How to create a Seurat Object from 10x Visium data if the feature-barcodes dataset is not in hdf5 file format?

[ssingh95jhu]

I am trying to create a Seurat Object from 10x Visium data. I have the following files for the tissue of interest:

1. matrix.mtx.gz
2. barcodes.tsv.gz
3. features.tsv.gz
4. and a folder called "spatial" containing the relevant files: scale factors_json.json, tissue_hires_image.png, tissue_positions.csv

I am following the steps provided here: https://satijalab.org/seurat/reference/load10x_spatial

However, I do not have my 10x Visium feature-barcodes matrix (data) in the hdf5 format (i.e., filtered_feature_bc_matrix.h5)

The strategy which I followed was:

1. First, created a Seurat object using the Read10X function using the matrix.mtx.gz, barcodes.tsv.gz and features.tsv.gz files.
2. Then, I tried to add the images to the above Seurat object but was not successful. This prevents me from implementing functions like SpatialFeaturePlot or SpatialDimplot.

Is there a work around for this? * I came across similar discussion threads here but couldn't find a solution *
I would really appreciate if you have any suggestions.

Regards

[lvmt]

hi
i meet the same question and i had solved it.

the error is because the 'tissue_lowres_image.png' lost

the Solution：
1 cp tissue_hires_image.png tissue_lowres_image.png
2 modify scalefactors_json.json file, let tissue_lowres_scalef.value = tissue_hires_scalef.value

matrix_dir = 'filtered_feature_bc_matrix/' 
counts = Seurat::Read10X(data.dir = matrix_dir)  

data = Seurat::CreateSeuratObject(
    counts = counts , 
    project = 'test', 
    assay = 'Spatial')

data$slice = 1 
data$region = 'test' 

imgpath = "spatial"
img = Seurat::Read10X_Image(image.dir = imgpath)  

Seurat::DefaultAssay(object = img) <- 'Spatial'  

img = img[colnames(x = data)]  
data[['image']] = img  

SpatialFeaturePlot(data, features = "nCount_Spatial")

hope this can help U.

[ssingh95jhu]

Thanks lvmt for your suggestion! I took your advice and made some changes to my workflow and achieved partial success.

1. Found out the "tissue_lowres_image.png" file and pasted it back in the "spatial" folder already containing the "tissue_hires_image.png" file. Never made any changes to the "scalefactors_json.json" file as you had suggested.

2. Implemented the code as provided by you. SpatialFeaturePlot function seems to be working when I pass the features argument, features="nCount_Spatial". (Image attached: It's a human kidney biopsy tissue)
   

3. But when I change it to a gene name, for example: features = "LCN2", I get an error as follows:
   Warning: No layers found matching search pattern provided
   Error in FetchData.Assay5(object = object[[DefaultAssay(object = object)]], :
   layer "data" is not found in the object

Have I made any mistake or missing anything here? Once again thanks so much for your help!

[lvmt]

hi,

you had successed.
the error 'layer "data" is not found in the object' is abot data processing。

because the created seurat obj just contain count layer

follow this tutorial https://satijalab.org/seurat/articles/pbmc3k_tutorial and start from Standard pre-processing workflow part

hope this can help U.