[BioC] how to get samoutput$mat.fdr and samoutput$delta from SAM object ? - help

Saurin Jani saurin_jani at yahoo.com
Thu Dec 22 13:53:51 CET 2005


Dear BioC,

I am using siggenes package. If anyone have used,
please let me know, how can I get access of all
samoutput$fdr and samoutput$delta values from
samoutput object?

This might be very easy..but and even I have also
opened Vignette but there is not information in
Vignette as well.(sam.out$mat.fdr  gives NULL)

I did following:
> samoutput <- sam(MatrixRMA,cl, rand = 123);
We're doing 90 complete permutations

>samoutput$mat.fdr
NULL

> samoutput
SAM Analysis for the Multi-Class Case with 3 Classes

    Delta    p0   False Called   FDR
1     0.1 0.854 408.578    426 0.819
2   259.8 0.854   4.400     14 0.268
3   519.5 0.854   1.733      6 0.247
4   779.2 0.854   0.722      3 0.205
5  1038.9 0.854   0.333      2 0.142
6  1298.6 0.854   0.333      2 0.142
7  1558.3 0.854   0.333      2 0.142
8  1818.0 0.854   0.111      1 0.095
9  2077.7 0.854   0.111      1 0.095
10 2337.5 0.854   0.000      0 0.000

>summary(samoutput)
SAM Analysis for the Multi-Class Case with 3 Classes

 s0 = 0.0048  (The 5 % quantile of the s values.)

 Number of permutations: 90 (complete permutation)

 MEAN number of falsely called genes is computed.

    Delta    p0   False Called   FDR cutlow    cutup
j2    j1
1     0.1 0.854 408.578    426 0.819   -Inf   16.599 
0 12063
2   259.8 0.854   4.400     14 0.268   -Inf  483.139 
0 12475
3   519.5 0.854   1.733      6 0.247   -Inf  786.998 
0 12483
4   779.2 0.854   0.722      3 0.205   -Inf 1185.619 
0 12486
5  1038.9 0.854   0.333      2 0.142   -Inf 2180.387 
0 12487
6  1298.6 0.854   0.333      2 0.142   -Inf 2180.387 
0 12487
7  1558.3 0.854   0.333      2 0.142   -Inf 2180.387 
0 12487
8  1818.0 0.854   0.111      1 0.095   -Inf 3254.402 
0 12488
9  2077.7 0.854   0.111      1 0.095   -Inf 3254.402 
0 12488
10 2337.5 0.854   0.000      0 0.000   -Inf      Inf 
0 12489


Thank you so much in advance,
Saurin



More information about the Bioconductor mailing list