Genome-wide analysis of fitness data and its application to improve metabolic models

BMC Bioinformatics

Table 1 Correlation and mmHG results for comparing co-fitness to co-expression. Empirical p-values for the mmHG tests were calculated based on shuffled data, where each list, used as functional class for the gene pairs, was shuffled (100 instances), preserving the original partition structure

	Spearman’s R	Corrected mmHG statistics	# of pairs (N)	B	n*	b*	Empirical p-value from shuffled data
All pairs	0.007	0.1	6,485,401	43	368	2	N/A
Same operons	0.186	1.18 × 10⁻¹³	2453	251	966	165	0
Paralog genes	0.048	1	4584	5	46	2	0.32
Same Pfam domain	0.051	1	10,626	948	155	31	0.47
Within 5 kb	0.058	2.53 × 10⁻⁴	7443	221	991	61	0
Within 10 kb	0.046	2.49 × 10⁻⁴	14,202	57	481	14	0

ISSN: 1471-2105