Figure 1
From: Significance analysis of lexical bias in microarray data
Binomial distribution of SPI-2 genes in a dataset The total filtered dataset consisted on 4290 unique elements. An SGL of 256 genes was generated using SAM and analyzed for 34 members of SPI-2. The arrow indicates the number of matches in the SGL, with P(x > 8) = 0.004. The binomial analysis required 5 seconds; Poisson analysis of the same datasets required 7 seconds. A 21,450 element dataset created by replicating the 4290 element dataset 5 times required 8 seconds for binomial analysis. The files used for this analysis are available at the LACK website or as supplementary data.