Comparing Order Of Genes In Two Lists Using Orderedlist

Question

[Nasir]

Hi

I would be very grateful for any suggestions on how to solve the following problem. I want to compare the order of genes in two ranked gene lists to determine if there is more similarity than would be expected by chance alone. Bioconductor package OrderedList should be able to do this, however I am getting an error. I have run the program using the example lists in the vignette supplied by the authors, and it works fine. I have also tried using two random lists of 1000 numbers each and, again, it works fine. However, when I try with my pair of gene lists (containing 41 genes each), I get the following error:

.>List1
[1] "1" "2" "3" "4" "5" "6" "7" "8" "9" "10" "11" "12" "13" "14" "15"
[16] "16" "17" "18" "19" "20" "21" "22" "23" "24" "25" "26" "27" "28" "29" "30"
[31] "31" "32" "33" "34" "35" "36" "37" "38" "39" "40" "41"
.>List2
[1] "7" "4" "1" "12" "20" "5" "14" "11" "9" "8" "23" "27" "13" "3" "35"
[16] "31" "6" "19" "24" "41" "18" "26" "2" "17" "25" "21" "36" "40" "34" "15"
[31] "22" "10" "38" "39" "32" "16" "30" "33" "28" "37" "29"
.>compareLists(List1,List2)
Error in 1:n : result would be too long a vector
In addition: Warning message: In max(nn) : no non-missing arguments to max; returning -Inf

I modified defaults as below, but got the same result

.>compareLists(List1, List2, mapping = NULL, two.sided=FALSE, B = 1000, alphas = NULL, invar.q = 0, min.weight = 1e-5, no.reverse=TRUE)

Perhaps my gene lists are too small? If so, is there any other way of comparing order of items in two lists containing the same items?

Thank you!

Nasir

Answer 1

[Martin Morgan]

For a reproducible example

library(OrderedList)
example(compareLists)
x = list1[1:40]; y = sample(x)

and then

> compareLists(x, y)
Error in 1:n : result would be too long a vector
In addition: Warning message:
In max(nn) : no non-missing arguments to max; returning -Inf

From looking at the code, try setting 'alphas' to something reasonable, e.g.,

>     compareLists(x, y, alphas=ceiling(seq(10, length(x), length.out=5)))
  Simulating random scores...
  0%.......:.........:.........:.........:......100%
  --------------------------------------------------
List comparison
  Assessing similarity of     : top and bottom ranks
  Length of lists             : 40
  Quantile of invariant genes : 0.5
  Number of random samples    : 1000
--------------------------------------
   Genes Scores p.values Rev.Scores Rev.p.values
10     1      0    0.099          0        0.099
18     0      0    0.000          0        0.000
25     0      0    0.000          0        0.000
33     0      0    0.000          0        0.000
40     0      0    0.000          0        0.000

Consider contacting the maintainer packageDescription("OrderedList") so that they are aware of this problem and can address it in a subsequent release.

[Nasir]

Many thanks Martin! I have now been able to run this analysis by manipulating alpha. Looking at the plots etc, I think I have got the analysis I wanted. I have re-read the manual a few times, but still can't quite understand what alpha is! Could anyone please help explain this to someone with no formal statistical training? Nasir