liangyy / deep_variant_annotation Goto Github PK

Hi @liangyy i took a look of your setup.sh on RCC. It seems you've configured a very specific environment that I cannot make it work easily on other machines for people without RCC access. I'm wondering if you are interested in helping me putting together a proper docker configuration to make your pipeline a lot more portable? For example eg I've done for the mash paper:

https://github.com/stephenslab/gtexresults/blob/master/workflows/mash.dockerfile

that case people do not need to worry about installing other software than docker itself.

I'm asking because one of our colleagues (you surely know her xD) is going to need to run this on her desktop. Would be good if we can sort this out one way or another. Thanks!

I got an error like this:

rule seq2input:
    input: /home/gaow/Documents/GIT/github/atac-gwas/test_var_deepsea/check.passed.ALT.tab.gz
    output: /home/gaow/Documents/GIT/github/atac-gwas/test_var_deepsea/input.ALT.hdf5
    jobid: 9
    wildcards: output_dir=/home/gaow/Documents/GIT/github/atac-gwas, name=test_var_deepsea, type=ALT


Traceback (most recent call last):
  File "/opt/deepann/seq2input.py", line 48, in <module>
    digit_seq[i, encode[seq[i]]] = 1
IndexError: index 1000 is out of bounds for axis 0 with size 1000

And looking at this line of code:

Just from reading the code it seems that if args.window is less than len(seq) there will be index overflow error. Is that the case? I'm not sure how to propose a fix because I'm not seeing how args.window is useful in this script. ...