model.train_df_and_spatial_deconvolution error

[zhangqi234]

Hi, thanks for coming up with this useful tool. When I conducted the model.train_df_and_spatial_deconvolution function to decompose ST data into spatially resolved single-cell transcriptomics data, I found the error like "pandas.errors.MergeError: No common columns to perform merge on. Merge options: left_on=None, right_on=None, left_index=False, right_index=False". I don't know what caused this error.

[AlexUOM]

Hi, I am also experiencing an error with the model.train_df_and_spatial_deconvolution function. In my case the error is ValueError: no types given as you can see below:

I am running the pdac tutorial with all the pre-provided data. Hope you can help with this.

[ForwardYang98]

Hi, thanks for coming up with this useful tool. When I conducted the model.train_df_and_spatial_deconvolution function to decompose ST data into spatially resolved single-cell transcriptomics data, I found the error like "pandas.errors.MergeError: No common columns to perform merge on. Merge options: left_on=None, right_on=None, left_index=False, right_index=False". I don't know what caused this error.

Hi, it seems there are some errors in your data format. You could refer to the Tutorial Handbook to check the format of input again.

[ForwardYang98]

ValueError: no types given

Hi, I have run the PDAC tutorial successfully again just now. I guess there are some errors in the format of your data, maybe you could check it by following the Tutorial Handbook

[AlexUOM]

Hi, I have run the PDAC tutorial successfully again just now. I guess there are some errors in the format of your data, maybe you could check it by following the Tutorial Handbook

Thank you for your feedback. I double checked again the format of the data as you adviced and everything is looking fine.

Here is a quick look of how the 4 input files for model.train_df_and_spatial_deconvolution look like (using .head()) generate_sc_data:

generate_sc_meta:

st_data.csv:

st_meta.csv:

Is this what they are supposed to look like? I suspect the files that throw the error are the st reference files (either st_meta or st_data, not sure which one).

[ForwardYang98]

Hi, I have run the PDAC tutorial successfully again just now. I guess there are some errors in the format of your data, maybe you could check it by following the Tutorial Handbook

Thank you for your feedback. I double checked again the format of the data as you adviced and everything is looking fine.

Here is a quick look of how the 4 input files for model.train_df_and_spatial_deconvolution look like (using .head()) generate_sc_data:

generate_sc_meta:

st_data.csv:

st_meta.csv:

Is this what they are supposed to look like? I suspect the files that throw the error are the st reference files (either st_meta or st_data, not sure which one).

According to the screenshots you provided, the column number of st_data (428 columns) is not same as the row number of st_meta (429 rows). The spots in st_data and st_meta are supposed to be same for Bulk2Space, as shown below: st_data.shape (19104, 428) st_meta.shape (428, 3)

[AlexUOM]

The number of rows and columns in the files is different you are correct. To resume, st_meta has 428 rows x 4 columns:

...while st_data has 19104 rows x 429 columns :

The reason why st_data has an extra column is because it counts the first column (without a name) that contains the gene names. Could this be the reason of the error?

[ForwardYang98]

The number of rows and columns in the files is different you are correct. To resume, st_meta has 428 rows x 4 columns:

...while st_data has 19104 rows x 429 columns :

The reason why st_data has an extra column is because it counts the first column (without a name) that contains the gene names. Could this be the reason of the error?

Yes. You could set the first column (gene names) as the index and then try it again.