Error: "None of ['WhiteListMatch'] are in the columns"Traceback:

[Selecton98]

Hello the Gotcha authors,

I'm trying your genotyping algorithm and encounter the following error:

"None of ['WhiteListMatch'] are in the columns"

Traceback:

1. GotchaLabeling(path = "/lab-share/Public/data/DNV/barcode3/",
   . infile = "Genemeta.csv", gene_id = "Gene", sample_id = "RM30")
2. reticulate::import_from_path(module = "gotcha_labeling", path = find.package("Gotcha"))$GotchaLabeling(path = path,
   . infile = infile, gene_id = gene_id, sample_id = sample_id)
3. py_call_impl(callable, call_args$unnamed, call_args$named)

My file has a format matching your demo:

Could you please tell me in which file should I 'WhiteListMatch'
Thank you,
Sunny

[fizzo13]

Hi Sunny, thank you for using GoTChA. Can you try replacing the column name from 'X' to "WhiteListMatch"?
Tagging @SanjayKottapalli for help with the python module.

[Selecton98]

Hi Sunny, thank you for using GoTChA. Can you try replacing the column name from 'X' to "WhiteListMatch"? Tagging @SanjayKottapalli for help with the python module.

Hello! Thank you for the reply. I tried to change 'X' to 'WhiteListMatch' in the list but it didn't help.
@SanjayKottapalli

[SanjayKottapalli]

Hi Sunny, I think the issue is actually with the "gene_id" and "sample_id" parameters that you set. Can you show me exactly how your own input .csv file looks?

[Selecton98]

Genemeta.csv
@SanjayKottapalli
Please find my file in the attachment. I used "RM30" to make this column same as your demo (for easier debugging).
Thank you for your help!

[SanjayKottapalli]

Hmm, I tried to read the file in python and it worked just fine for me. Can you try the following?

In an interactive python session:

1. Navigate to the folder where you installed Gotcha, then cd into inst/python
2. import gotcha_labeling
3. test = gotcha_labeling.read_data("/lab-share/Public/data/DNV/barcode3/Genemeta.csv", "Gene", "RM30")
4. print(test)

Let me know what output this gives and if any errors come up.

[Selecton98]

@SanjayKottapalli The files is successfully read by the "gotcha_labeling.read_data".

[fizzo13]

@Selecton98 was the issue fixed?

[SanjayKottapalli]

Since it seems to be working when using python only and doesn't work when you call python through the R pipeline, it seems like it could be a reticulate issue @fizzo13 . Probably related to formatting of dataframes.

@Selecton98 could you also try commenting/changing this code in your gotcha_labeling.py from

try:
genotyping = pd.DataFrame(index=cell_line.index)
if sample_id in np.unique(cell_line['Sample'].values):
cell_line = cell_line.loc[cell_line['Sample']==sample_id, :]
genotyping['WTcount'] = cell_line[gene_id+'_WTcount']
genotyping['MUTcount'] = cell_line[gene_id+'_MUTcount']
except:
cell_line = cell_line.set_index("WhiteListMatch")
genotyping = pd.DataFrame(index=cell_line.index)
genotyping['WTcount'] = cell_line['WTcount']
genotyping['MUTcount'] = cell_line['MUTcount']

to only what's in the try block?

genotyping = pd.DataFrame(index=cell_line.index)
if sample_id in np.unique(cell_line['Sample'].values):
cell_line = cell_line.loc[cell_line['Sample']==sample_id, :]
genotyping['WTcount'] = cell_line[gene_id+'_WTcount']
genotyping['MUTcount'] = cell_line[gene_id+'_MUTcount']

The error is strange because the try block shouldn't fail to begin with.

[Selecton98]

@SanjayKottapalli
@fizzo13
Hi everyone,
I finally get this bug solved.

One problem is the R "GotchaLabeling" doesn't need a "sample_column" information which is required by the python "GotchaLabeling" and "read_data". So I manually defined "sample_column='Sample'" in each of these two functions.

Another problem is the csv file should include an index column and a "WhiteListMatch" column. Both of these two columns should be index/barcode information.

Thank you,
Sunny

[fizzo13]

Thank you @SanjayKottapalli and @Selecton98 . I'll close the issue.