which qvalue to use in pathway analysis using gage, same.dir?

Question

[Daisy]

Hello Guys

If we use same.dir = True in Gage, we got the following stats:

 greater.p.geomean  greater.stat.mean   greater.p.val   greater.q.val   greater.set.size    greater.exp1    
 less.p.geomean less.stat.mean  less.p.val  less.q.val  less.set.size  less.exp1    stats.stat.mean stats.exp1

If I want to pick the significant pathways, which qvalue I should use? For example, If it is significant in the less.q.val but not significant in the greater.q.val what does this mean? Is there any global pvalue to use?

Thanks

Answer 1

[h.mon]

The same.dir = T in gage performs two tests, one for enrichment of up-regulated genes (the greater table), one for enrichment of down-regulated genes (the less table). So, you should use both, but understand their difference.

[Daisy]

Thanks. I understand it dies 2 tests. What if the greater is insignificant, but the less is significant, does this mean, it is down-regulated, and ignore the greater? They are both output in one table not two tables.

[h.mon]

No, it is one object, but two tables of results, represented by the $greater and $less slots.

?gage

The result returned by gage function is a named list, with either 3 elements ("greater", "less", "stats") for one-directional test (same.dir = TRUE) or 2 elements ("greater", "stats") for two-directional test (same.dir = FALSE). Elements "greater" and "less" are two data matrices of the same structure, mainly the p-values, element "stats" contains the test statistics. Each data matrix here has gene sets as rows sorted by global p- or q-values.

[Daisy]

What I meant is that I have just one table that contains both, but that's not the main point concern, my concern is if the same pathway is significant in the less and not significant in the greater, it means it is a downregulated pathway that i should take it and discard the fact it is not significant in the greater? Thanks

[h.mon]

discard the fact it is not significant in the greater?

I am not understanding if you are talking about interpreting the results, or presenting the results.

If you are talking about presenting the results, then yes, you may (in fact, you should) present only the results deemed significant at the chosen level. So if a pathway is significant for the $less but not for the $greater, include it on the less and exclude from the greater.

If you are talking about interpretation, then I do not even understand the question.

[Daisy]

You exactly answered my questions, thanks so much h.mon :)