Error in if (currH <= zeroLow) { : missing value where TRUE/FALSE needed

[xingguowu]

Dear gao,
Thanks for your great software!
Here I got some issue about running FastEPRR 2.0 at step2.
This is the error_log.

This is the output_file in /step1.

I get an error when I run the second step directly, but when I "grep NA" to process the result file from the first step, I can avoid this error, even if the result is not complete.

Could you please help me solve this problem?

[zhou-sumei]

Hi,
Thank you for your interest in this software! We have received this error from other users, we speculate that the error may be due to higher R version. I don't know which version of R you are using, at present the most adaptive version for FastEPRR2.0 is R3.6.3. Maybe this will be a solution.
We are testing the FastEPRR2.1,which can be better adapted to R4.x version, sorry for the inconvenience.

About the "NA" in the output:
NA means this window doesn't contain missing data.

Please let me know if you have any questions.

[xingguowu]

Thank you very much！
I used R4.1.2 because I couldn't install mboost in an R3.6.3 environment.
And I want to know the effect of my removal of non-NA on the step1-results.
This is my email address. I want to communicate with you in more details
xgwu23@cemps.ac.cn

[zhou-sumei]

I guess maybe you set window length as 500bp, it is not very big and I think dop these window (contain missing data) will not have a significant impact on the results.

[Mimi33Mimi]

Dear developers,

I would like to let you know my situation in R-4.4.2(I know you have recommended R-3.6.3.).
I got the same error but for a different reason in OP, and I finally seem to have solved it.

I checked the source code and found that estimateRho function in FastEPRR_VCF_step2.R did not treat non-numeric chromosome codes appropriately (After assignment to aWin variable, all aWin are NA, then currH <= zeroLow goes NULL.).
Then I renamed chromosome codes in my outputs of step1, it worked well.

I hope this information will be helpful.

Best Regards.