Ba/Iyondji and TL2 populations, respectively. Bonobo females transfer among groups whereas males stay in their natal group for life [10,11]. The existence of certain haplotypes in male samples suggested that those haplotypes had been maintained over generations rather than representing occasional transfer of females, because a haplotype is found in male samples only when females who brought the haplotype produced male offspring. The results of purchase AKT inhibitor 2 clustering suggested that the observed clades were evolutionarily related to each other, with a substantial number of nucleotide substitutions, in which the mean number of pairwise haplotype differences between the clades (36.0367.69) was 4.7 times larger than that within each clade (7.7462.61) (Table S2). The 
unique haplotypes of the C clade have been reported previously, but their geographical distribution in the wild was not identified [1]. This study confirmed distribution of haplotypes in the C clade in the Wamba and Iyondji populations. One of the haplotypes of the C clade that was previously found in an exported bonobo (Accession Number AF176762 [12]) was found in the Wamba population (PPCR26 type), but several other haplotypes in the same clade that were reported in individuals in captivity (AF137491 [1], GU189665 as “PP56” and GU189670 as “PP69”Genetic Structure of BonobosFigure 2. Molecular phylogeny of haplotypes and their distribution in study populations. Left is a tree constructed by neighbor-joining (NJ) method. Three numbers on a tree path indicate percent bootstrap values (1,000 replications) obtained from statistical assessments by neighborjoining (NJ), maximum likelihood (ML) and maximum parsimony (MP) algorithms in order. Branches corresponding to partitions reproduced in less than 70 bootstrap replicates were collapsed. Each of black bars aside the tree shows a mtDNA clade inferred from the cluster analyses. Right score illustration summarized distribution of mtDNA haplotypes in study populations. Numbers with closed, open circles and without circle mean the observed number of male, female and sex-unknown samples, respectively, for each study population with different color. doi:10.1371/journal.pone.0059660.g[1]) were absent in either the Wamba or Iyondji populations. This suggests that the distribution of bonobos having the C clade haplotypes may have a broader range than that confirmed in this study. We named the sole clade 1934-21-0 consisting of autoapomorphs in TL2 as D clade. 1662274 Their related sequences have previously been reported (AJ829464 J829466 as “E3 to E5” [7]), and there was one case of type matching between a sample from this study and one from the same report (PPCR24 and AJ8294564 as “E3” [7]). However, this isolated D clade was not found in previous studies with full confidence probably due to comparison based on the short sequence lengths. Thus the component haplotypes of D clade have already been reported and we confirmed that these haplotypes comprise an independent clade.Application of long sequence reading revealed a finer image of mtDNA phylogeny in bonobos than that in a previous study [7]. However, geographical distribution of mtDNA haplotypes needs future assessment because numbers of available samples were particularly small 
(n = 7) for Salonga and Lac Tumba in this study.Genetic Diversity within PopulationsThe diversity of mtDNA in terms of haplotype diversity, mean number of pairwise differences, and nucleotide diversity varied among populations (Table.Ba/Iyondji and TL2 populations, respectively. Bonobo females transfer among groups whereas males stay in their natal group for life [10,11]. The existence of certain haplotypes in male samples suggested that those haplotypes had been maintained over generations rather than representing occasional transfer of females, because a haplotype is found in male samples only when females who brought the haplotype produced male offspring. The results of clustering suggested that the observed clades were evolutionarily related to each other, with a substantial number of nucleotide substitutions, in which the mean number of pairwise haplotype differences between the clades (36.0367.69) was 4.7 times larger than that within each clade (7.7462.61) (Table S2). The unique haplotypes of the C clade have been reported previously, but their geographical distribution in the wild was not identified [1]. This study confirmed distribution of haplotypes in the C clade in the Wamba and Iyondji populations. One of the haplotypes of the C clade that was previously found in an exported bonobo (Accession Number AF176762 [12]) was found in the Wamba population (PPCR26 type), but several other haplotypes in the same clade that were reported in individuals in captivity (AF137491 [1], GU189665 as “PP56” and GU189670 as “PP69”Genetic Structure of BonobosFigure 2. Molecular phylogeny of haplotypes and their distribution in study populations. Left is a tree constructed by neighbor-joining (NJ) method. Three numbers on a tree path indicate percent bootstrap values (1,000 replications) obtained from statistical assessments by neighborjoining (NJ), maximum likelihood (ML) and maximum parsimony (MP) algorithms in order. Branches corresponding to partitions reproduced in less than 70 bootstrap replicates were collapsed. Each of black bars aside the tree shows a mtDNA clade inferred from the cluster analyses. Right score illustration summarized distribution of mtDNA haplotypes in study populations. Numbers with closed, open circles and without circle mean the observed number of male, female and sex-unknown samples, respectively, for each study population with different color. doi:10.1371/journal.pone.0059660.g[1]) were absent in either the Wamba or Iyondji populations. This suggests that the distribution of bonobos having the C clade haplotypes may have a broader range than that confirmed in this study. We named the sole clade consisting of autoapomorphs in TL2 as D clade. 1662274 Their related sequences have previously been reported (AJ829464 J829466 as “E3 to E5” [7]), and there was one case of type matching between a sample from this study and one from the same report (PPCR24 and AJ8294564 as “E3” [7]). However, this isolated D clade was not found in previous studies with full confidence probably due to comparison based on the short sequence lengths. Thus the component haplotypes of D clade have already been reported and we confirmed that these haplotypes comprise an independent clade.Application of long sequence reading revealed a finer image of mtDNA phylogeny in bonobos than that in a previous study [7]. However, geographical distribution of mtDNA haplotypes needs future assessment because numbers of available samples were particularly small (n = 7) for Salonga and Lac Tumba in this study.Genetic Diversity within PopulationsThe diversity of mtDNA in terms of haplotype diversity, mean number of pairwise differences, and nucleotide diversity varied among populations (Table.
