Difference in test3.bed for cyp2d6 hg38

[abdulfaqihalm]

Describe the bug I got warning on the call_star process with the following log:

Command error: Traceback (most recent call last): File "bin/stellarpgx.py", line 29, in cn = get_total_CN(cov_file)[0] File "/home/faqih/Documents/pgx_prototype/scripts/stellarpgx/cyp2d6/hg38/bin/sv_modules.py", line 24, in get_total_CN av_2d7_ex2_in8 = float(all_reg[9][3])/(float(all_reg[9][2]) - float(all_reg[9][1])) IndexError: list index out of range

I know that the default of errorStrategy for call_stars is ignore. However, this results in not producing the expected outcomes

To Reproduce I change from a test profile to a standard profile

Expected behavior Expecting the non-test behave correctly

Screenshots Somehow when I dug down into the code, I found that the test3.bed is different between the test and standard profile res_init.

Additional context I created my own test data where I only have references of chr6, 19, and 22 to simulate some PGx variants (HLA, cyp2d6, and others) in my local machine. Is it expected? Since, as my understanding of the get_depth, it tries to find the depth of the region overlapped with the bed file. However, not sure why we need to put the reference here. My hypothesis is that since my reference file does not include chr 7 and 12, the outputted depth data is not complete (less than 11).

[twesigomwedavid]

@abdulfaqihalm

Thanks. Yes, you need to have coverage on the EGFR (on chr7) and VDR (on chr12) genes. StellarPGx is expecting this (control region data) to help with computing the copy number changes in the target genes. Also, at present we don't have support for HLA.

David

[abdulfaqihalm]

@twesigomwedavid, Thank you for your reply! For the HLAs and other components, I am utilizing different tools, not StellarPGx. I mentioned this only to provide context for the regions I am using in my case.

Thanks again for the clarity