Recombination events in fish species usually occur once per chromosome arm, indicating the existence of interference after the formation of a single chiasma . Japanese flounder is a male-determined gonochoristic species  and presents the same pattern of recombination as other fish species, in which the male presents a higher recombination rate closer to the putative telomere, and the female presents a higher recombination rate closer to the centromere. Evidence of this phenomenon was reported in rainbow trout ; brown trout , zebrafish  and Atlantic halibut . The A2 map  and BACE map of Japanese flounder present the same pattern of recombination as other fish species. Figure 3 is a schematic representation of linkage group 2, and shows a clear example of this phenomenon, where a set of markers (bin) is separated by 2.3 cM from the centromeric region in the male map and 36.0 cM in the male. In contrast, the first recombination point in the female map is located at 2.2 cM from telomeric regions and 33.6 cM in the male map. Even though the lengths of the female and male maps were similar, Coimbra et al.  found an unexpected F:M 1:7.4. In that study, because the male was gynogenetically produced, it is unclear whether the ratio was influenced by the genetic origin of the male. In addition, only a few markers from 16 linkage groups were used to perform the analyses. The actual positions of those markers in the chromosomes were determined in the marker-centromere map built by Castaño-Sánchez et al. . All marker pairs analyzed by Coimbra et al. were placed far from the estimated centromeric regions, which are characterized by less recombination activity in the Japanese flounder male map and higher recombination rates in the female map. Reid et al.  reported the recombination rate in female Atlantic halibut to be twice that of the male, and observed a significant difference of F:M 1.6:1. Analysis of overall recombination rates between males and females in the BACE map confirmed a F:M ratio of 1:0.7. The improved male map is 1.4 times longer than the female map. Conversely, the male sheep map (Ovis aries) is 1.2 times longer than the female map and cattle (Bos taurus) maps present a very similar rate between sexes [35, 36]. While all chromosomes in Japanese flounder are acrocentric, the cattle karyotype contains 29 acrocentric autosomes and the sheep has 23 acrocentric and 3 metacentric autosomes . In humans, there is evidence that recombination along the chromosomes depends on the chromosome structure . The presence of acrocentric karyotypes in sheep and Japanese flounder could explain the fact that the male map is slightly longer than the female map in those species, and accordingly, similar in length in the cattle map. These findings, together with the reported existence of gaps longer than 20 cM between adjacent markers in some linkage groups, might indicate poor coverage in certain regions of the female map. Incomplete female maps might reflect that a higher proportion of crossovers in female generated maps will be missed, causing an underestimation of recombination rates in females relative to males, and therefore artificially decreasing F:M recombination ratios.
Synteny among species or genera may provide opportunities to complement initial QTL experiments with candidate gene approaches from homologous chromosomal locations identified in related model organisms . Based on the sequence homology analysis, more of Japanese flounder chromosomes were associated with T. nigroviridis chromosomes than D. rerio chromosomes; accordingly, Japanese flounder is phylogenetically more closely related to T. nigroviridis than to D. rerio . In addition, analysis results suggest that, during evolution, some chromosomes and regions have remained intact and others have been broken up. Ancient Actinopterygii (ray-finned fish) were postulated to have a 13 chromosome karyotype, composed of 52 A'-J' segments. Those blocks were mosaically arrayed within the proto-Actinopterygian karyotype and subsequently designated A-M (reviewed by ). Based on Danzmann et al. , the association of JF9 to D. rerio chromosomes Dr2, 6 and 22 and JF16 with Dr6 and 11, might indicate a relation of those linkage groups to "M" ancestral grouping of Actinopterygians. Moreover, the association of JF11 to Dr18, 19 and 25 might suggests its relation to the "J" ancestral linkage groups, while JF3 is associated with Dr9 and 23 and could be related to either ancestral "C" or "L" lineages, being "L" more likely. This data indicates that those linkage groups are likely remnants of regions that share a high degree of 3R duplicated segments.
Low-density genetic linkage maps have been published for P. olivaceus [26, 27, 29]. The map developed in the present study was built with 1,375 markers including 1.268 microsatellites, 105 SNPs and two genes, which makes it more portable to other strains and families. This facilitates its application to QTL analyses as well as comparative mapping to reference animals. The average inter-marker distances (5.0 cM and 4.4 cM in the male and female maps, respectively) offer sufficient marker density for QTL studies .
The improved maps, in addition to being useful for improving aquaculture strains, could be of assistance in the study of wild stocks in Japan, where cultured P. olivaceus are being released into the wild. Maintaining genetic variability is essential for the conservation of the species, not only to prevent inbreeding and bottleneck effects, but also to protect the genetic structure of natural stocks. Several microsatellite markers included in the improved maps have been previously used in population studies, genetic tagging, parentage determination and genetic diversity [40–44].
With 1,375 markers, the new map is presently the densest flatfish linkage map. The number of genetic markers available for other flatfish species is relatively limited. In this report, we describe the production of a large number of polymorphic microsatellite markers for P. olivaceus which could be amplified in other closely related species. Japanese flounder markers have already been used in the construction of the Atlantic halibut linkage map . Despite the limited number of comparison points between Japanese flounder and Atlantic halibut, Reid et al.  found evidences of conserved syntenic regions as well as regions of chromosome rearrangements. The markers mapped in this study, could be an important tool for future comparative map studies and to establish the correspondence between linkage groups of different flatfish species.
The microsatellite markers included in previous versions of the map [26, 27, 29] were consistently assigned to the same linkage groups in the newly developed maps. The order of those markers was conserved in most of the linkage groups. However, several markers co-segregate in clusters, preventing the determination of their precise order. Several regions in the maps remain poorly covered. JF19 in both female and male maps is short and has only a small number of markers. The A2 map presented several gaps, which tended to occur towards the putative telomere in the male map and centromere in the female . By adding more markers, several gaps were filled, but there is still the need to improve the centromeric regions of some linkage groups in the female map (Figure 1 : JF3f, JF6f, JF16f). Further studies with segregating data from different families and larger number of progeny will be necessary to enhance the distribution of the markers in the linkage maps. Physical maps could be constructed based on an existing BAC (Bacterial Artificial Chromosome) library  and they could be useful to determine the precise distribution and order of the markers in the genome.