search

digitalcytometry / cytotrace2

CytoTRACE 2 is an interpretable AI method for predicting cellular potency and absolute developmental potential from scRNA-seq data.
Other
85 stars 6 forks source link

Error : Checking zero-variance data... #25

Closed lvmt closed 3 months ago

lvmt commented 3 months ago

Dear Team:

i use this tool for spatial transcriptomics. use log-normalized data.

the error message 

cytotrace2: Started loading data

Dataset contains 36601 genes and 4199 cells.

Warning message in cytotrace2(expression_data):
"Species is most likely human. Please revise the 'species' input to the function."
Windows OS can run only on 1 core

The passed subsample size is greater than the number of cells in dataset.
Now setting subsample size to 4199

cytotrace2: Running on 1 subsample(s) approximately of length 4199

cytotrace2: Started running on subsample(s). This will take a few minutes.

cytotrace2: Started preprocessing.

11 input genes mapped to model genes.

Warning message in preprocessData(dt, species):
"The number of input genes mapped to the model is too low. Please verify the input species is correct.
In case of a correct species input, be advised that model performance might be compromised due to gene space differences."
cytotrace2: Started prediction.

This section will run using  1 / 16 core(s).
...
cytotrace2: Running with fast mode (subsamples are processed in parallel)

This section will run on 4 sub-sample(s) of approximately 1050 cells each using 1 / 16 core(s).

Output is truncated. View as a [scrollable element](command:cellOutput.enableScrolling?968e0f40-2da5-4152-a032-36a68a1bcaf8) or open in a [text editor](command:workbench.action.openLargeOutput?968e0f40-2da5-4152-a032-36a68a1bcaf8). Adjust cell output [settings](command:workbench.action.openSettings?%5B%22%40tag%3AnotebookOutputLayout%22%5D)...
---> Checking zero-variance data...
--->     Total number of variables:  1049
--->     WARNING: 145 variables found with zero variance

Error in if (mean(abs(cur_score - prev_score))/(mean(init_score) + 1e-06) < : missing value where TRUE/FALSE needed
Traceback:

1. cytotrace2(expression_data)
2. lapply(subsamples, subsample_processing_f)
3. FUN(X[[i]], ...)
4. smoothData(ranked_data, predicted_df, top_genes, ncores = ncores, 
 .     smooth_batch_size = smooth_batch_size, parallelize_smoothing = parallelize_smoothing, 
 .     seed = seed)
5. parallel::mclapply(subsamples, mc.cores = min(chunk_number, ncores), 
 .     get_score)
6. lapply(X, FUN, ...)
7. FUN(X[[i]], ...)
8. smoothing_by_diffusion(nn_pred_order[subsample], markov_matrix)

need your help and thanks.

lvmt commented 3 months ago

i have solve the problem, my sample is from human and i forget add the params species.

the tool is really simple to use.