Eigen Decompsition

[Sagnik-Epigen]

I was trying the pipeline with HT1080 HI-C data (.mcool generated from distiller.nf). Unfortunately, eigdecomp rule never worked. I think it is the problem with partition. I am sending you my SRA code (SRR7586416, SRR7586418). Please tell me what to do.

[nvictus]

Hello. Can you post the traceback of the error you see?

[Sagnik-Epigen]

Yeah sure... first one when I have to tried to run it in vscode individually each rule. Second one, is within the conda environment.

[Sagnik-Epigen]

the assembly was hg38

[nvictus]

The chromosome order in your coolers might be different than what is expected.

What is the output of cooler dump -t chroms {your_data}.mcool::resolutions/50000 ?

[Sagnik-Epigen]

Ok. What is the solution? I am sending you the cooler dump output
HT1080.txt

[Sagnik-Epigen]

Need to go for sorting?

[nvictus]

Easiest thing is to rebuild your coolers from pairs using a properly sorted chromsizes file:

https://github.com/open2c/distiller-nf/blob/master/sorted_chromsizes/hg38.reduced.chrom.sizes

[Sagnik-Epigen]

Ok. let me try that. Thanks.

[Sagnik-Epigen]

I got a new error

[nvictus]

Did you "look above for the error message" as it says?

[Sagnik-Epigen]

Yeah but it's just the log like the previous image

[nvictus]

The log should contain somewhere the actual error that caused the workflow to terminate. You can browse the complete log file printed on the last line of what you posted.

There's no way I can solve your issue without that information.

[Sagnik-Epigen]

This is the complete log

[nvictus]

Hmm. That's strange. Looks like it might be a resource problem. What are the specs on your machine, cores & memory?

[Sagnik-Epigen]

I am working in cluster under conda environment. We can go as much as 40 vCPUs core and maximum RAM = 372 GB