search

aertslab / scenicplus

SCENIC+ is a python package to build gene regulatory networks (GRNs) using combined or separate single-cell gene expression (scRNA-seq) and single-cell chromatin accessibility (scATAC-seq) data.
Other
165 stars 27 forks source link

normalize_scores() process is killed #225

Closed elesanesc closed 9 months ago

elesanesc commented 9 months ago

Hello,

I am using SCENIC+ package and I am trying to determine the differentially accessibile regions (DARs), but when running normalize_scores() the process get killed. I do not have any error message, just "killed".

Here is the code I run and the output.

>>> from pycisTopic.diff_features import *
>>> imputed_acc_obj = impute_accessibility(cistopic_obj, selected_cells=None, selected_regions=None, scale_factor=10**6)
2023-08-10 16:15:01,691 cisTopic     INFO     Imputing region accessibility
2023-08-10 16:15:01,691 cisTopic     INFO     Impute region accessibility for regions 0-20000
2023-08-10 16:15:03,560 cisTopic     INFO     Impute region accessibility for regions 20000-40000
2023-08-10 16:15:05,387 cisTopic     INFO     Impute region accessibility for regions 40000-60000
2023-08-10 16:15:07,171 cisTopic     INFO     Impute region accessibility for regions 60000-80000
2023-08-10 16:15:09,044 cisTopic     INFO     Impute region accessibility for regions 80000-100000
2023-08-10 16:15:10,929 cisTopic     INFO     Impute region accessibility for regions 100000-120000
2023-08-10 16:15:12,774 cisTopic     INFO     Impute region accessibility for regions 120000-140000
2023-08-10 16:15:14,657 cisTopic     INFO     Impute region accessibility for regions 140000-160000
2023-08-10 16:15:16,527 cisTopic     INFO     Impute region accessibility for regions 160000-180000
2023-08-10 16:15:18,415 cisTopic     INFO     Impute region accessibility for regions 180000-200000
2023-08-10 16:15:20,311 cisTopic     INFO     Impute region accessibility for regions 200000-220000
2023-08-10 16:15:22,206 cisTopic     INFO     Impute region accessibility for regions 220000-240000
2023-08-10 16:15:23,962 cisTopic     INFO     Impute region accessibility for regions 240000-260000
2023-08-10 16:15:25,665 cisTopic     INFO     Impute region accessibility for regions 260000-280000
2023-08-10 16:15:27,356 cisTopic     INFO     Impute region accessibility for regions 280000-300000
2023-08-10 16:15:29,049 cisTopic     INFO     Impute region accessibility for regions 300000-320000
2023-08-10 16:15:30,732 cisTopic     INFO     Impute region accessibility for regions 320000-340000
2023-08-10 16:15:32,617 cisTopic     INFO     Impute region accessibility for regions 340000-360000
2023-08-10 16:15:34,497 cisTopic     INFO     Impute region accessibility for regions 360000-380000
2023-08-10 16:15:36,394 cisTopic     INFO     Impute region accessibility for regions 380000-400000
2023-08-10 16:15:36,899 cisTopic     INFO     Done!
>>> 
>>> # Log-normalize the imputed data
>>> normalized_imputed_acc_obj = normalize_scores(imputed_acc_obj, scale_factor=10**4)
2023-08-10 16:15:36,900 cisTopic     INFO     Normalizing imputed data
Terminado (killed)

I am working with ≈16000 cells in total from 4 different samples which I have merged.

Does anyone know why this error is happening and how to resolve it?

Thank you very much for your help.

Best, Elena

SeppeDeWinter commented 9 months ago

Hi @elesanesc

You probably ran out of memory, this step is quite memory intense. Could you retry on a node/machine with more memory?

Best,

Seppe

elesanesc commented 9 months ago

Hi @SeppeDeWinter

I have checked and it is what you said, a memory issue. I am working with 64Gb RAM. I will try to increase my RAM or change to another server.

Thank you very much for your help.

Best, Elena

SeppeDeWinter commented 9 months ago

No worries. Good luck with the analysis.

Best,

Seppe