SGoF+ program allows to obtain the multiple test adjustment probability for Sequential Bonferroni (Holm 1979),
Benjamini-Hochberg (1995) FDR, SGoF (Carvajal-Rodriguez et al 2009), Fisher combined probability test (SFisher, Perez-Diz et al 2011)
and SGoF+ (Carvajal-Rodriguez and de Uña, in preparation) methods. It also estimates the proportion of true null hypothesis
and computes the corresponding q-values (Tibshirani & Storey 2003; Storey et al 2004).
The SGoF adjustment can be computed without a list of p-values via interactive input of the number of comparisons and significant cases detected.
The remaining methods (including SGoF+) need a list of probability values in a text file (the list does not need to be sorted).
 
Input File
 
The input file is called by default
PvalSGoF.dat but any other name can be given from the corresponding option in the menu.
The input file should have an integer number indicating the total number of tests and after it two columns with pairs of
identifiers and p-values. The identifier can be a number or a character string. The list of p-values does not need to be sorted.
 
See an
SGoF+ input file example:
 
7
1 0.003
id2 5e-3
gen3 1.0
id4 0.34
5 0.01
6 0.004
7 0.445
There is no limit on the number of p-values other than computer memory.
Using interactive input, the program also allows to compute the expected number of true positives in a sample of tests
when a given number of significant p-values was found.
 
Run the program
 
a)
Windows: double click the SGoF+.exe file.
b)
Unix/Linux: Just type ./SGoF+. The linux executable was compiled in Ubuntu 10.04 (Lucid Lynx).
Source will be provided upon request to be compiled in any other system.
c)
Mac: Just type ./SGoF+ in the command line (in a Terminal window) from the appropriate directory or double click if executable is linked to be opened with Terminal.app.
Note that by using the double click way, the input file must be in the home directory. The Mac executable was compiled under OsX 10.5.6.
 
Output File
 
The program provides with two kinds of output files:
 
1)
SGoF+_cutoff.txt which includes only the tests remaining significant after the different adjustments.
2)
SGoF+_Fulloutput.html and
SGoF+_Fulloutput.xls which provide to the user with the full list of a priori significant
p-values and the adjusted p and estimated q-values. The .html one returns just the output coming from the last program execution
while the .xls accumulate the output through different executions.
 
The adjusted p-values for B-H and SB are computed following Yekutieli & Benjamini 1999. Note that the SGoF, SFisher and SGoF+ values are
not adjusted individual p-values. They are just p-values from the metatest.
 
See an
SGoF+_Fulloutput file example: