ValueError: too many values to unpack (expected 2)

[kayihui]

Hi scDRS team,

I got the following error message. Please let me know how to resolve this.

Thank you. Ka Yi

scdrs compute-score \
    --h5ad-file NAc.combinded.h5ad \
    --h5ad-species mouse \
    --gs-file processed_geneset.gs \
    --gs-species mouse \
    --cov-file None \
    --flag-filter-data False \
    --flag-raw-count False \
    --flag-return-ctrl-raw-score False \
    --flag-return-ctrl-norm-score False \
    --out-folder data/
******************************************************************************
* Single-cell disease relevance score (scDRS)
* Version 1.0.2
* Martin Jinye Zhang and Kangcheng Hou
* HSPH / Broad Institute / UCLA
* MIT License
******************************************************************************
Call: scdrs compute-score \
--h5ad-file NAc.combinded.h5ad \
--h5ad-species mouse \
--cov-file None \
--gs-file processed_geneset.gs \
--gs-species mouse \
--ctrl-match-opt mean_var \
--weight-opt vs \
--adj-prop None \
--flag-filter-data False \
--flag-raw-count False \
--n-ctrl 1000 \
--flag-return-ctrl-raw-score False \
--flag-return-ctrl-norm-score False \
--out-folder data/

Loading data:
--h5ad-file loaded: n_cell=3579, n_gene=55416 (sys_time=1.9s)
First 3 cells: ['N1_AAACCCACATCTGTTT', 'N1_AAAGAACAGCACCGAA', 'N1_AAAGAACCAAGGTACG']
First 5 genes: ['Gm37671', 'Gm19087', 'Gm8941', 'Gm38212', 'Gm7449']
Traceback (most recent call last):
  File "/opt/anaconda3/envs/scDRS_py3_8/bin/scdrs", line 740, in <module>
    fire.Fire()
  File "/opt/anaconda3/envs/scDRS_py3_8/lib/python3.8/site-packages/fire/core.py", line 143, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/opt/anaconda3/envs/scDRS_py3_8/lib/python3.8/site-packages/fire/core.py", line 477, in _Fire
    component, remaining_args = _CallAndUpdateTrace(
  File "/opt/anaconda3/envs/scDRS_py3_8/lib/python3.8/site-packages/fire/core.py", line 693, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/opt/anaconda3/envs/scDRS_py3_8/bin/scdrs", line 186, in compute_score
    dict_gs = scdrs.util.load_gs(
  File "/Users/kayi/scDRS/scdrs/util.py", line 239, in load_gs
    for i, (trait, gs) in df_gs.iterrows():
ValueError: too many values to unpack (expected 2)

[martinjzhang]

Hi,

It seems your .gs file doesn't have the correct format. Can you double check it to make sure it follows the same format as the example .gs file? https://martinjzhang.github.io/scDRS/file_format.html

[kayihui]

Hi Martin,

Thank you for your quick reply. The .gs file is a subset of the .gs file from the paper. I visually inspected the file, and didn't seem to see a difference.

I have rerun using the original magma_10kb_top1000_zscore.74_traits.rv1.gs file. It's working without error.

I attached the code for subsetting the .gs file.

# Define the subset and the rename dictionary
subset_values = ["PASS_MDD_Howard2019", "PASS_BIP_Mullins2021", "PASS_SleepDuration_Dashti2019", "UKB_460K.other_MORNINGPERSON"]
rename_dict = {
    "PASS_MDD_Howard2019": "MDD",
    "PASS_BIP_Mullins2021": "BP",
    "PASS_SleepDuration_Dashti2019": "SleepDuration",
    "UKB_460K.other_MORNINGPERSON": "MorningPerson"
}

# Subset the DataFrame
subset_df = df_gs[df_gs['TRAIT'].isin(subset_values)]

# Rename the values in the "Trait" column
subset_df['TRAIT'] = subset_df['TRAIT'].replace(rename_dict)

subset_df.to_csv("processed_geneset.gs", sep="\t")

Thank you very much, Ka Yi

[martinjzhang]

Can you remove the first indexing column from the gs file and try running the code again?

[kayihui]

It worked without the error. However, here's the result:

Computing scDRS score:
trait=BP: skipped due to small size (n_gene=0, sys_time=2.8s)
trait=MDD: skipped due to small size (n_gene=3, sys_time=2.8s)
trait=SleepDuration: skipped due to small size (n_gene=0, sys_time=2.8s)
trait=MorningPerson: skipped due to small size (n_gene=1, sys_time=2.8s)

Why it looks like the number of gene in the list is not correct?

[martinjzhang]

Yes, it seems scDRS didn't recognize the genes. Did you specify the genes to be human genes?

[kayihui]

my dataset is mouse, do I have to convert the gene name in the list?

[martinjzhang]

But I see all your genes are upper-case, suggesting the disease genes are human genes. Use this argument to specify the genes to be human genes --gs-species human scDRS will do the auto-conversion.

[kayihui]

Great! Everything is working fine now. Thank you!