@***.***$sample_name <- ifelse( xenium_obj @meta.data$barcode %in% xenium_obj ***@***.******@***.***, "annotation" , ***@***.***$sample_name ) I hope I get it right this time. Thanks, KyuTae 2024년 5월 20일 (월) 오후 2:56, Kelvin Mo ***@***.***>님이 작성:

…

Hi @ilovegreentea <https://github.com/ilovegreentea> Sorry for the confusion. So what I did was to crop the image from the xenium data and stored it in xenium.obj[["cropped"]]. However, I was not able to access the metadata of the cropped subset. Can I know how would you use the *subset_opt* function to subset the xenium.obj based on the cropped image? Thanks, Kelvin — Reply to this email directly, view it on GitHub <#7947 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXOTFKOWWNKYXCVM7HVKO73ZDJBNHAVCNFSM6AAAAAA6UVLX6WVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TIOJZHA4DO> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[kelvinmo0513]

Thank you so much @ilovegreentea ! This is helpful.

One side question: After you crop the image and place it in xenium.obj[["xenium_obj_1"]], how do you access the cell identities or barcode information? For me, it's just an image under the "fov" of the xenium object without any info.

[ilovegreentea]

Why is my code obscured?? ***@***.******@***.*** 2024년 5월 20일 (월) 오후 3:17, Kelvin Mo ***@***.***>님이 작성:

…

Thank you so much @ilovegreentea <https://github.com/ilovegreentea> ! This is helpful. One side question: After you crop the image and place it in xenium.obj[["xenium_obj_1"]], how do you access the cell identities or barcode information? For me, it's just an image under the "fov" of the xenium object without any info. — Reply to this email directly, view it on GitHub <#7947 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AXOTFKPGIEIIFZZKVWEBDATZDJD3VAVCNFSM6AAAAAA6UVLX6WVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TKMBQGAZTQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[ilovegreentea]

xenium_obj_1(at sign)boundaries$centroids(at sign)cells

[kelvinmo0513]

Omg thank you so much! This solves my issue. You're a life saver:)

[aksaikumarjayalatha]

Hi I am sorry but I am unable to follow these codes as they're obsecured. For the same setup of Kelvin, how can I subset my Xenium data, using the cropped coordinates information?
From this code xenium_obj<- Crop( xenium_obj [["fov"]], x = c(300, 2500), y = c(000, 2300), coords = "plot") xenium_obj[[" xenium_obj _1"]] <- xenium_obj _1 ***@***.***$sample_name <- ifelse( xenium_obj @meta.data$barcode %in% xenium_obj ***@***.******@***.***, "annotation" , ***@***.***$sample_name )

What is this @.***$sample_name?

Thank you!

[HansLambda]

You can use the standard subset function for Seurat objects. Once the cropped image is stored as "zoom", just pull out the cells.

cropped.cells <- xenium.obj@images$zoom$centroids@cells
cropped.obj <- subset(xenium.obj, cells = cropped.cells)

[msraredon]

My solution to this has been to build a new metadata column just for this operation, because subsetting on metadata works just fine.

So, if we have a big object called 'sub' and we want to make a small object called 'demo', defined by a custom FOV/crop:

define small crop

cropped.coords <- Crop(sub[["PNX3_M2_Sp"]],
x = c(6900, 7075),
y = c(4450, 4600),
coords = "plot")

set a new field of view (fov)

sub[["demo"]] <- cropped.coords

isolate demo into solitary object

initialize new metadata column

sub$demo.cells <- NA

identify cells to pull

cropped.cells <- sub@images$demo$centroids@cells

annotate the empty metadat acolumn to identify these desired cells as a group

sub$demo.cells[cropped.cells] <- 'demo'

subset based on this group value within the new metadata column

demo <- subset(sub, subset = demo.cells == 'demo')

inspect output

demo