pyDESeq2 Index Error?

Question

[Emily]

I was trying to do a DEG analysis using DESeq2 but keeps giving me InvalidIndexError when my index object shouldn't be a problem running the pyDESeq2 package. Here is my code:

counts = pd.read_csv('filename.csv')
counts = counts.set_index('GeneID')
counts = counts.T

metadata = pd.DataFrame(zip(counts.index, ['Ctr','Ctr','Ctr',Ctr', 'KO','KO','KO','KO', 'C2','C2','C2','C2'], columns = ['Sample', 'Condition'])

metadata = metadata.set_index('Sample')

dds = DeseqDataSet(counts=counts, metadata = metadata, design_factors ="Condition")
dds.deseq2() # at this step is where I get "InvalidIndexError"

I did get a message when running DeseqDataSet command

/Users/anaconda3/lib/python3.11/site-packages/anndata/_core/anndata.py:1900: UserWarning: Variable names are not unique. To make them unique, call `.var_names_make_unique`.
  utils.warn_names_duplicates("var")
/Users/anaconda3/lib/python3.11/site-packages/pydeseq2/dds.py:257: UserWarning: Some factor levels in the design contain underscores ('_').
                    They will be converted to hyphens ('-').
  self.obsm["design_matrix"] = build_design_matrix(

I'm wondering if that step is preventing me from executing dds.deseq2() Do I need to make variable names unique or can I ignore that message?

[Mark]

Try to make the variable names unique and see if it fixes the issue. I would have done just that before posting. Also you need to post the full error message, which will help people assist you.

[Emily]

I figured it out, the duplicates were causing the problems so I removed the ones that were low expressing.

[ATpoint]

It's concerning your matrix even has dups, you should figure out why that is.