# Spatial datasets integration

**URL:** <https://discourse.scverse.org/t/spatial-datasets-integration/437>\
**Category:** scvi-tools\
**Tags:** scvi\
**Created:** [May 5, 2022, 2:19am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437 "2022-05-05T02:19:42Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![wangjiawen2013](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/wangjiawen2013/32/25_2.png) [@wangjiawen2013](https://discourse.scverse.org/u/wangjiawen2013)\
**Post date:** [May 5, 2022, 2:19am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437/1 "2022-05-05T02:19:42Z")

</div>

Hi,  
Can I use scvi-tools to integrate spatial transcriptomics datasets ?

---

<div class="post-metadata">

**Author:** ![adamgayoso](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/adamgayoso/32/100_2.png) [@adamgayoso](https://discourse.scverse.org/u/adamgayoso)\
**Post date:** [August 8, 2022, 7:07am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437/2 "2022-08-08T07:07:22Z")

</div>

What kind of integration do you want to do? You could ignoring the spatial information and integrating like normal?

---

<div class="post-metadata">

**Author:** ![wangjiawen2013](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/wangjiawen2013/32/25_2.png) [@wangjiawen2013](https://discourse.scverse.org/u/wangjiawen2013)\
**Post date:** [August 8, 2022, 9:16am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437/3 "2022-08-08T09:16:44Z")

</div>

I wanna integration different spatial datasets without regard to the spaital coordinates, just like what scvi do on single cell datasets.

---

<div class="post-metadata">

**Author:** ![wangjiawen2013](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/wangjiawen2013/32/25_2.png) [@wangjiawen2013](https://discourse.scverse.org/u/wangjiawen2013)\
**Post date:** [November 18, 2022, 9:01am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437/4 "2022-11-18T09:01:51Z")

</div>

cell2location integration different spatial datasets using bbknn, so I wnna know if I can use scvi-tools. There is a close relation between cell2location and scvi.

> <https://github.com/BayraktarLab/cell2location/issues/164#issuecomment-1319718565>
>
> Hi,
> I am using cell2location to deconvolution my single cell datasets. Now I ha…ve four spatial samples and I have concatenated them in one object. Can I run them together using
> cell2location.models.Cell2location.setup\_anndata(adata=adata\_vis, layers="counts",batch\_key="sample") ?
> In the tutorial its "cell2location.models.Cell2location.setup\_anndata(adata=adata\_vis, batch\_key="sample")", but there are only one sample in adata\_vis, so I wanna know if it supports more samples.

---

<div class="post-metadata">

**Author:** ![wangjiawen2013](https://yyz1.discourse-cdn.com/flex035/user_avatar/discourse.scverse.org/wangjiawen2013/32/25_2.png) [@wangjiawen2013](https://discourse.scverse.org/u/wangjiawen2013)\
**Post date:** [November 18, 2022, 9:03am UTC](https://discourse.scverse.org/t/spatial-datasets-integration/437/5 "2022-11-18T09:03:10Z")

</div>

Yes, this is what I mean. I want to ignore the spatial information and integrate like normal. Could I use scvi-tools for this purpose ?