Eukaryotic transcriptomics in silico: Optimizing cDNA-AFLP efficiency

BMC Genomics

Table 2 The relative contribution of enzyme combinations to cDNA pool coverage.

Source	Num df	Den df	F	Sig.	Partial R²
Model	58	2248.56	134.76	<0.001	77.66
Total pool size (bp)	1	85.40	14.25	<0.001	14.30
Average sequence length	1	86.29	52.87	<0.001	37.99
GC content	1	86.60	34.89	<0.001	28.72
Non-ACGT content	1	84.49	0.05	0.823	<0.01
Enzyme combination	27	2428.48	199.26	<0.001	68.90
Enzyme combination*GC content	27	2428.48	112.38	<0.001	55.55

Variance partitioning addressing the contribution of enzyme combination (28 combinations) on pool coverage for 92 eukaryotic species (see additional file 1). Species was included as a random factor in a mixed model analysis which aimed to determine the influence of individual factors or interactions (Source). cDNA pool coverage was weighted by the number of sequences per species to account for variation in available sequence data. The numerator and denominator (Kenward-Roger corrected) degrees of freedom (Num df/Den df) are provided. F statistics (F) and the significance (Sig.) of the overall model, factors and interactions are reported. The proportion of the variation in cDNA pool coverage which is explained by each factor/interaction is indicated as Partial R-square values [27].

ISSN: 1471-2164