ValueError: Invalid expression matrix in .X. nb mode expects raw UMI counts in .X of the AnnData."

[wuxiangning]

hello i have the same issue,when run tnode = sct.train.Trainer(adata)tnode = sct.train.Trainer(adata)
its error like this "ValueError Traceback (most recent call last)
Cell In[22], line 1
----> 1 tnode = sct.train.Trainer(adata)

File ~\AppData\Roaming\Python\Python312\site-packages\sctour\train.py:168, in Trainer.init(self, adata, percent, n_latent, n_ode_hidden, n_vae_hidden, batch_norm, ode_method, step_size, alpha_recon_lec, alpha_recon_lode, alpha_kl, loss_mode, nepoch, batch_size, drop_last, lr, wt_decay, eps, random_state, val_frac, use_gpu)
166 X = self.adata.X.data if sparse.issparse(self.adata.X) else self.adata.X
167 if (X.min() < 0) or np.any(~np.equal(np.mod(X, 1), 0)):
--> 168 raise ValueError(
169 f"Invalid expression matrix in .X. {self.loss_mode} mode expects raw UMI counts in .X of the AnnData."
170 )
172 self.n_cells = adata.n_obs
173 self.batch_size = batch_size

ValueError: Invalid expression matrix in .X. nb mode expects raw UMI counts in .X of the AnnData."

but my .X and .raw.X all integers and my seurat counts are also integers，.help!thk!

[LiQian-XC]

Can you run the following code on your .X? If it returns True, it then means that not all of the values in .X are integers and you need to check what causes some non-integer values in your .X.

np.any(~np.equal(np.mod(X, 1), 0))

[wuxiangning]

I run it np.any(~np.equal(np.mod(adata.x, 1), 0)) this code seems to check if adata.x is an integer
It return error like this:
TypeError: unsupported operand type(s) for %: 'csr_matrix' and 'int'
this error seems my adata.X type isnt right,why?

This adata is convert from seurate by using MuDataSeurat::WriteH5AD(seurate, "F:/数据分析/R.h5ad", assay="RNA").
Is there a problem with this conversion function?but this adata can use Scanpy as normal! Thanks!

[LiQian-XC]

Can you try this np.any(~np.equal(np.mod(adata.X.data, 1), 0))?

[wuxiangning]

Sure 。It seems something is wrong in my anadata.because when i use the standard data . It return false

[wuxiangning]

[wuxiangning]

I dont know why my adata has this wrong,because it seems every object is right and ture

[LiQian-XC]

I think you can first check which values in your .X are non-integers. Then you can go back to your original Seurat object to see where the issue occurred.

[wuxiangning]

I don't quite know how to determine which value makes my .X are non-integers. Can you help me!help!!!

[LiQian-XC]

One quick, though possibly inefficient, way is:

np.where(~np.equal(np.mod(adata.X.toarray(), 1), 0))

The first array returned is the row index and the second one is column index for non-integer values in .X

[wuxiangning]

It looks like there are a lot of non-integers, how do I go and change these in Seurate