TABLE 4.

One-way ANOVA statistics for linear regression models^a for invasion efficiency and intracellular survival of S. Saintpaul strains

Model and factor	dfb	Sum sqc	Mean sqd	F valuee	Pr (>F)f
INV model
Biological replicate	3	1.27	0.42	21.17	<0.001
Epi-type	1	0.16	0.16	8.01	0.010
Lineage	2	0.99	0.50	24.75	<0.001
bglA/bepFg	1	0.92	0.92	45.84	<0.001
ICS-1 model
Biological replicate	2	0.26	0.13	3.42	0.060
Lineage (IC vs IA/IB)h	1	0.32	0.32	8.45	0.011
bglA/bepFg	1	0.15	0.15	4.05	0.063
ygcH/stfCg	1	0.11	0.11	2.87	0.111
ICS-2 model
Biological replicate	2	0.17	0.08	7.80	0.004
Lineage	2	0.12	0.06	5.46	0.016
ICS-3 model
Biological replicate	2	0.84	0.42	11.77	<0.001
sodCI	1	0.51	0.51	14.17	0.002

^aThe models were fitted (using reference coding [R default]) to the data for invasion efficiency (the INV model), as well as intracellular survival between 0 and 2 hpi (the ICS-1 model), 2 and 6 hpi (the ICS-2 model), and 6 and 24 hpi (the ICS-3 model).

^bdf, degree of freedom.

^cSum sq, sum of squares due to the factor.

^dMean sq, mean of the sum of squares due to the factor.

^eF value, F statistic.

^fPr (>F), P value of the F test.

^gGenomic signatures grouped together due to identical genotypic patterns across S. Saintpaul strains.

^hStrains representing lineages IA and IB were grouped together and compared with strains representing lineage IC, as feature selection suggested that only the dummy variable for lineage IC should be included in the final model.