[R] qvalue package: checking I'm using the right output values.

[aoife doherty]

Can I ask, when I run the below code:


library(qvalue)

pvals
<-c(0.01,0.02,0.03,0.10,0.67,0.08,0.55,0.10,0.81,0.08,0.94,0.10,0.81,0.08,0.72,0.10,0.22,0.08,0.72,0.01,0.02,0.03,0.04,0.05,0.06,0.07,0.01,0.02,0.03,0.10,0.67,0.08,0.55,0.10,0.81,0.08,0.94,0.10,0.81,0.089,0.72,0.10,0.22,0.08,0.72,0.01,0.02,0.03,0.04,0.05,0.06,0.07)

qobj <-qvalue(p=pvals)

pvals

qobj$qvalues

qobj$lfdr


the qobj$qvalues list is:

 [1] 0.03592311 0.03592311 0.03592311 0.03991457 0.22490992 0.03991457

 [7] 0.19757712 0.03991457 0.23278177 0.03991457 0.25975174 0.03991457

[13] 0.23278177 0.03991457 0.22490992 0.03991457 0.08319037 0.03991457

[19] 0.22490992 0.03592311 0.03592311 0.03592311 0.03991457 0.03991457

[25] 0.03991457 0.03991457 0.03592311 0.03592311 0.03592311 0.03991457

[31] 0.22490992 0.03991457 0.19757712 0.03991457 0.23278177 0.03991457

[37] 0.25975174 0.03991457 0.23278177 0.03991457 0.22490992 0.03991457

[43] 0.08319037 0.03991457 0.22490992 0.03592311 0.03592311 0.03592311

[49] 0.03991457 0.03991457 0.03991457 0.03991457


and the qobj$lfdr list is:

 [1] 0.03070397 0.04312982 0.05458114 0.13925884 0.79081046 0.11311424

 [7] 0.79081046 0.13925884 0.79081046 0.11311424 0.79081046 0.13925884

[13] 0.79081046 0.11311424 0.79081046 0.13925884 0.34265237 0.11311424

[19] 0.79081046 0.03070397 0.04312982 0.05458114 0.06583569 0.07718949

[25] 0.08879583 0.10074920 0.03070397 0.04312982 0.05458114 0.13925884

[31] 0.79081046 0.11311424 0.79081046 0.13925884 0.79081046 0.11311424

[37] 0.79081046 0.13925884 0.79081046 0.12463425 0.79081046 0.13925884

[43] 0.34265237 0.11311424 0.79081046 0.03070397 0.04312982 0.05458114

[49] 0.06583569 0.07718949 0.08879583 0.10074920


Can I check that the qobj$lfdr are the FDR Q values (i.e. the number of
expected false positives over the number of significant results) for each
gene?


Thanks

	[[alternative HTML version deleted]]