Table 3.

Protein homologies found for 23 candidate SNP loci under diversifying selectionb identified by nonhierarchicalc and hierarchicalf Arlequin analysis and BayeScand after outlier analysis of datasets comprising AQUA populations and the TOB_WILD population. Homologies are also shown for the 17 “consistent” loci found using all three Arlequin F _ST methodse. Genome position of transcript is from Ssa ICSASG_v2 (http://salmobase.org)

SNP Name	Nonhierarch. Arlequinc F ST (p < .01)	BayeScand F ST (q < 0.05)	All three Arlequin F ST methodse	Hierarch. Arlequinf F CT (p < .01)	Chrg on NA Maph	Position NA maph (cM)	Chrg on EU maph	Position EU maph (cM)	%seq. similarity	Genome position (transcript)i	E‐value	Matches (Gene or Protein)
ESTV_15118_153		√			1	13.7	1	94.9	70%	CIGSSA_003028.t4	1E‐75	Mitochondrial ribosomal protein s5
ESTNV_16380_81				√	3	11.5	3	0.9	99%	CIGSSA_119709.t2	2E‐125	WW domain‐containing adapter protein with coiled‐coil‐like isoform X2
GCR_cBin8095_Ctg1_103k					3	103.2	3	87.2	99%	CIGSSA_018951.t1	4E‐138	Protein phosphatase 1 regulatory subunit 1B‐like
ESTNV_36457_2447			√	√	4	0	4	6.1	99%	CIGSSA_024082.t6	0	Protogenin B‐like isoform X1
ESTNV_35352_77	√		√		4	101.6	4	109.9	72%	CIGSSA_024847.t1	1E‐163	RPA‐interacting protein rpain‐a
GCR_hBin33595_Ctg1_191	√		√	√	6	7.9	–	–	98%	CIGSSA_032258.t2	1E‐10	ATPase family AAA domain protein 5‐like isoform X1
ESTV_16674_300					6	63.7	–	–	100%	CIGSSA_034497.t1	3E‐14	Zona pellucida sperm‐binding protein 4‐like
ESTNV_33402_1114				√	6	67.4	6	66	100%	CIGSSA_032952.t1	0	Cleavage and polyadenylation specificity factor subunit 3
ESTNV_34703_1491			√	√	9	67.4	9	52.7	100%	CIGSSA_042894.t2	0	HCLS1‐binding protein 3
ESTNV_17881_371	√		√√		9	96.2	9	91.9	99%	XM_014214565	0	Protein phosphatase, Mg2+/Mn2+ dependent
ESTNV_24797_128	√	√	√		12	38.0	12	68.7	61.6%	CIGSSA_063980.t1	3.41E‐27	MHC class ii alpha chain
ESTNV_16810_167	√		√	√	12	70.4	12	108.3	99.8%	CIGSSA_062376.t1	0	CD34b1
ESTNV_27237_290		√			13	37.1	–	–	94%	CIGSSA_069043.t1	2.39E‐66	Orm1‐like protein 2
ESTNV_22997_260	√		√	√	13	52	13	85	97%	XM_014138527.1	0	RILP‐like protein 1 (LOC106568309), transcript variant X4,
ESTNV_28880_179	√			√	14	–	14	62	100%	CIGSSA_073641.t4	2E‐151	Probable 39S ribosomal protein L24, mitochondrial isoform X1
ESTNV_29368_168	√				15	34.9	15	44.2	83%	CIGSSA_079971.t5	2.00E‐161	Translocation associated membrane protein 2
ESTNV_34978_699			√	√	16	42.9	16	43.5	100%	CIGSSA_085177.t1	4E‐102	Glycine cleavage system H protein, mitochondrial‐like
ESTV_14714_122	√		√	√	19	61.5	19	52.7	98%	XM_014158407.1	0.0	Zinc transporter ZIP11‐like transcript variant X2, mRNA
ESTNV_31104_129	√				20a	–a	20	46.2	100%	CIGSSA_101213.t1	3E‐45	Rab5 gdpgtp exchange factor
ESTNV_28135_402	√			√	22	58.2	22	48.5	100%	CIGSSA_109518.t2	1E‐26	Claudin‐19 isoform X2
ESTNV_28135_544	√			√	22	58.2	22	48.5	100%	CIGSSA_109518	1E‐26	Claudin‐19 isoform X2
ESTNV_31411_621				√	22	44.5	22	35.4	100%	CIGSSA_108471.t1	1E‐92	DNA‐binding protein inhibitor ID‐1
ESTNV_31110_261j	√				1/23	63.3	23	13.2	91%	CIGSSA_112247.t1	2E‐26	Thrombopoietin receptor‐like
ESTNV_31110_721j	√				1/23	63.3	23	13.2	91%	CIGSSA_112247.t1	2E‐26	Thrombopoietin receptor‐like
ESTNV_31210_275l	√		√		1/23	118.1	23	51.4	87%	CIGSSA_112663.t1	0	Palmitoyltransferase ZDHHC7‐like isoform X1
ESTNV_35075_2399	√			√	25	73	25	52.7	92%	CIGSSA_117588.t3	0	Type I inositol 3,4‐bisphosphate 4‐phosphatase‐like isoform X3
ESTNV_25950_262				√	26/28	49.6	26	22.9	81%	CIGSSA_119802.t1	4E‐140	Phosphatidylinositol 3‐kinase regulatory subunit alpha‐like

^a p < .001.

^bAfter individuals shown by STRUCTURE to have European subspecies ancestry had been removed from the dataset.

^cNon‐hierarchical Arlequin outlier analysis (Appendix S1) was done using the nonhierarchical option and included five AQUA populations: 2010PG_AQUA, (2011PG_AQUA + 2011PGN_AQUA), 2012PG_AQUA as well as the TOB_WILD population.

^dBayeScan analysis (Figure 2; Appendix S2) is always nonhierarchical and included five AQUA populations: 2009PO_AQUA, 2010PG_AQUA, 2011PG_AQUA, 2011PGN_AQUA, 2012PG_AQUA, as well as the TOB_WILD population.

^eHierarchical Arlequin outlier analysis (Appendix S5) was performed using the hierarchical option. The three large populations of AQUA populations (2010PG, {2011PG + 2011PGN}, and 2012 PG) were placed in one group, and the TOB_WILD population was randomly split into two random populations (see text) and placed in a second group.

^fSignificant F _ST (p < .01) in three methods of Arlequin analysis including in at least one of the pairwise comparisons (Figure 2; Appendix S3).

^g“Chr” means the chromosome number on which the outlier locus was located.

^h“NA” means the position on the North American Atlantic salmon female linkage map by Brenna‐Hansen et al. (2012); “EU” means the position on the European Atlantic salmon female linkage map by Lien et al. (2011).

ⁱGenome position is shown as the reference transcript name and number containing the SNP where possible as numbering of the base pairs may change in subsequent releases of Atlantic salmon (Salmo salar) genome.

“√” means the significance value is smaller than the significance level (see text) for that method.

^jThese two SNPs are 460 base pairs apart on in the same mRNA transcript and, consequently, were in perfect linkage disequilibrium with each other.

^kThis SNP is also an outlier locus for nonhierarchical and hierarchical comparisons of the Stewiacke populations with the AQUA populations.

Chromosome containing a SNP inferred from European map (Lien et al., 2011) after known translocations were accounted for (Brenna‐Hansen et al., 2012); consequently, the exact position on the North American chromosome is unknown.

^lESTNV_31210_275 codes for a protein that plays a role in follicle stimulating hormone activation of testicular Sertoli cells (Pedram et al. 2012).