The collapsed dataset was normally distributed (Kolmogorov-Smirnov test = 0.162, p = 0.100).
GSEAs for the five signaling pathways and the positive control tissue(s) were run using GSEA software http://www.broad.mit.edu/gsea/ at the probe level (Collapse dataset to gene symbols = false) with 1,000 gene set permutations.
Given this condition, it is possible to 'collapse' each split dataset across covariate patterns (for example, separately collapse the dataset for males and females) [ 30].
In addition the average growth velocity for each child over the first year, obtained through collapsing the dataset to child level was used.
After collapsing the dataset from adjusted probes to symbols by GSEA, the expression levels of 2,870 of 5,923 genes were changed in the WAT of FSP27-/ mice (48.5%).
β4 phenotype was compared to mock phenotype by first collapsing the dataset to gene symbols and then using a weighted, difference of classes metric for ranking genes.
In the present study we used GSEA version 2.0.12 (www Broad Institute at MIT), with the C2 curated canonical pathway gene set (i.e. c2.cp.v3.1.symbols.gmt download), collapsing of dataset to gene symbols, weighted enrichment statistic and real gene list sorting.
The A. oryzae genome contains more than one thousand extra genes as compared to A. nidulans and A. fumigatus, but when we attempted to use phylogenetic methods to determine the origin of these genes the dataset collapsed to only 35 informative loci (Table 1).
Additional bootstrap analyses (1000 replicates) using neighbor-joining (K2P model, pairwise deletion) and parsimony (dataset collapsed using FaBox [ 67], complete deletion) were conducted in MEGA5.
Firstly, pre-processed Affymetrix datasets were collapsed to the probeset with the maximum expression value for each gene using the Gene Set Enrichment Analysis software (details below).
