yunzhang813
commented
5 years ago Hi Pef,
In general, we recommend to conduct the analysis in individual replicate and then combine the p-values using some meta data analysis techniques, such as Fisher’s p-value combination (as we did in the paper).
Answering your specific question about the data input, X is the data matrix, tt is the UNIQUE time points, you are right about rr. And you would have to group repeats for a given time point in adjacent columns.
Thanks,
Yun (Renee) Zhang, PhD Staff Scientist - Biostatistician Informatics Department J. Craig Venter Institute 4120 Capricorn Lane, La Jolla, CA 92037 Email: zhangy@jcvi.orgmailto:zhangy@jcvi.org Phone: (858) 750-4003
From: PFRoux notifications@github.com Reply-To: yunzhang813/FUNNEL-GSEA-R-Package reply@reply.github.com Date: Tuesday, November 13, 2018 at 9:37 AM To: yunzhang813/FUNNEL-GSEA-R-Package FUNNEL-GSEA-R-Package@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [yunzhang813/FUNNEL-GSEA-R-Package] Failing to include replicates (#4)
Hi !
I managed so far to reproduce nicely the example you provide in the vignette.
When shifting to my data and analyzing one single timecourse (no replicate) it worked like a charm. Nevertheless I struggling to include replicate.
What is the expected structure of the X matrix, the tt vector, and the rr vector ? I have two repeats of a 6 time points experiments.
Here is what I am doing now :
head(x) RAS_1_0 RAS_1_24 RAS_1_48 RAS_1_72 RAS_1_96 RAS_1_144 RAS_2_0 RAS_2_24 RAS_2_48 RAS_2_72 RAS_2_96 RAS_2_144 DDX11L1 6.173987 6.292460 6.161247 6.371116 6.282635 6.698949 6.068301 6.129756 6.135309 6.270198 6.236951 6.464467 LINC01001 8.006745 7.991134 7.965366 7.859454 8.034330 8.668004 8.040542 7.766224 7.867446 7.729937 7.988560 8.263325 LINC01061 11.145774 11.284128 11.355694 11.368836 11.654988 12.466725 11.277207 11.181921 11.315475 11.170272 11.607917 12.105991 LOC101928626 4.798979 4.473942 4.636451 4.604835 4.502793 4.377194 4.783271 4.699812 4.664304 4.354646 4.477432 4.323595 LINC01128 7.935098 8.281702 8.308267 8.341863 8.518952 8.825222 7.924112 8.284762 8.328809 8.390836 8.511379 8.723604 LOC284600 7.449482 7.003634 7.334723 7.458990 7.349386 7.203119 7.502387 7.236217 7.376226 7.420036 7.361937 7.377836
tt [1] 0 24 48 72 96 144 0 24 48 72 96 144
funnel_test <- FUNNEL.GSEA(x, tt, genesets, rr = rep(2,6)) Error in rep(1:mm, rr) : invalid 'times' argument
I also tried sorting the X matrix and the tt vector by increasing time point (grouping repeats for a given time point in adjacent columns) but I am getting the same issue.
Thanks a lot for this really nice tool !
Cheers,
Pef
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/yunzhang813/FUNNEL-GSEA-R-Package/issues/4, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ANi8MtOjTnMaoaafKDhQIYe9daejS_PPks5uuwM9gaJpZM4YcKKv.
Hi !
I managed so far to reproduce nicely the example you provide in the vignette.
When shifting to my data and analyzing one single timecourse (no replicate) it worked like a charm. Nevertheless I struggling to include replicate.
What is the expected structure of the X matrix, the tt vector, and the rr vector ? I have two repeats (1 and 2) of a 6 time points (T0, 24, 48, 72, 96 and 144) experiment.
Here is what I am doing now :
I also tried sorting the X matrix and the tt vector by increasing time point (grouping repeats for a given time point in adjacent columns) but I am getting the same issue.
Thanks a lot for this really nice tool !
Cheers,
Pef