Did glacial advances during the Pleistocene influence differently the demographic histories of benthic and pelagic Antarctic shelf fishes? – Inferences from intraspecific mitochondrial and nuclear DNA sequence diversity

Table 3 Results of demographic analyses performed with various methods.

	MD	W-H	Taj. D	Fu's F	FLUCTUATE	LAMARC	SWEEP-BOTT
Cyt b
T. bernacchi	θi = 0 (0–0.7), θ_f= 550 (0.97–3890)	θ_i= 0.21, θ_f= 2.5	-1.8 *	-7.7 **	3969 (s.d. 199)‡	1008 (36–1974)	-11.9 vs.0 **
T. bernacchi (Ross Sea)	θ_i = 0 (0–0.45), θ_f = 0.356 (0–2365)	θ_i = 0.0001, θ_f = 19.	-2.18 **	-8.11**	10000 (s.d. 1670) ‡	4454 (3575–5625)	-10.57 vs. 0 **
T. penelli	θ_i = 0.1 (0–0.34), θ_f = 0.141 (0–776)	θ_i = 0, θ_f = 3.8	-1.31	-2.29 **	5717 (s.d. 745) ‡	2246 (16–4064)	-2.94 vs. 0.56 **
P. borchgrevinki	θ_i= 0 (0–1.5), θ_f= 2110 (15.8–7666)	θ_i= 0, θ_f= 15.1	-1.95 **	-19.04 **	8350 (s.d. 135) ‡	4407 (3158–4981)	NA
T. newnesi	θ_i= 0.005 (0–2.1), θ_f= 6.58 (3–1645)	θ_i= 0, θ_f= 5.2	-0.83	-7.81 *	889 (s.d. 116) ‡	482 (180–696)	NA
S7
T. bernacchi	θ_i = 0 (0–0.5), θ_f = 552 (1.3–3678) **	θ_i = 0, θ_f = 2.6	-1.21	-3.95	1587 (s.d. 156)		-16.4 vs. 0 **
T. bernacchi (Ross Sea)	θ_i = 0 (0–0.6), θ_f = 492 (1–3406)**	θ_i = 0, θ_f = 2.3	-0.88	-2.16	1387 (s.d. 191) ‡		-10.4 vs. -7.3 *
T. penelli	θ_i= 0.004 (0–0.7), θ_f= 1.354 (0.1–4081)	θ_i= 0, θ_f= 1.0	-1.15	-2.88	1813 (s.d. 274) ‡		-22.02 vs. 0 **
P. borchgrevinki	θ_i = 0 (0–0.8), θ_f = 33.1 (5.6–6823)	θ_i = 0, θ_f = 2.38	-0.15	-2.77	1013 (s.d. 151) ‡		NA
T. newnesi	θ_i = 0.006 (0–7.8), θ_f = 4.08 (1.3–29)*	θ_i = 11.6, θ_f = 0	2.09*	3.97	33 (s.d. 86)		-26.1 vs. 24.2

For the Mismatch distribution (MD) and Wakeley and Hey's (W-H) methods, we indicate the pre- (θ_i) and post- (θ_f) expansion population parameters. Significant departures from expectation under the sudden expansion model of MD are indicated with asterisks (* means P < 0.05, ** < 0.01). Tajima's D and Fu's F_s are indicated with their significance as above (in the latter case, * stands for P < 0.02 and ** for < 0.004). ML values of growth parameters estimated separately for each locus in FLUCTUATE are noted together with their standard deviation in parentheses. ‡ denotes cases where the zero growth value were excluded from the the 95% C.I. We further indicate the ML estimates of growth parameters for joined dataset as estimated by LAMARC together with 95% C.I. Finally, we also indicate the Log Likelihood values of no-founder and bottleneck models of Galtier et al. as estimated by SWEEP-BOTTLENECK software. Asterisks denote the significance of the LRT test as above. NA denotes cases, where this method was not applicable.

ISSN: 2730-7182