Issue with generateTypicalParamTable() and pd.concat function

[drewjbeh]

Hi there,

While trying to run the pipeline outlined in Library_design_walkthrough.ipynb, I ran into an error at line [9] paramTable_trainingGuides = generateTypicalParamTable(libraryTable_training,sgInfoTable_training, tssTable, p1p2Table, genomeDict, bwhandleDict)

The error I got was
AssertionError: Cannot concat indices that do not have the same number of levels coming from the pd.concat function.

I went manually, line by line through the code for generateTypicalParamTable() and eventually found the problem was with doubleBaseTable. It seems that this table is created with a MultiIndex for columns, and so when you try concat it to other tables which all have single level indices for columns you get this error. Is there something I am doing wrong?

[drewjbeh]

Hi there,

Still really need a solution for this. Get the feeling this repo isn't being maintained anymore?

[Gowthamvarma30]

Faced similar issue, your information is helpful.
Did you make any further progress on this one?

Is there any alternative git for this process?

[inesHellmann]

Hi there,
just ran into exactly the same issue. Pandas is happier if one if sort=True is added to the pd.concat in generateTypicalParamTable. However, concat that I have does not collapse the multiple columns of an object, as seems to be assumed in the version used here -- which means that there are too few column names -- ( there are 3 accessibility columns, not only 1). Anybody knows when that behavior changed?
It would be really helpful to get the package versions, of the working script.

[MaxActionWang]

Hi there,

I have similar problems with this line of codes about parameter calculation.

paramTable_trainingGuides = generateTypicalParamTable(libraryTable_training,sgInfoTable_training, tssTable, p1p2Table, genomeDict, bwhandleDict)

the errors were like

....
<string>:619: RuntimeWarning: Mean of empty slice
<string>:595: RuntimeWarning: All-NaN slice encountered
<string>:619: RuntimeWarning: invalid value encountered in float_scalars
......
ValueError: Wrong number of items passed 0, placement implies 18380

It seems that the problem is about pandas module and my pandas version is 0.19.2.

So can anyone tell me your package versions so that I could use the correct one. Thanks.

[jsacco1]

Hi there,

I have similar problems with this line of codes about parameter calculation.

paramTable_trainingGuides = generateTypicalParamTable(libraryTable_training,sgInfoTable_training, tssTable, p1p2Table, genomeDict, bwhandleDict)

the errors were like

....
<string>:619: RuntimeWarning: Mean of empty slice
<string>:595: RuntimeWarning: All-NaN slice encountered
<string>:619: RuntimeWarning: invalid value encountered in float_scalars
......
ValueError: Wrong number of items passed 0, placement implies 18380

It seems that the problem is about pandas module and my pandas version is 0.19.2.

So can anyone tell me your package versions so that I could use the correct one. Thanks.

Make sure you have ViennaRNA 2.2.5 installed.

When this line runs

subprocess.call('RNAfold --noPS < %s > %s' % (tempfile_fa.name, tempfile_rnafold.name), shell=True) the tempfile_rnafold.name is likely empty as you do not have ViennaRNA installed and in your path