nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

JMP Genomics r2 >1.0

ulipkin

How can it be, that I am getting CorrCoeff2 values above 1.0, up to 2.0?

Accepted answers

All comments

statman

Can't tell from what you posted. Could you provide some info? What version of JMP, which OS, and perhaps screen shots or the journal of the analysis you performed.

ulipkin

Thanks for the reply,

Attached is an Excel file telling the story. I am using JMP Genomics whose
data are presented in the first worksheet 'JMP'. I submitted the genotypes
presented in the worksheet 'Genotypes' to the LD analysis (worksheet 'JMP
Linkage Disequilibrium'), and got the LD r2 matrix presented in the last
worksheet. Values above 1.0 are labeled. By definition, r2 cannot have a
value above 1.0. I am using JMP Genomics for some years, many LD analyses,
and I never saw something like this. What is going on?

Thanks,

Ehud Lipkin

Thierry_S

Hi,
I think you forgot to include the file(s) you referred to in your last message or there might be a problem with the JMP forum attachment module.
Best,
TS

ulipkin

Hello TS,

It was attached, and attached again. If you didn't get it again, please let
me what to do.

All the best,

Ehud Lipkin

Chris_Kirchberg

Hi @ulipkin,

Can you tell us which version of JMP Genomics you are using? Look under Help>About JMP

Also, try uploading the file as a .csv

Thanks,

Chris_Kirchberg

Also, it would help if you could let us know the options you have selected in each of the tabs of the Linkage Disequilibrium dialog.

ulipkin

Hi Chris,

Found the problem, my old recording of missing genotypes by "n/n". I am always deleting it in JMP, but forgot this time. Now all is well. Nevertheless, (1) there should be a warning of invalid genotypes, and (2) even with these, r2 can't be above 1.0.

Regards,

Ehud Lipkin

Chris_Kirchberg

That is correct. Correlation Coefficients cannot be >1 or <-1 and thus Correlation Coefficients squared cannot be >1. I agree that a message that invalid genotypes should throw a message and it is likely the cause of the R2 issue.

Could you let me know what version of JMP Genomics you are on so I can report this to development?

Thanks,