[BioC] edgeR question

[chunlai at umich.edu]

1. If I enter source("http://bioconductor.org/biocLite.R")
              biocLite("edgeR") in R package, 
   Does the R package automatically install the latest version, which I
believe is 1.6.9?
2. While using readDGE function, there are sub-function in the guide like
(targets, skip=5,    
   comment.char="#").
   What does it mean by skip=5 and comment.char="#" ?
   If I don't type skip=5 or type something else like skip=10, the count
libraries sizes and counts
   reported differently
   So I wonder if this "skip" function makes any difference?
3. Again, using d<-readDGE(targets, skip=5, comment.char="#") function, the
result of $counts    
   showed incorrect read counts from the original txt. file
   Is this a bug?
   For example:
                                  3MO1      3MO2     3MO3      18MO1   
18MO2     18MO3
miR-192/CTGACCTATGAATTGACAGC    192720.6 23615.286 5085.223      0.0   
0.000     0.000

However, the 18MO1, 18MO2, and 18MO3 do have reads for
miR-192/CTGACCTATGAATTGACAGC in the original txt. file

4. Since I have tags have very few read counts, I filter them out by
d<-d[rowSums(d$counts) > 1000,]
   I wonder if edgeR compare tags which exist in one sample but are
filtered out in another sample?
   For example:
                                  3MO1      3MO2     3MO3      18MO1      
18MO2      18MO3
miR-192/CTGACCTATGAATTGACAGC      exist     exist    exist   filter out 
filter out  filter out
3MO1, 3MO2 and 3MO3 are in one group, and 18MO1, 18MO2 and 18MO3 are in
another group
I want to compare the reads expression difference in these two groups   
Does edgeR treat the 18MO1, 18MO2 or 18MO3 as 0 count and compare between
3MO and 18MO? or edgeR does not compare between them?

   Another example:
                                   3MO1     3MO2      3MO3      18MO1      
18MO2      18MO3
miR-192/CTGACCTATGAATTGACAGC      exist   filter out  exist   filter out 
filter out   exist
Does edgeR compare reads between 3MO and 18MO groups?