3 featureMSEA Analysis

Step 2 performs feature rank-based metabolite set enrichment scoring. Features are ranked according to a user-provided phenotype-associated statistic, and for each metabolite set, a running enrichment score is computed across the ranked feature list to evaluate whether features annotated to metabolites in that set are non-randomly concentrated toward the top of the ranking.

Figure 3.1: Step 2: featureMSEA enrichment parameters

3.1 Parameters

Parameter	Description	Default
Threads	Parallel processing threads	3
Min Compounds	Minimum pathway size (compounds)	15
Max Compounds	Maximum pathway size (compounds)	300
Permutations	Statistical permutations for p-value estimation	1000
Max Iterations	Algorithm iterations	1
FDR Thr	FDR significance threshold	0.05

Note: Increasing Permutations improves p-value precision but increases runtime proportionally. 1000 permutations gives a good balance for exploratory analysis; use 10,000+ for publication-quality results.

Click Run Step 2. Run time scales with the number of permutations and the size of the pathway database. A progress bar is shown during computation.