ValueError: max() arg is an empty sequence

[zonzero]

Hello ! When I use this code: SEVtras.sEV_recognizer(input_path='./', sample_file='./skin1/sample_file.txt', out_path='./skin1/
outputs', species='Homo')
to anylsis my 10x-scRNAseq raw file, after a few minutes run it return the error "ValueError: max() arg is an empty sequence".
I try to input another file but still not working.
I need some help, thank you!

[RuiqiaoHe]

This means that SEVtras cannot find a representative gene to identify sEVs. There are two options you can choose. One is that you can try to lower the parameter "alpha", e.g. 0.09. The other is that you can use more samples to get a comprehensive result of representative genes.

[zonzero]

This means that SEVtras cannot find a representative gene to identify sEVs. There are two options you can choose. One is that you can try to lower the parameter "alpha", e.g. 0.09. The other is that you can use more samples to get a comprehensive result of representative genes.

Thank you! I retry and successfully run the code. But in the output file, only have two file, "raw_sample.h5ad" and "skin1_ev.txt". It's very different between your test flie output. There is no "sEV_sample.h5ad" file output. I try to lower the parameter 'alpha' in 0.09, but still no "sEV_sample.h5ad" file. It is strange.
There must something wrong ……

[RuiqiaoHe]

The log file of SEVtras may contain "sEV is hard to detect in xxx". You can check this in the running log file. This means that SEVtras cannot detect a strong sEV signal in your sample. But if you really want to characterize potential sEVs, you can try to set the parameter "score_t" lower, e.g. '10'. Note: this parameter must be a string.

[zonzero]

The log file of SEVtras may contain "sEV is hard to detect in xxx". You can check this in the running log file. This means that SEVtras cannot detect a strong sEV signal in your sample. But if you really want to characterize potential sEVs, you can try to set the parameter "score_t" lower, e.g. '10'. Note: this parameter must be a string.

OK, thanks a lot. And another question is my 'Sample_file' list have 3 sample dir name, but when I run the code it stuck here:

I can confirm that it not working anymore. The code has been running for an hour and has not progressed to the next sample
It is my "Sample_file.txt" file format:
Skin1
Skin2
Skin3
...
All the row is name of a 10x_scRNAseq sample folder which contain the raw output file from cellranger in right pathway.
Is strange because I run your test file It won't stuck.

[RuiqiaoHe]

Could you please first check that the process of SEVtras is still running, and has not been killed or shut down in the cluster? The first sample has been completed, which means that SEVtras can be run successfully in your environment.

[zonzero]

Could you please first check that the process of SEVtras is still running, and has not been killed or shut down in the cluster? The first sample has been completed, which means that SEVtras can be run successfully in your environment.

I finally managed to run the code, although it took a long time. I feel that it may be the problem of the system. I rented the server of the linux native system yesterday, and it succeeded. Previously, it was running in my windows WSL linux, and there were various bugs.

[zonzero]

Could you please first check that the process of SEVtras is still running, and has not been killed or shut down in the cluster? The first sample has been completed, which means that SEVtras can be run successfully in your environment.

And, thanks for you help!

[RuiqiaoHe]

Thanks for your testing. There may be some bugs incompatible with Windows because I develop SEVtras on a Linux server. I will add this in the notes of the SEVtras tutorial.