ordinal phenotype path

75 views
Skip to first unread message

LMB

unread,
Jul 2, 2021, 2:58:30 AM7/2/21
to Genomic SEM Users
If you have ordinal regression or categorical regression GWAS summary statistics (with effect size and their standard errors), what path in 3. would you take? 

Michel Nivard

unread,
Jul 2, 2021, 3:11:17 AM7/2/21
to LMB, Genomic SEM Users
Hi,

Did you run. GWAS where the data was ordinal but you used OLS or did you run a GWAS using a tool that uses propped ordinal regression?

Michel

Op vr 2 jul. 2021 om 08:58 schreef LMB <whatawo...@gmail.com>
If you have ordinal regression or categorical regression GWAS summary statistics (with effect size and their standard errors), what path in 3. would you take? 


--
You received this message because you are subscribed to the Google Groups "Genomic SEM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genomic-sem-us...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/genomic-sem-users/b103c2e5-e6a4-4adc-a346-efb747f47033n%40googlegroups.com.
Message has been deleted

LMB

unread,
Jul 2, 2021, 4:25:51 PM7/2/21
to Genomic SEM Users
I used the latter for ordinal regression and there's output of effect sizes, standard errors, and p-values

LMB

unread,
Jul 10, 2021, 4:29:41 PM7/10/21
to Genomic SEM Users
For categorical/ordinal GWAS summary statistics with effect sizes/se/pvalues as outputs, do you think it's okay to go with the "continuous" path? Or is there probably no option for this in genomic sem? 

Michel Nivard

unread,
Jul 10, 2021, 5:07:03 PM7/10/21
to LMB, Genomic SEM Users
You can probably convert the bèta’s to a place where they can be used in GenomicSEM. 

Now if you are interest in standardized model (and don’t care about the h2 snp) then you can really just treat as continuous I think, it’s only when you need to pick an effective N that I don’t really know what to do as I know no formula for effective N from ordered logistic.

Best,

Michel

Op za 10 jul. 2021 om 22:29 schreef LMB <whatawo...@gmail.com>
Reply all
Reply to author
Forward
0 new messages