sample assignment incorrect in glMDPlot 1.8.1 vs. 1.6.0
Closed this issue · 2 comments
jdrnevich commented
I encountered a bug in Glimma 1.8.1 that was not present in 1.6.0. In the top right plot, the group argument is pulling the correct expression values and grouping them correctly, but the sample.cols and the pop-up sample information is not correct. I used the same code to make the attached screenshots:
glMDPlot(fit.bad.tr, coef = 2, counts=logCPM.filt, transform = FALSE, sample.cols = as.numeric(bad.samp),
groups=bad.samp, ylab = "log2(bad / good)", xlab = "Average expression level",
status=decideTests(fit.bad.tr), folder = "glimma-plots-R3.5.0", side.ylab = "Individual expression level",
main="R 3.5.0", html = "Bad_vs_good_speciesD", samples = targets$Sample)
data.frame(sample = colnames(logCPM.filt), TRINITY_DN28488_c1_g3 = logCPM.filt["TRINITY_DN28488_c1_g3",],
bad.samp)
sample TRINITY_DN28488_c1_g3 bad.samp
gfw_a gfw_a 3.524879 FALSE
gfw_b gfw_b 2.810853 FALSE
gfw_c gfw_c 3.287497 FALSE
gb_a gb_a 3.190447 FALSE
gb_b gb_b 2.545063 FALSE
gb_c gb_c 3.056966 FALSE
gm_a gm_a 3.588505 FALSE
gm_b gm_b 3.437603 FALSE
gm_c gm_c 3.179978 FALSE
wfw_a wfw_a 2.509955 FALSE
wfw_b wfw_b 1.950160 FALSE
wfw_c wfw_c 2.280330 FALSE
wb_a wb_a 2.604763 FALSE
wb_b wb_b 7.901653 TRUE
wb_c wb_c 9.101571 TRUE
wm_a wm_a 2.843367 FALSE
wm_b wm_b 2.274999 FALSE
wm_c wm_c 9.278173 TRUE
Shians commented
😱😱😱
Thanks as always for letting me know about these serious bugs. This should be fixed here with the extra reordering calls.
I've pushed to Bioconductor Devel as 1.9.2 and Release as 1.8.2.
jdrnevich commented
No problem - I love the package!!!