[R-sig-ME] lme4 analysis_2ANOVA repeated measure

Tanay Ghosh tg369 @end|ng |rom c@m@@c@uk
Thu Jun 13 23:16:27 CEST 2019


Dear Emmanuel,

Many thanks for this reply.

I now added a column "subject id" (attached).

Atg3 is the name of a gene whose expression level was measured.

I know that we can not detect any difference here, also aware of the 
outlier.  However, using this data set as an example, I want to learn 
using lme4 for RM ANOVA specially when some values are NA.

Please do not hesitate to ask if anything is unclear and I am looking 
forward to hearing from you.

Kind regards,

Tanay

On 2019-06-13 13:45, Emmanuel Curis wrote:
> Hi Tanay,
> 
> In your data file, there seems to be nothing like « subject id », but
> only time, sex and Atg3 (=Abt3 ?) measure so it's not possible to do
> any random-effect modeling.
> 
> Besides, you should give more explaination about what is the Atg3
> column, because the syntax will change if there are CTs or equivalent
> measures [does not look like, except the atypical 31.355 value],
> absolute quantities or relative quantities compared to a (set of)
> reference genes.
> 
> Last, comparing all times to see a time effect is not a very powerful
> method, because you quickly increase the number of comparisons, and at
> the extreme, the more time you have, the more precise is your time
> curve, the more difficult it is to detect anything...
> 
> Note however that looking « by eye » your data, you can't expect much
> since you have a small dataser, huge outliers (31.355, 5.081, 4.804)
> and a quite high intra-groupe variability (values ranging from 0.087
> to 2.237 for the 3 males at t=7, for instance).
> 
> est regards,
> 
> On Mon, Jun 10, 2019 at 03:31:53PM +0100, Tanay Ghosh wrote:
> « Dear All,
> «
> « I wonder if I can ask your advise to analyse this data (attached). 
> Please
> « note that some of the values are "NA" i.e. I do not have any 
> measurement
> « taken.
> «
> « Here I have four different time (days) points. I measured expression 
> of Abt3
> « gene at these time points in male and female.
> «
> « I want to see if gender wise expression of Atg3 differ at any time 
> point?
> «
> « I further want to do a posttest to get p value for Male at time 7 vs 
> female
> « at time 7, Male at time 15 vs Female at time 15, and like wise all 
> other
> « time points.
> «
> «
> « Please could you advise me how to do this analysis using lmer 
> function? I
> « would very much appreciate if you please write the commands in R for 
> this
> « specific case.
> «
> « Thank you in advance for your time.
> «
> «
> « Kind regards,
> «
> « Tanay
> «
> «
> « --
> « Dr. Tanay Ghosh, PhD
> « Research Associate
> « Department of Clinical Neurosciences,
> « Wellcome Trust-Medical Research Council Cambridge Stem Cell 
> Institute,
> « Clifford Allbutt Building,
> « Cambridge Biomedical Campus,
> « University of Cambridge,
> « Cambridge CB2 0AH
> 
> «
> time	Gender	Atg37	Male	0.2127	Male	2.2377	Male	0.0877	Female	0.0927	Female	2.8057	Female	0.10315	Male	0.07115	Male	2.73315	Male	2.41415	Female	0.1515	Female	2.43815	Female	1.43424	Male	0.07924	Male	2.21224	Male	31.35524	Female	0.21124	Female	0.81224	Female	5.081231	Male	0.073231	Male	0.505231	Male	0.737231	Female	0.178231	Female	0.319231	Female	4.804420	Male	0.29420	Male	0.424420	Male	NA420	Female	0.337420	Female	0.966420	Female	NA
> 
> « _______________________________________________
> « R-sig-mixed-models using r-project.org mailing list
> « https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

-- 
Dr. Tanay Ghosh, PhD
Research Associate
Department of Clinical Neurosciences,
Wellcome Trust-Medical Research Council Cambridge Stem Cell Institute,
Clifford Allbutt Building,
Cambridge Biomedical Campus,
University of Cambridge,
Cambridge CB2 0AH
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Ginez_data_Atg3.txt
URL: <https://stat.ethz.ch/pipermail/r-sig-mixed-models/attachments/20190613/2c158520/attachment.txt>


More information about the R-sig-mixed-models mailing list