comparison between two condition (label_compare)

[DoYup]

Dear Jordi

Thanks for your help the other day.

I have question on Welch's ttest between different condition. Because my own data is not working I have tried it with example data "mtbl231". But I was not able to have p value and fold change between two condition.

Here is the script I am using. I have tried it with different combination of control.con = (e.g. "05mM_Nor)

s4 <- label_compare(geoRgeR = s2, XCMSet = xset3, ppm.s = NULL,rt.win.min = 1, control.cond = NULL, fc.vs.Control = 1, p.value.vs.Control = 0.05, Show.bp = T)

str(s4)
'data.frame': 1105 obs. of 5 variables:
$ Comparison : Factor w/ 2 levels "","Base peak": 2 1 2 1 1 2 1 2 1 1 ...
$ X05mM_Normo_MEAN: num 61.89 38.11 81.43 7.36 11.21 ...
$ X25mM_Normo_MEAN: num 67.12 32.88 82.01 7.92 10.07 ...
$ X05mM_Normo_SD : num 7.17 7.17 1.6 0.37 1.92 ...
$ X25mM_Normo_SD : num 0.87 0.87 0.723 0.437 1.142 ...

Here is the file name that I downloaded for mtbl231
CELL_Glc12_05mM_Normo_04
CELL_Glc12_05mM_Normo_05
CELL_Glc12_05mM_Normo_06
CELL_Glc12_25mM_Normo_16
CELL_Glc12_25mM_Normo_17
CELL_Glc12_25mM_Normo_18
CELL_Glc13_05mM_Normo_01
CELL_Glc13_05mM_Normo_02
CELL_Glc13_05mM_Normo_03
CELL_Glc13_25mM_Normo_13
CELL_Glc13_25mM_Normo_14
CELL_Glc13_25mM_Normo_15

Thanks in advance.

[jcapelladesto]

Dear DoYup,

Thank you for contacting again.
First, I am aware that label_compare does not return p-values and foldchanges as PuInc_seeker does. I will try to add this feature to the function ASAP (I actually cannot understand why we did not do that in the first place).
The main feature of label_compare is the first column ('Comparison') which will include "UP_" or "DOWN_" values when significant results are found. Since I see that when you run

s4 <- label_compare(geoRgeR = s2, XCMSet = xset3, ppm.s = NULL,rt.win.min = 1, control.cond = NULL, fc.vs.Control = 1, p.value.vs.Control = 0.05, Show.bp = T)

you have 'ppm.s' and 'control.cond' set to NULL. This might be the cause that this result comes up:
$ Comparison : Factor w/ 2 levels "","Base peak": 2 1 2 1 1 2 1 2 1 1 ...
Please try to run again this function, like this (or similar):

s4 <- label_compare(geoRgeR = s2, XCMSet =mtbls213, rt.win.min = 1, 
										control.cond = "05mM_Normo",ppm.s=6.5,
										fc.vs.Control = 1, p.value.vs.Control = 0.05, Show.bp = T)

This should return (see number of levels of $ Comparison):

str(s4)
'data.frame':	1105 obs. of  5 variables:
 $ Comparison      : Factor w/ 4 levels "","Base peak",..: 2 1 2 1 1 2 1 2 1 1 ...
 $ X05mM_Normo_MEAN: num  61.89 38.11 81.43 7.36 11.21 ...
 $ X25mM_Normo_MEAN: num  67.12 32.88 82.01 7.92 10.07 ...
 $ X05mM_Normo_SD  : num  7.17 7.17 1.6 0.37 1.92 ...
 $ X25mM_Normo_SD  : num  0.87 0.87 0.723 0.437 1.142 ...

Last but not least, you indicated that this function does not work for your data. I would like to ask you if at least the % values are calculated.

Yours sincerely,

Jordi

[DoYup]

Great. It works.

For my data, % are calculated. I wish to see p value and fold change soon. :)
Now I will try to do it with my data. (I may return to you at that moment with question)
Thank you again

[jcapelladesto]

The package was updated a few weeks ago, the p-value and fold-change should be reportable now. I wonder if you have been able to run it for your data.

[DoYup]

Hi Jordi It sounds great! And it works very well with my real data thanks to your help !!! Sincerely, Do Yup

…

On Mon, Aug 28, 2017 at 5:59 PM, Jordi Capellades Tomas < ***@***.***> wrote: The package was updated a few weeks ago, the p-value and fold-change should be reportable now. I wonder if you have been able to run it for your data. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#8 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AbShbEs652ksKQ16OIzdgNdbWmDt1v4Jks5scoFrgaJpZM4OpxOW> .