R help - Jul 2013 - Multiple Comparison Kruskal-Wallis Test on a dataframe

Hi
I have a dataframe of genomic positions something like below:

chr    start    end    strand    1a    2a    3a    1b    2b    3b    1c
2c    3c    1d    2d    3d
chr1    201199522    201199523    -    51    62    63    46    53    58
53    78    37    27    65    62
chr1    201199584    201199585    -    42    49    35    54    62    55
43    56    26    21    64    63
chr1    201199603    201199604    -    45    46    55    49    41    57
45    52    26    21    62    69
chr1    201199608    201199609    -    40    45    47    63    54    50
40    45    25    22    64    63
chr1    37944086    37944087    -    100    92    96    78    74    76
76    61    65    84    100    100
chr1    37944119    37944120    -    89    95    93    81    67    76
74    51    63    100    65    100

I would like to use a kruskal test with multi-test correction, probably
from the agricolae package, on each genomic coordinate (i.e. row) to
calculate whether treatment groups a,b,c & d are different (where 1,2 &
3
would be replicates within the groups a,b,c, & d).
What is the best way to achieve this? I would have thought the options are:
i) use transform in plyr
ii) assign group numbers by genomic coordinate and treatment group and
perform the kruskal test on each genomic coordinate group
Does anyone have any suggestions to achieve this in the quickest way?
Thanks in advance

	[[alternative HTML version deleted]]