[BioC] "romer"ing and "roast"ing around gene sets

Robert M. Flight rflight79 at gmail.com
Wed Jul 14 15:57:48 CEST 2010


Hi Di,

Thank you for the clarification. I think that helps me to better know
which one of the two to use.

Cheers,

-Robert

On Tue, Jul 13, 2010 at 23:40, Di Wu <di.wu at med.monash.edu.au> wrote:
> Hi Robert,
>
> Thank you for asking. I am sorry if the two methods caused confusing.
>
> You are right about the understanding of the "roast" paper and "romer" in
> your first paragraph.
>
> The null hypothesis in "roast" is for the focused gene sets as demonstrated
> in the real data example of the "roast" paper. "mroast" function in limma
> can test multiple gene sets at the same time. Essentially "mroast" is doing
> the same as you run a series of "roast" followed by multiple testing
> adjustments for p values. "mroast" works in a more computing efficiently way
> than a series of "roast".
>
> The null hypothesis in roast is equivalent as the self-contained hypothesis
> discussed by Goeman and Bühlmann (2007)
> http://www.ncbi.nlm.nih.gov/pubmed/17303618
> Bioinformatics. 2007 Apr 15;23(8):980-7
>
> From my point of view, GSEA tests the combbined hypothesis, as it uses rank
> (a test score for competitive test, relative to other genes)  and sample
> labeling permutations (which generates the null distribution under the
> self-contained hypothesis, "significantly associated with a phenotypic
> comparison").
>
> Roast is can handle both small sample size data as well as larger samples
> size data. Sample size is a concern for self-contained hypothesis (focus
> set).
>
> In your case, I think romer is more suitable. Let me know if you have more
> questions regarding these two functions in limma.
>
> I am still using my email address in Monash in the mailing list though I am
> in WEHI now.
>
> Cheers,
> Di
>
> On Wed, Jul 14, 2010 at 12:58 AM, Robert M. Flight <rflight79 at gmail.com>
> wrote:
>>
>> Hi All,
>>
>> I am having trouble with the distinction between the functions "roast"
>> and "romer" in the limma package. From the publication describing
>> "roast" (http://dx.doi.org/10.1093/bioinformatics/btq401), it seems
>> that it tests a particular gene set for differential expression,
>> whereas "romer" tests a battery of sets to find those that are
>> differentially expressed compared to the rest?
>>
>> I am really having trouble discerning the true difference between
>> these two, and how they compare to GSEA. I always thoght that the
>> primary purpose of GSEA was to determine those gene sets that are
>> significantly associated with a phenotypic comparison, i.e. those gene
>> sets showing differential expression.
>>
>> If any one can help me clear this up, that would be great, because as
>> of now I am thoroughly confused. To me, if I have a dataset, and I
>> want to know which gene sets (from say MSigDB) are differentially
>> expressed, then it sounds like I would use "roast", but the way it is
>> described in the publication (and the help in limma), this isn't what
>> I would do, but rather I should use "romer", and see if any of the
>> sets show differential expression compared to the rest in the
>> database.
>>
>> Color me confused,
>>
>> -Robert
>>
>> Robert M. Flight, Ph.D.
>> Bioinformatics and Biomedical Computing Laboratory
>> University of Louisville
>> Louisville, KY
>>
>> PH 502-852-0467
>> EM robert.flight at louisville.edu
>> EM rflight79 at gmail.com
>>
>> Williams and Holland's Law:
>>        If enough data is collected, anything may be proven by
>> statistical methods.
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>



More information about the Bioconductor mailing list