[BioC] Gene Set Definition GSCA package

[Seb]

hi all

i'm trying to construct a gene definition list like the one in the
GSCA package (availabler here
http://www.biostat.wisc.edu/~kendzior/GSCA/), the first 10 elements of
it will look like this:


> head(str(LungCancer3$info$GSdef[1:10]))
List of 10
 $ GO:0008150: Named chr [1:3631] "5595" "8850" "4137" "644" ...
  ..- attr(*, "names")= chr [1:3631] "Row1" "Row10" "Row100" "Row1000" ...
 $ GO:0008152: Named chr [1:2348] "5595" "8850" "4137" "644" ...
  ..- attr(*, "names")= chr [1:2348] "Row1" "Row10" "Row100" "Row1000" ...
 $ GO:0002526: Named chr [1:46] "2683" "4153" "717" "12" ...
  ..- attr(*, "names")= chr [1:46] "Row1052" "Row1097" "Row1193" "Row1291" ...
 $ GO:0006629: Named chr [1:248] "11119" "2629" "23049" "11120" ...
  ..- attr(*, "names")= chr [1:248] "Row1004" "Row1005" "Row1007" "Row1018" ...
 $ GO:0006950: Named chr [1:437] "23049" "6387" "1965" "7057" ...
  ..- attr(*, "names")= chr [1:437] "Row1007" "Row1014" "Row102" "Row105" ...
 $ GO:0006952: Named chr [1:204] "7716" "6387" "2683" "3276" ...
  ..- attr(*, "names")= chr [1:204] "Row1003" "Row1014" "Row1052" "Row1070" ...
 $ GO:0006953: Named chr [1:16] "12" "197" "5265" "1401" ...
  ..- attr(*, "names")= chr [1:16] "Row1291" "Row2036" "Row2107" "Row2231" ...
 $ GO:0006954: Named chr [1:134] "6387" "2683" "4153" "717" ...
  ..- attr(*, "names")= chr [1:134] "Row1014" "Row1052" "Row1097" "Row1193" ...
 $ GO:0009605: Named chr [1:277] "6387" "1794" "5880" "7057" ...
  ..- attr(*, "names")= chr [1:277] "Row1014" "Row1033" "Row1041" "Row105" ...
 $ GO:0009611: Named chr [1:205] "6387" "7057" "2683" "5359" ...
  ..- attr(*, "names")= chr [1:205] "Row1014" "Row105" "Row1052" "Row1053" ...



however, i don't really know how to create it for my list of genes, i
was able to obtain my list of genes paired with a number of GO IDs but
it seems i can't do the opposite.
any idea? or do you know any other package that will help doing a
similar gene co-expression analysis and with the chance to visualize a
network?

thanks

Seb