Abstract
The demographic history of the Japanese archipelago was shaped by major episodes of migration and admixture, most notably the transition approximately 3,000 years ago from the Jomon period, a long-established hunter-gatherer tradition, to the Yayoi period, when migrants from the Eurasian continent introduced agriculture and new technologies. However, the timing, extent, and regional variability of the admixture between continental migrants and indigenous Jomon populations remain poorly understood. Northwestern Kyushu has drawn particular attention because skeletal analyses have indicated that Yayoi individuals from this region retain Jomon-like morphological features. However, their genetic background remains unclear. Here, we report whole genome sequences from four Northwestern Kyushu Yayoi individuals excavated from Nagasaki Prefecture, Japan. The two individuals retained nearly full Jomon ancestry, demonstrating that unadmixed Jomon descendants persisted alongside the admixed populations until the transitional phase between the Early and Middle Yayoi periods. The other two individuals showed clear evidence of an admixture, and genetic analyses indicated that gene flow between migrants and the indigenous Jomon population had already begun in this region approximately 2.5–2.6 kya. These findings suggest that the admixture in northwestern Kyushu occurred gradually rather than through a single large-scale migration event at the onset of the Yayoi period, offering new insights into the complex demographic dynamics that shaped the genetic landscape of ancient Japan.
Introduction
Understanding how distinct human populations admix and form new genetic and cultural identities is a central question in population history. Episodes of migration and admixture have repeatedly shaped human diversity worldwide, influencing genetic ancestry as well as social, cultural, and technological transformations. For example, population movements accompanying the spread of agriculture during the Holocene fundamentally reshaped the genetic landscapes of regions such as Europe, South Asia, Southeast Asia, and Africa. The Japanese archipelago offers a particularly well-documented case study of such a transformation, in which a long-lasting hunter-gatherer culture was replaced by an agricultural society through migration and admixture.
The prehistory of the Japanese archipelago is divided into two major cultural periods: Jomon and Yayoi. The Jomon period (ca. 16,500 years ago to ca. 10th century BCE) represents a Neolithic society characterized by foraging, fishing, and the production of distinctive “cord-marked” pottery1. Archaeological and genetic evidence indicate that the Jomon people maintained a unique cultural and biological identity for millennia, largely isolated from the Eurasian continent1,2,3,4. In contrast, the Yayoi period, beginning around the 10th century BCE and named after the site where its characteristic pottery was first identified, marks the introduction of wet rice agriculture and new technologies by migrants from continental East Asia, most likely via the Korean Peninsula5,6,7.
This migration led to admixture between the indigenous Jomon and incoming continental populations, resulting in major demographic, cultural, and technological transformations. This admixture is widely regarded as the primary process shaping the ancestry of modern mainland Japanese, often summarized in the literature as the “Dual Structure Model”8, and population genetic studies have firmly established that the continental ancestry component in modern Japanese and Yayoi individuals derives mainly from groups related to the Korean Peninsula9,10,11,12.
These findings raise the questions of where and how the initial arrival of migrants from the Korean Peninsula occurred within the Japanese archipelago. Archaeological evidence indicates that northern Kyushu was the earliest region to adopt wet rice agriculture5 and exhibits numerous Korean-style artifacts6. Because the Korean Peninsula and northern Kyushu are separated by only a narrow strait, this region, including Nagasaki Prefecture, is widely regarded as the primary entry point for continental migrants. In particular, the Tsushima and Iki Islands, located between the peninsula and mainland Kyushu, are considered stepping stones along this migration route7. However, although northern Kyushu as a whole served as the major gateway for these migrants, archaeological and morphological evidence suggest that the demographic impact of continental ancestry was not uniform across the region, with northwestern Kyushu exhibiting particularly distinctive patterns.
The northwestern region of northern Kyushu has attracted considerable attention owing to the distinctive morphological traits of the remains found in this region. Analyses of ancient human skeletal remains have distinguished the people of the Yayoi period in northwestern Kyushu (referred to here as the “Northwestern Kyushu Yayoi”) from those in other regions. Northwestern Kyushu Yayoi exhibit morphological features resembling those of the Jomon people, such as a low, wide face and short stature13,14,15. In contrast, Yayoi individuals from surrounding regions, such as the Northern Kyushu Yayoi and those from the Doigahama site in Yamaguchi Prefecture, located across the Kanmon Strait from Kyushu, display greater stature and increased facial height16,17. Although these morphological characteristics are well-documented, the genetic background of Northwestern Kyushu Yayoi remains poorly understood.
A recent ancient DNA study provided important initial insights by analyzing two Northwestern Kyushu Yayoi individuals from the Shimomotoyama rock shelter site, revealing substantial genomic affinity between the Northwestern Kyushu Yayoi and the Jomon, despite the presence of continental Asian ancestry18. However, these conclusions are derived from only two individuals from a single locality, even though numerous Yayoi archaeological sites are known in this region. To achieve a more comprehensive understanding of the genetic diversity and population structure of the Northwestern Kyushu Yayoi, genomic data from a broader range of sites and individuals are required.
To address this issue, we sequenced Yayoi genomes from two additional archaeological sites associated with Northwestern Kyushu Yayoi: one from the Shomura site on Iki Island19 and three from the Neshiko site on Hirado Island20,21 (Fig. 1). The individuals excavated from Neshiko exhibited morphological features similar to those of Jomon, consistent with previous reports, whereas the individuals from Shomura could not be definitively classified due to incomplete skeletal remains (Figure S1). Through comprehensive population genetic analyses, we characterized the genetic profiles of these Northwestern Kyushu Yayoi individuals, quantifying the extent of their retained Jomon ancestry and estimating the timing of admixture based on individuals who carry ancestry derived from migrants.
Geographical location of the excavation sites for the newly analyzed samples. Location of Kyushu Island within the Japanese archipelago. Neshiko and Shomura, the excavation sites of the analyzed Northwestern Kyushu Yayoi individuals, are highlighted by red rectangles.
Results
Chronological and dietary context of the Yayoi individuals
To establish the temporal and dietary contexts of Yayoi individuals, we performed radiocarbon dating and stable isotope analyses of bone collagen to clarify when they lived and the subsistence strategies they practiced during the early Yayoi period. The measured atomic carbon and nitrogen ratios (C/N), δ13C and δ15N values, 14C ages, and calibrated ages of the four bone collagen samples are summarized in Table S1. The C/N ratio range from 3.3 to 3.7. Three of these values were within the recommended range (2.9–3.6) for well-preserved bone collagen22, suggesting that the extracted collagen samples were suitable for radiocarbon dating. The value for sample MSMR-No5 (Shomura) deviates slightly from the recommended range, possibly indicating minor contamination by biogenic substances such as humic acids. The δ13C values range from − 16.3‰ to − 13.5‰, with a mean of − 15.1‰ and a standard deviation (SD) of 1.2‰. The δ15N values range from 10.6‰ to 16.7‰, with a mean of 13.2‰ and an SD of 2.7‰. Figure S2 shows the relationship between δ13C and δ15N isotope values in bone samples, compared with reference data reported by Yoneda et al.23. These values are higher than those expected for terrestrial herbivore consumption. This suggests that individuals may have relied not only on terrestrial ecosystems, but also on protein sources with higher nitrogen isotope ratios, such as aquatic resources or rice-based agricultural products.
The radiocarbon (14C) ages of the four samples ranged from 2,441 ± 20 year before present (BP) to 2,227 ± 20 year BP (Table S1). Since these humans obtained part of their dietary proteins from marine resources, bone collagen contains a proportion of marine-derived carbon. Therefore, the marine reservoir effect should be considered when obtaining accurate calendar ages from collagen-based radiocarbon dates. The proportion of marine resources in each collagen sample was estimated from the carbon isotope ratio (δ¹³C), following the methodology of Arneborg et al.24. These estimated values were used to construct a mixed calibration curve based on the IntCal2025 and Marine2026 datasets, and calibration was performed using OxCal. The local reservoir effect was corrected using a ΔR value of − 296 ± 3527, as employed in the analysis of human bones from the Otomo site in the same Saga Prefecture28. Figure S3 shows the probability distributions of the calibrated dates for the four collagen samples. The 95.4% probability ranges of the calibrated calendar dates are as follows: NSK–11 (Neshiko 11), 345–281 cal BC (7.8%) and 258–34 cal BC (87.6%); NSK–13 (Neshiko 13), 392–149 cal BC; NSK–16 (Neshiko 16), 396–166 cal BC; and MSMR-No5 (Shomura), 506–184 cal BC. The median calibrated ages of the samples ranged from 348 to 148 cal BC.
Maternal and paternal lineages reflecting dual origins
To assess the maternal and paternal genetic lineages of the Yayoi individuals and evaluate potential contributions from both continental migrants and indigenous populations, we analyzed the mitochondrial DNA (mtDNA) and Y-chromosome haplogroups of the four Yayoi individuals (Table 1). The mtDNA haplogroup of the Shomura Yayoi sample from Iki Island was identified as D4c1a, whereas the Y chromosome belonged to the O1b2a1a1a lineage. Both mtDNA haplogroup D4 and Y chromosome haplogroup O1b2 (formerly O2b) are common lineages not only among the Japanese, but also among continental East Asians, including Koreans, Chinese, and Mongolians29,30. O1b2a1a1 (formerly O2b1) is prevalent in Japan and Korea30. These haplogroups may have been introduced into the Japanese archipelago from continental East Asia during the Yayoi period. Because the mtDNA haplogroups typically observed in Jomon individuals are M7a and N9b31,32, the presence of haplogroup D4c1a in the Shomura individual suggests that this person carried continental ancestry and was likely a descendant of admixed populations of migrants and indigenous Jomon people.
In contrast, the mtDNA of three Northwestern Kyushu Yayoi individuals from the Neshiko site was identified as belonging to haplogroup M7a1a, whereas two male individuals carried the Y chromosome haplogroup D1a2a1c1. Both mtDNA and Y chromosome haplogroups are rare in continental Eurasia but are frequently found in present-day Japanese populations and are considered to be associated with Jomon ancestry haplogroups31,33,34,35. It is worth noting, however, that one of these individuals (Neshiko 11) carries continental-derived ancestry according to our nuclear genome analyses, highlighting that uniparental markers do not necessarily capture the full complexity of admixture history.
Kinship relationships among individuals
To establish whether they represented independent members of the population or closely related family members, we calculated the kinship coefficients (KING-robust36 among the four Yayoi individuals (Table 2). Neshiko 13 and Neshiko 16 may have a third-degree familial relationship, because their kinship coefficient (0.0801138) falls between 0.08839 and 0.04419. This third-degree relationship was further supported by an IBD1 value of 0.264 for this pair, whereas all other pairs exhibited extremely low IBD1 and IBD2 values (Table S7). Because all six pairs of the four Yayoi individuals exhibited kinship coefficients below 0.08839, confirming the absence of second-degree familial relationships, we included all four Yayoi individuals in subsequent population genetic analyses.
Genetic affinities of Northwestern Kyushu Yayoi individuals
To investigate how Northwestern Kyushu Yayoi individuals are genetically related to other ancient and modern East Asian populations, we conducted principal component analysis (PCA) and ADMIXTURE37 analyses. The PCA plot shown in Fig. 2 includes data from the newly sequenced Northwestern Kyushu Yayoi genomes along with those from ancient and modern Northeast and East Asians. In this plot, PC1 appears to represent the north-south genetic cline of East Eurasians, while PC2 seems to distinguish East Eurasians from the Jomon people. At the bottom of the PCA plot, Jomon individuals from the Japanese archipelago clustered together with genetically Jomon-like individuals, including Japan_Nagabaka_2800BP from the Nagabaka site in the Ryukyu Archipelago and Korea_Yokchido from the Yokchido site, the southernmost island of Korea38. Present-day Japanese are positioned between the continental East Asian and Jomon clusters.
Principal component analysis of Northeast and East Asians. Modern individuals are labeled with text, whereas ancient individuals are plotted as symbols, with colors indicating population affiliation.
The Northwestern Kyushu Yayoi individuals (Shomura, Neshiko 11, Neshiko 13, and Neshiko 16) did not cluster with present-day Japanese; rather, they were located toward the Jomon cluster (Fig. 2). Notably, the extent of their affinity to the Jomon cluster differed among individuals. Shomura and Neshiko 11 were positioned between the Jomon and present-day Japanese clusters, similar to the Shimomotoyama Yayoi individuals, who also belong to the Northwestern Kyushu Yayoi18. In contrast, the two other individuals from the Neshiko site, Neshiko 13 and Neshiko 16, clustered directly with Jomon individuals.
We also confirmed that using diploid genotype data for the four Yayoi individuals yielded nearly identical PCA results (Figure S5), indicating that the overall pattern was robust to genotype calling methods. Given this consistency, we expected that analyses based on pseudo-haploid data would produce comparable results; therefore, subsequent ADMIXTURE analyses were conducted using pseudo-haploid genotypes for the four Northwestern Kyushu Yayoi individuals. A previous study showed that ancient and present-day Japanese populations can be characterized by three major ancestral components: East Asian-related, Northeast Asian-related, and Jomon-related ancestries12. In the ADMIXTURE analysis for K = 3 (Figure S6), the Shomura and Neshiko 11 individuals exhibited three major ancestry components: an East Asian-related lineage (blue), a Northeast Asian-related lineage (yellow), and a Jomon-related ancestry (red). This pattern was similar to that observed in the Northern Kyushu Yayoi individual (KumaNishioda) and the individual from the Doigahama site in the Yamaguchi Prefecture (Doigahama), as reported in a previous study12. These results suggest that both Shomura and Neshiko 11 were affected by admixture, carrying genetic contributions from continental ancestries while retaining a substantial proportion of Jomon-derived ancestry. In contrast, Neshiko 13 and Neshiko 16 displayed an ancestry profile indistinguishable from that of the Jomon, with no detectable contribution from either East Asian-related or Northeast Asian-related lineages, suggesting that they were likely unadmixed descendants of the Jomon people.
Genome-wide evidence for Jomon-related ancestry
To quantify the genetic affinity of Yayoi individuals to the Jomon and other populations, we performed outgroup f3 and f4 analyses. These statistics evaluate shared ancestry and help identify individuals and populations that are genetically close to the target individual or population. The outgroup f3-statistics of form f3(Mbuti, target, X), where the target corresponds to each Northwestern Kyushu Yayoi individual, showed that all four Yayoi individuals exhibited close genetic affinities with one another and consistently higher affinities to Jomon groups, including those from the Ryukyu Archipelago (Japan_Nagabaka_2800BP) and Yokchido in Korea (Korea_Yokchido)38, compared to present-day Japanese or any other populations (Fig. 3). Neshiko 13 and Neshiko 16 were grouped together in this analysis because they were suggested to share a close familial relationship (Table 2) and showed the highest f3 values (Figure S7). Neshiko 13 and Neshiko 16, which clustered with Jomon individuals in the PCA (Fig. 2) and exclusively displayed Jomon-related ancestry in the ADMIXTURE analysis (Figure S6), had a pronounced contrast in f3 values between the Jomon and non-Jomon populations (Fig. 3C), which is fully consistent with these patterns. Although Shomura and Neshiko 11 were inferred to have been affected by the admixture between the Jomon and continental migrant groups based on the ADMIXTURE analysis (Figure S6), both individuals were found to be genetically close to Neshiko 13 and Neshiko 16. Consistent with this interpretation, outgroup f3-statistics calculated using Shimomotoyama Yayoi individuals, another Northwestern Kyushu Yayoi group inferred to have been affected by admixture18, also showed a similarly high affinity to Neshiko 13 and Neshiko 16 (Figure S8B). Taken together, these results are most plausibly explained by the Northwestern Kyushu Yayoi individuals sharing genetic components derived from the Jomon people who lived in this region. In addition, the close genetic affinity observed between the Northwestern Kyushu Yayoi individuals and the Japan_Nagabaka_2800BP individuals (Figs. 3, S7, S8B, and S8C), which have been shown in previous studies to be genetically close to mainland Jomon populations38, suggests that Late Jomon people from northwestern Kyushu and the Japan_Nagabaka_2800BP individuals may have shared a relatively recent common ancestry.
f3(Mbuti; target, X). Outgroup f3(Mbuti; target, X) was calculated for Yayoi individuals from Northwestern Kyushu, with the target set as (A) Shomura, (B) Neshiko 11, and (C) Neshiko 13 and Neshiko 16 (combined). For each target, the top 30 populations with the highest f3 values in the dataset are shown. The error bars represent the range of |Z|≤1.
Another noteworthy observation from the outgroup f3 analysis is that present-day Japanese individuals show a tendency to be genetically closer to Yayoi individuals who carry continental-derived ancestry (i.e., individuals such as Shomura, Neshiko 11, and the Doigahama Yayoi) than to continental populations, even though the error bars overlap and the differences are not statistically significant (Figure S8A). This trend is comparable to the affinity observed between present-day Japanese and Kofun-period individuals, indicating that both the Yayoi and Kofun groups are similar to present-day Japanese individuals. This pattern suggests that the formation of the ancestral population underlying present-day Japanese may have been underway by the Yayoi period.
The outgroup f3 analysis revealed that Shomura and Neshiko 11 are genetically close to the Jomon-like Neshiko 13 and Neshiko 16 individuals. To further evaluate which individual or population is genetically closer to each Northwestern Kyushu Yayoi individual than to the reference mainland Jomon group, we next applied f4-statistics of the form f4(Mbuti, Jomon; target, X), where “Jomon” refers specifically to the pooled mainland Jomon population (Honshu, Shikoku, Kyushu, and Hokkaido) (Fig. 4A and B). The Shomura and Neshiko 11 individuals showed significantly positive f4 values with Jomon-like individuals, such as Neshiko 13 and Neshiko 16 (Nagasaki_Neshiko13_16), Japan_Nagabaka_2800BP, and Korea_Yokchido, indicating that both individuals have a substantially stronger genetic affinity to these groups than to the mainland Jomon population.
f4(Mbuti, Jomon; target, X). f4(Mbuti, Jomon; target, X) was calculated for Yayoi individuals from Northwestern Kyushu, with the target set as (A) Shomura, (B) Neshiko 11, and (C) Neshiko 13 and Neshiko 16 (combined). The “Jomon” population refers to previously reported Jomon genomes from mainland Japan (Honshu, Shikoku, Kyushu, and Hokkaido), which were grouped into a single population for this analysis. For each target, the top 15 populations with the highest f4 values in the dataset are shown. Thick error bars indicate the range of |Z|≤1, while thin error bars indicate the range of |Z|≤3.
In the above analyses, all four Northwestern Kyushu Yayoi individuals were suggested to be genetically close to the Jomon-like individuals from the Nagabaka site in the Ryukyu Archipelago (Japan_Nagabaka_2800BP). To evaluate whether the Late Jomon people from northwestern Kyushu and the Japan_Nagabaka_2800BP individuals shared a relatively recent common ancestry, we calculated f4-statistics for the Neshiko 13 and Neshiko 16 individuals, both of whom exhibited genome-wide profiles equivalent to those of the Jomon individuals (Fig. 4C). The f4 value being nearly zero for Japan_Nagabaka_2800BP suggests that Neshiko 13 and Neshiko 16 are at least as genetically close to Japan_Nagabaka_2800BP as they are to the mainland Jomon population.
Regional variation in genetic affinity to Jomon populations
In the preceding f-statistic analyses, all mainland Jomon individuals were grouped into a single population labeled “Jomon.” However, Jomon populations are expected to exhibit regional genetic heterogeneity across the Japanese archipelago. To determine whether Northwestern Kyushu Yayoi individuals exhibit differential affinities to Jomon populations from various regions of the Japanese archipelago (Figure S9), we calculated the f4(Mbuti, target; Jomon1, Jomon2) statistics (Figures S10, S11, and S12). Fundamentally, the f4-statistic takes a negative value when the target (i.e., the Yayoi individual) is genetically closer to Jomon1 than to Jomon2 and a positive value when the target is closer to Jomon2 than to Jomon. The f4 analysis indicated that the Northwestern Kyushu Yayoi individuals exhibited asymmetric genetic relationships with Jomon individuals from different regions. In particular, they showed the highest genetic affinity with a Late Jomon individual from the Shikoku region, which was the Jomon sample in our dataset and was both geographically and temporally closest to the Kyushu region. This result is consistent with a previous analysis of a Yayoi individual excavated from the Doigahama site in Yamaguchi Prefecture, which is adjacent to Kyushu12.
Because this analysis relies on a small number of Jomon genomes from each region, and because the temporal and regional population structure of the Jomon has not yet been rigorously characterized using high-quality genomes, these results should be interpreted as preliminary. This limitation is evident for the “Japan_Shikoku_LateJomon” group, which is represented by only one low-coverage individual (< 1×)39, making the signal especially susceptible to stochastic variation. Even so, the emerging pattern suggests the possibility of regional differentiation among the Jomon, a hypothesis that can be more robustly evaluated as additional high-quality genomes from across the Japanese archipelago become available.
Admixture modeling between Jomon and continental lineages
To model the proportion of Jomon and continental ancestry components in the Yayoi and later Japanese populations, we applied qpAdm40 (Fig. 5; Table S8). The analysis showed that both present-day and ancient Japanese can be modeled as deriving from an admixture between the Jomon and Korean-related lineages, with P-values exceeding 0.05. Furthermore, we found that, in addition to the previously published data from Jomon individuals (Fig. 5A), data from the Neshiko 13 and Neshiko 16 individuals, whose genome-wide profiles are essentially indistinguishable from those of the Jomon individuals (Figs. 2 and S6), can serve as potential sources of Jomon-related ancestry (Fig. 5B). Using Neshiko 13 and Neshiko 16 as Jomon sources led to a slight, although not statistically significant, increase in the estimated proportion of Jomon-related ancestry across the analyzed populations (Fig. 5).
Admixture proportions estimated by qpAdm analysis. Error bars represent the range of |Z| ≤ 1. (A) Admixture proportions were estimated using previously reported mainland Jomons as the source of Jomon-related ancestry. (B) Admixture proportions were estimated using the combined data of Neshiko 13 and Neshiko 16, which were genetically classified as Jomon individuals, as sources of Jomon-related ancestry.
Northwestern Kyushu Yayoi individuals who show evidence of mixed ancestry (i.e., Shomura, Neshiko 11, and Shimomotoyama) exhibited higher proportions of Jomon-related ancestry than Yayoi individuals from northern Kyushu (i.e., KumaNishioda Yayoi) and Yamaguchi Prefecture (i.e., Doigahama Yayoi). Despite the geographic proximity of northwestern Kyushu to the Korean Peninsula, this pattern suggests that the scale of migration into this region may have remained relatively limited up to this point in time.
Timing of admixture events
Finally, to estimate when admixtures between Jomon- and continent-derived lineages occurred, we applied the DATES method41 based on the patterns of linkage disequilibrium decay. The Shomura individual was estimated to have lived approximately 4–5 generations after the admixture event (4.860 ± 1.824 or 4.222 ± 1.566 generations), whereas Neshiko 11 was estimated to have lived about 10–11 generations after the admixture (9.632 ± 2.082 or 10.602 ± 2.623 generations) (Table 3). In the DATES analysis, assuming a Jomon-Korean admixture model, the estimated admixture times for various Japanese populations ranged from approximately 2,600 to 1,900 BP. These findings are consistent with the historical view that migration from the Eurasian continent occurred gradually throughout the Yayoi period, and declined after the Kofun period.
Discussion
The ancient genomes of Northwestern Kyushu Yayoi excavated in Nagasaki Prefecture exhibited significantly higher proportions of Jomon-derived ancestry than present-day Japanese populations, and the individuals themselves showed substantial variation in the proportions of this ancestry. The results of the admixture dating further provided important insights into the timing of migration from the Korean Peninsula to the Japanese archipelago.
The beginning of the Yayoi period was defined primarily based on archaeological criteria, such as the appearance of distinctive Yayoi pottery styles characterized by burnished reddish surfaces and sunken patterns (incised line decorations), wet rice agriculture, and other cultural elements, and remains a subject of scholarly debate, with estimates ranging from the 10th to the 3rd century BCE. Recent radiocarbon dating of charred botanical remains and other archaeological materials has suggested that the onset of the Yayoi period dates back as early as the 10th to 9th centuries BCE42,43. While such chronologies reflect cultural changes, they do not directly reveal when biological interactions such as admixtures between populations began. In this regard, our study adds a genetic perspective by estimating the timing of admixture between the Jomon people and migrants from the Korean Peninsula. Specifically, admixture dating of the ~ 2,441-year-old Shomura individual suggests that gene flow had already begun in this region just under 2.6 kya. Although this genetic evidence does not redefine the beginning of the Yayoi period, it provides additional insight into the timing of demographic interactions that likely accompanied the associated cultural transformations.
Among ancient Japanese genomes from the post-Yayoi period, the Ikawazu individual (IK002), dated to approximately 2,681 years before present4, uniquely displays a genetic profile that is largely consistent with that of the Jomon. In this study, we identified two Northwestern Kyushu Yayoi individuals, Neshiko 13 and Neshiko 16, dated to 2,340 and 2,355 years ago, whose genomes are composed almost entirely of Jomon ancestry. These findings demonstrate that during this period, the Japanese archipelago was inhabited not only by individuals of mixed ancestry between the Jomon and continental migrants but also by people who were direct descendants of the Jomon. Notably, such Jomon-descendant individuals were present even in northwestern Kyushu, a region geographically closest to the Korean Peninsula and regarded as a primary entry point for continental migrants. This situation parallels other cases worldwide in which indigenous genetic lineages persist locally despite major cultural and demographic transitions. For example, the Mesolithic hunter-gatherer ancestry remained in parts of Iberia during the Neolithic, and a similar persistence of forager ancestry was observed in Scandinavia and sub-Saharan Africa44,45,46. Such examples highlight that cultural transitions do not necessarily entail complete genetic replacement but can result in complex mosaics of continuity and admixture. In Japan, this meant that during the Middle Yayoi period, populations included both admixed descendants of continental migrants and direct Jomon lineages, representing one of the most genetically diverse moments in Japanese prehistory.
Previous genetic studies have shown that after the Yayoi period, the admixture between the Jomon and continental-derived populations progressed throughout much of mainland Japan. Although present-day populations in the Tohoku region retain slightly higher proportions of the Jomon ancestry, the overall ratio of Jomon to continental ancestry is remarkably uniform among mainland Japanese47. In contrast, peripheral groups, such as the Ainu and Ryukyuans, appear to have been less affected by continental migrants. Genome-wide analyses have shown that they form distinct genetic clusters and retain higher proportions of Jomon ancestry than mainland Japanese48. These broader patterns provide an important contextual framework for interpreting our findings in northwestern Kyushu, which captures one of the early local stages of this demographic transition.
Our study revealed substantial genetic diversity within the Northwestern Kyushu Yayoi, a group previously defined largely on the basis of shared Jomon-like morphological traits. The wide variation in Jomon ancestry observed among individuals from the Shomura, Neshiko, and Shimomotoyama sites indicates that the Northwestern Kyushu Yayoi cannot be regarded as a homogeneous population. Although the individuals analyzed here may not fully capture the genetic diversity of the entire population, increasing the number of samples in future studies will be essential for refining our understanding of the genetic landscape of the Yayoi people in this region.
Materials and methods
Ancient sample information
In this study, we first conducted radiocarbon dating of bone collagen extracted from four Northwestern Kyushu Yayoi individuals excavated from two Middle Yayoi period sites in Nagasaki Prefecture, Japan, to confirm their chronological placement. One adult male (Shomura) was recovered from a stone coffin at the Shomura site on Iki Island19. Although his humerus and femur were well preserved, most of the skull was missing, precluding morphological determination. In addition, three individuals were excavated from the Neshiko site, a Middle Yayoi period settlement on Hirado Island20,21: one adult male (Neshiko 11), one child (Neshiko 13), and one adult female (Neshiko 16). Following radiocarbon dating, all four individuals were sequenced for population genomic analyses. This study was approved by the Ethics Committee of Toho University School of Medicine (A23103_A20110_A18099_A18056).
Radiocarbon dating and stable isotope analysis
Each bone fragment was initially cleaned by sandblasting to remove surface contaminants, such as adhering sediments. This was followed by an ultrasonic treatment in deionized water and an 8-hour immersion in acetone. To further eliminate humic substances, the samples were soaked in 0.2 M sodium hydroxide (NaOH) solution for 8 h. After thorough rinsing and drying, collagen was extracted from the treated bone material according to established protocols. Radiocarbon dating of the extracted collagen was conducted using a compact accelerator mass spectrometry (AMS) system equipped with a 0.5 MV Pelletron accelerator (National Electrostatics Corporation, Middleton, WI, USA). The measurements were performed at the Yamagata University AMS facility (YU-AMS), Japan49. The obtained radiocarbon ages were calibrated using OxCal version 4.450. Stable carbon and nitrogen isotope ratios in collagen were analyzed using an elemental analyzer (EA; Vario MICRO cube, Elementar Analysensysteme GmbH, Langenselbold, Hesse, Germany) coupled with an isotope ratio mass spectrometer (IRMS; IsoPrime, Isoprime Ltd., Manchester, UK). Both instruments were housed at the YU-AMS facility51. The carbon (13C/12C) and nitrogen (15N/14N) isotopic values are expressed as δ13C and δ15N relative to internationally accepted standards.
DNA extraction
DNA was extracted from the petrous bones of four individuals: Shomura, Neshiko 11, Neshiko 13, and Neshiko 16. The left petrous bone was used for Shomura and Neshiko 11 and the right petrous bone was used for Neshiko 13 and Neshiko 16 (Figure S1). The sampling targeted the dense region of the petrous bone surrounding the cochlea. The outer surface of each bone was removed using a sanding tool (Dremel, Robert Bosch Tool Corporation, Mt. Prospect, IL, USA) and the cleaned inner portion was ground into a fine powder using a bead mill (Multi-Beads Shocker MB601U; Yasui Kikai, Osaka, Japan). The powder was decalcified in 0.5 M EDTA (pH 8.0) at 56 °C for 2 h in a rotating oven. After centrifugation, the supernatant was discarded, and decalcification was repeated three times. DNA was extracted using phenol: chloroform: isoamyl alcohol (25:24:1), followed by additional chloroform extraction. The aqueous phase was recovered, concentrated to 200 µL using Amicon Ultra-15 centrifugal filters (Merck Millipore, Darmstadt, Germany), and further purified using a MiniElute silica spin column (QIAGEN, Hilden, Germany).
Library preparation and target enrichment
Next-generation sequencing libraries were constructed as previously described52,53. For the Shomura individual, both double-stranded (DS) and single-stranded (SS) libraries were prepared and used for shotgun sequencing. In contrast, due to poor DNA preservation, only SS libraries were constructed for the three Neshiko individuals (Neshiko 11, Neshiko 13, and Neshiko 16). These libraries were subjected to in-solution nuclear DNA target enrichment using a Twist Ancient DNA Panel (Twist Bioscience, South San Francisco, CA, USA), which captures approximately 1.2 million genome-wide single nucleotide polymorphisms (SNPs). Enrichment was performed according to the manufacturer’s protocol and the procedures described by Rohland et al.54 using the Twist Hybridization and Wash Kit.
Sequencing and data processing
Enriched libraries were amplified and sequenced on an Illumina MiSeq platform (150-cycle v3 kit) for preliminary evaluation. Based on the assessment of DNA preservation and authenticity, high-coverage sequencing was conducted on the Illumina NovaSeq 6000 platform (150 bp paired-end reads) at SRL Inc. (Tokyo, Japan). Initially, Illumina adapter sequences were trimmed from the raw FASTQ files using AdapterRemoval v2.3.055, with filters applied to exclude reads shorter than 30 base pairs and to remove low-quality bases at the ends of the reads using the following parameters: ‘--trimns --trimqualities --minlength 30 --minquality 20 --minadapteroverlap 1 --collapse’. For SS libraries, the parameter ‘--trim5p 0 15’ was additionally used to remove the low-diversity tail at the 5′ end of R2 reads, based on quality assessments conducted using fastQC v0.11.956. Following this, the well-merged, adapter-trimmed reads were mapped to the human reference genome hs37d5 using the aln and samse modules in Burrows-Wheeler Aligner (BWA) v0.7.1757, with the following parameters: ‘-l 16500 -n 0.01 -o 2’. The output BAM files were sorted, and reads with a Phred-scaled mapping quality score below 30 were discarded using SAMtools v1.1958. PCR duplicates were subsequently removed using the MarkDuplicates tool from Picard v3.1.1 (https://broadinstitute.github.io/picard/). Ancient DNA damage patterns were assessed using mapDamage v2.2.259 (Figure S4). Based on the mapDamage plots, 10 bp were clipped from both ends of all reads in the DS library data and 3–4 bp were clipped from both ends of all reads in the SS library data using the trimBam function on bamUtils v1.0.1560 to minimize the impact of ancient DNA damage.
Authentication and contamination estimates
Modern DNA contamination was assessed using various methods. For male individuals, X chromosome–based contamination was estimated with ANGSD v0.94161 using both the ‘MoM’ and ‘new_llh’ approaches. Autosomal contamination was evaluated using verifyBamID v2.0.162 with the 1000 Genomes 100 K SNP reference panel. All the estimates indicated low contamination levels (Table S2). After verification, the DS and SS BAM files of the Shomura individual were merged using SAMtools v1.1958. Contamination was reassessed for the merged files to ensure consistency across libraries. The mean genome-wide coverage of the Shomura individual was computed using Qualimap v2.363 (Table S3).
Genotype calling and filtering
Diploid genotypes for the newly sequenced target individuals were called for each locus using the Genome Analysis Toolkit (GATK) v3.864. A GVCF file was generated using the HaplotypeCaller module with the options ‘-ERC GVCF -rf BadCigar -mbq 30’, and a VCF file was then produced from the GVCF file using the GenotypeGVCFs module with the ‘-allSites’ option. Variant sites were filtered using BCFtools v1.2058, retaining only those with a minimum depth of 10 and both a QUAL score and a genotype quality score (GQ or RGQ) of at least 30 on the Phred scale. The final number of obtained 1240K autosomal SNPs is presented in Table S4.
Pseudo-haploid genotypes were also generated by randomly sampling a single high-quality base for each autosomal SNP in the 1240 K panel, using SAMtools mpileup with the ‘-R -B -q30 -Q30’ options, followed by the randomHaploid mode in pileupCaller v1.5.2 (https://github.com/stschiff/sequenceTools). In this study, pseudo-haploid data were used only for the PCA (Figure S5) and the ADMIXTURE analysis (Figure S6). The quality metrics for the 1240 K autosomal SNPs for each sample are presented in Table S5.
Sex identification
The sex of each sample was estimated based on the number of reads mapped to the X and Y chromosomes (Table S6). The read counts for each chromosome were computed using the ‘idxstats’ command in SAMtools v1.1958. Based on these results, the \(\:{R}_{Y}\) index, Y/(X + Y), of each sample was calculated, and the genetic sex was determined following the method described in a previous study65.
MtDNA and Y chromosome haplogroup
For mtDNA and Y chromosome haplogroup analysis, mitochondrial and Y chromosomal sites were extracted with ‘FORMAT/DP > = 1’ from non–quality-filtered VCF files using BCFtools v1.20. The mtDNA haplogroup was determined using Haplogrep v3.2.166 and the Y-chromosome haplogroup was identified using Y-LineageTracker v1.3.067.
Dataset compilation
For population genetic analysis, the newly sequenced ancient Yayoi genomes were merged with the 1240 K SNP dataset (v54.1.p1), which was downloaded from the Allen Ancient DNA Resource (AADR)68 (https://reich.hms.harvard.edu/allen-ancient-dna-resource-aadr-downloadable-genotypes-present-day-and-ancient-dna-data), and processed in the same manner as described in our previous study12. Human Genome Diversity Project (HGDP) data69 and Simons Genome Diversity Project (SGDP) data70 were extracted from modern population panels. One individual from each pair with a KING kinship coefficient greater than 0.088436 was excluded using PLINK v2.00a571. The China_Lahu outlier (HGDP01319) was also excluded based on our previous analysis. In addition, ancient Eurasian genomes4,38,39,41,72,73,74,75,76,77,78,79,80 were extracted from the 1240 K dataset, and additional ancient genomes from Japan and Korea3,12,18,81 were merged.
Genetic kinship analysis
Kinship coefficients (KING-robust36 were calculated for all six possible pairs among the four Yayoi individuals, based on 1,150,639 autosomal SNPs, using PLINK v2.00a71. As no second-degree familial relationships were detected, all four Yayoi individuals were included in the subsequent population genetic analyses. To further assess possible third- to sixth-degree familial relationships, identity-by-descent (IBD) was estimated using TRUFFLE v1.38 program82.
Population clustering analyses
Principal component analysis (PCA) was conducted using smartpca v16000 from the EIGENSOFT package v7.2.183. Ancient genomes were projected onto the PC space calculated from modern individuals using the ‘lsqproject: YES’ option. Additionally, the ‘shrinkmode: YES’ option was applied when the present-day population set included only East Eurasian populations.
The ADMIXTURE v1.3.0 program40 was used for unsupervised model-based genetic clustering. To avoid biases arising from the coexistence of diploid and haploid genomes, pseudo-haploid-called modern individuals from the 1000 Genomes Project84, as available in the AADR 1240 K SNP dataset, were used instead of the diploid-called HGDP and SGDP data. Five East Asian populations (CDX, CHB, CHS, KHV, and JPT) from the 1000 Genomes Project were selected as modern reference panels.
Before running ADMIXTURE, the absence of bias between the pseudo-haploid and diploid datasets was confirmed by principal component analysis (Figure S5). The haploid genotype dataset was then merged, and SNPs with a minor allele frequency below 0.01 were filtered out using PLINK v1.90b85 with the ‘--maf 0.01’ option. Linkage disequilibrium (LD) pruning was performed using PLINK v1.9 with the ‘--indep-pairwise 200 25 0.2’ option, applying a window size of 200 SNPs, a step size of 25 SNPs, and an r² threshold of 0.2. This process retained 368,630 SNPs. ADMIXTURE was run with the ‘--haploid=‘*’ --cv’ options.
f-statistics
To assess the genetic relationships between the target ancient and other East Eurasian populations, we computed the f-statistics using the qp3Pop v651 and qpDstat v980 programs in the ADMIXTOOLS v7.0.2 package40.
First, outgroup f3-statistics of the form f3(Mbuti; target, X) were calculated, where Mbuti pygmies served as the outgroup and X represented one of the populations in the dataset. In addition, f4-statistics of the form f4(Mbuti, Jomon; target, X) were computed to identify the Jomon population that was genetically closer to each Northwestern Kyushu Yayoi individual than to the reference Jomon group from mainland Japan. Here, previously reported Jomon genomes from mainland Japan (Honshu, Shikoku, Kyushu, and Hokkaido) were grouped into a single reference population labeled “Jomon.” For all f-statistics, Z-scores representing the deviation from zero in units of standard error were also calculated.
Furthermore, f4-statistics of the form f4(Mbuti, target; Jomon1, Jomon2) were calculated to test whether any Jomon subpopulations exhibited exceptionally high genetic affinity for the target individuals. The archaeological locations of the Jomon groups used in this analysis are shown in Figure S9.
Admixture modeling
The qpWave and qpAdm v1520 programs in the ADMIXTOOLS v7.0.2 package were used to model newly sequenced ancient individuals as an admixture of Jomon and continental migrants. These programs were run without the “allsnps: YES” option to avoid potential bias introduced by the option86.
The following populations were used as outgroup references for qpAdm modeling: Mbuti, Italy_North_Villabruna_HG73, Iran_GanjDareh_N41, Tianyuan74, USR175, Papuan, Ami, Bianbian_EN, Liangdao2_EN78, and She.
The qpWave program was run to assess the independence of the outgroup populations using a significance threshold of P = 0.01. After confirming the independence of the outgroups, a two-way admixture model involving Jomon and present-day Koreans (from the SGDP dataset) was tested for both ancient and present-day Japanese target populations using qpAdm, in accordance with our previous findings12. Two Jomon–Korean admixture models were tested: one using mainland Jomon genomes (Honshu, Shikoku, Kyushu, and Hokkaido) grouped into a single population, and the other using the combined Neshiko 13 and Neshiko 16 data as the Jomon-related source. All qpAdm models were computed using more than 50,000 SNPs after dataset intersection; therefore, we considered the results to be reliable without applying an explicit minimum SNP threshold.
Estimation of admixture date
Admixture dates were estimated for the Yayoi individuals from the Shomura and Neshiko sites, as well as for present-day Japanese and Kofun-period Japanese, using DATES v75341 with the following parameters: “jackknife: YES, maxdis: 1.0, binsize: 0.001, runmode: 1, mincount: 1, runfit: YES, lovalfit: 0.45, afffit: YES, and qbin: 100”. Two Jomon–Korean admixture models were examined: one using mainland Jomon genomes as the Jomon source and another using the combined Neshiko 13 and Neshiko 16 data as the Jomon-related source. A generation time of 29 years, which is the default value in the DATES program, was assumed to convert the number of generations into calendar years. The estimated radiocarbon dates for Japan_Honshu_Kofun were calculated as the average of those of the three individuals: 1378 BP, 1340 BP, and 1325 BP.
Data availability
The fastq-format raw sequencing reads and BAM-format genome files of Shomura, Neshiko 11, Neshiko 13, and Neshiko 16 were deposited in the DDBJ Sequence Read Archive under the BioProject accession number PRJDB35646 (https://ddbj.nig.ac.jp/resource/bioproject/PRJDB35646).
References
Habu, J. Ancient Jomon of Japan (Cambridge Univ. Press, 2004).
Kanzawa-Kiriyama, H. et al. A partial nuclear genome of the Jomons who lived 3000 years ago in Fukushima, Japan. J. Hum. Genet. 62, 213–221 (2017).
Kanzawa-Kiriyama, H. et al. Late Jomon male and female genome sequences from the funadomari site in Hokkaido. Japan Anthropol Sci. 127, 83–108 (2019).
Gakuhari, T. et al. Ancient Jomon genome sequence analysis sheds light on migration patterns of early East Asian populations. Commun. Biol. 3, 437 (2020).
Crema, E. R., Stevens, C. J. & Shoda, S. Bayesian analyses of direct radiocarbon dates reveal geographic variations in the rate of rice farming dispersal in prehistoric Japan. Sci. Adv. 8, eadc9171 (2022).
Hudson, M. & Barnes, G. L. Yoshinogari. A Yayoi settlement in Northern Kyushu. Monumenta Nipponica. 46, 211–235 (1991).
Daifuku, H. The early cultures of the Island of Kyushu, Japan. Southwest. J. Anthropol. 5, 253–271 (1949).
Hanihara, K. Dual structure model for the population history of the Japanese. Nichibunken Jpn Rev. 2, 1–33 (1991).
Kim, Y. J. & Jin, H. J. Dissecting the genetic structure of Korean population using genome-wide SNP arrays. Genes Genomics. 35, 355–363 (2013).
Wang, Y., Lu, D., Chung, Y. J. & Xu, S. Genetic structure, divergence and admixture of Han Chinese, Japanese and Korean populations. Hereditas 155, 1–12 (2018).
Jeon, S. et al. Korean genome project: 1094 Korean personal genomes with clinical information. Sci. Adv. 6, eaaz7835 (2020).
Kim, J. et al. Genetic analysis of a Yayoi individual from the Doigahama site provides insights into the origins of immigrants to the Japanese Archipelago. J. Hum. Genet. 70, 47–57 (2025).
Naito, Y. On the human skeletons of Yayoi period excavated at the sites in North-Western Kyushu. J. Anthropol. Soc. Nippon. 79, 236–248 (1971). (in Japanese).
Saiki, K., Wakebe, T. & Nagashima, S. Cranial nonmetrical analyses of the Yayoi people in the Northwestern Kyushu area. Anthropol. Sci. 108, 27–44 (2000).
Watanabe, T., Saiki, K., Okamoto, K. & Wakebe, T. Metrical and nonmetrical analyses of modern female Crania in the Northwestern Kyushu area. Anthropol. Sci. 112, 147–159 (2004).
Kanaseki, T. The physical characteristics of Japanese in the Yayoi-period. Proc. 15th Gen. Assembly Jpn Med. Congr. 1, 167–174 (1959). (in Japanese).
Kaifu, Y. Differences in mandibular morphology of the Jomon people from Kanto and the Yayoi people from Northern Kyushu and Yamaguchi. Anthropol. Sci. 103, 291–308 (1995).
Shinoda, K. I., Kanzawa-Kiriyama, H., Kakuda, T. & Adachi, N. Genetic characteristics of Yayoi people in Northwestern kyushu: ancient genome analysis of human bones excavated from Shimomotoyama rock shelter, Sasebo, Nagasaki prefecture, Japan. Anthropol. Sci. (Jpn Ser). 127, 25–43 (2019). (in Japanese).
Matsushita, M., Mizuno, F. & Matsushita, T. Yayoi human skeletal remains from a stone coffin adjacent to the Shomura site, Iki City, Nagasaki Prefecture. Bull. Doigahama Site Anthropol. Museum. 18, 49–463 (2023). (in Japanese).
Matsushita, T. Yayoi-period human skeletal remains from the Nezushi site. Hirado City History: Nat. Archaeological, 405–441 (Hirado City, 1996). (in Japanese).
Matsushita, M. & Matsushita, T. Yayoi human skeletal remains from the Neshiko site, Hirado City, Nagasaki Prefecture. Hirado Bull. 5, 1–46 (2017). (in Japanese).
DeNiro, M. J. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806–809 (1985).
Yoneda, M. et al. Isotopic evidence of inland-water fishing by a Jomon population excavated from the Boji site, Nagano, Japan. J. Archaeol. Sci. 31, 97–107 (2004).
Arneborg, J. et al. Change of diet of the Greenland Vikings determined from stable carbon isotope analysis and 14 C dating of their bones. Radiocarbon 41, 157–168 (1999).
Reimer, P. J. et al. The IntCal20 Northern hemisphere radiocarbon age calibration curve (0–55 cal kBP). Radiocarbon 62, 725–757 (2020).
Heaton, T. J. et al. Marine20—the marine radiocarbon age calibration curve (0–55,000 cal BP). Radiocarbon 62, 779–820 (2020).
Kong, G. S. & Lee, C. W. Marine reservoir corrections (∆R) for Southern coastal waters of Korea. The sea. J. Korean Soc. Oceanogr. 10, 124–128 (2005). (in Korean).
Sakamoto, M. & Takigami, M. Recalibration of the radiocarbon ages of the ‘Yaponesia genome’ using IntCal20 and Marine20. Res. Bull. Natl. Museum Japanese History. 237, 173–186 (2022). (in Japanese).
Nohira, C., Maruyama, S. & Minaguchi, K. Phylogenetic classification of Japanese MtDNA assisted by complete mitochondrial DNA sequences. Int. J. Legal Med. 124, 7–12 (2010).
Kim, S. H. et al. High frequencies of Y-chromosome haplogroup O2b-SRY465 lineages in korea: a genetic perspective on the peopling of Korea. Investig Genet. 2, 1–11 (2011).
Adachi, N. et al. Mitochondrial DNA analysis of Hokkaido Jomon skeletons: remnants of archaic maternal lineages at the Southwestern edge of former Beringia. Am. J. Phys. Anthropol. 146, 346–360 (2011).
Kanzawa-Kiriyama, H., Saso, A., Suwa, G. & Saitou, N. Ancient mitochondrial DNA sequences of Jomon teeth samples from Sanganji, Tohoku district, Japan. Anthropol. Sci. 121, 89–103 (2013).
Mizuno, F. et al. Diversity in matrilineages among the Jomon individuals of Japan. Ann. Hum. Biol. 50, 324–331 (2023).
Tajima, A. et al. Genetic origins of the Ainu inferred from combined DNA analyses of maternal and paternal lineages. J. Hum. Genet. 49, 187–193 (2004).
Hammer, M. F. et al. Dual origins of the japanese: common ground for hunter-gatherer and farmer Y chromosomes. J. Hum. Genet. 51, 47–58 (2006).
Manichaikul, A. et al. Robust relationship inference in genome-wide association studies. Bioinformatics 26, 2867–2873 (2010).
Alexander, D. H., Novembre, J. & Lange, K. Fast model-based Estimation of ancestry in unrelated individuals. Genome Res. 19, 1655–1664 (2009).
Robbeets, M. et al. Triangulation supports agricultural spread of the Transeurasian languages. Nature 599, 616–621 (2021).
Cooke, N. P. et al. Ancient genomics reveals tripartite origins of Japanese populations. Sci. Adv. 7, eabh2419 (2021).
Patterson, N. et al. Ancient admixture in human history. Genetics 192, 1065–1093 (2012).
Narasimhan, V. M. et al. The formation of human populations in South and central Asia. Science 365, eaat7487 (2019).
Shoda, S. Y. Radiocarbon and archaeology in Japan and korea: what has changed because of the Yayoi dating controversy? Radiocarbon 52, 421–427 (2010).
Sakamoto, M., Hakozaki, M., Nakatsuka, T. & Ozaki, H. Radiocarbon dating of tree rings from the beginning and end of the Yayoi period, Japan. Radiocarbon 66, 1991–1999 (2024).
Skoglund, P. et al. Genomic diversity and admixture differs for Stone-Age Scandinavian foragers and farmers. Science 344, 747–750 (2014).
Schlebusch, C. M. et al. Southern African ancient genomes estimate modern human divergence to 350,000 to 260,000 years ago. Science 358, 652–655 (2017).
Olalde, I. et al. The genomic history of the Iberian Peninsula over the past 8000 years. Science 363, 1230–1234 (2019).
Watanabe, Y. & Ohashi, J. Modern Japanese ancestry-derived variants reveal the formation process of the current Japanese regional gradations. Iscience 26, 106130 (2023).
Jinam, T. et al. The history of human populations in the Japanese Archipelago inferred from genome-wide SNP data with a special reference to the Ainu and the Ryukyuan populations. J. Hum. Genet. 57, 787–795 (2012).
Tokanai, F. et al. Present status of YU-AMS system. Radiocarbon 55, 251–259 (2013).
Ramsey, C. B. Bayesian analysis of radiocarbon dates. Radiocarbon 51, 337–360 (2009).
Kato, K., Tokanai, F., Anshita, M., Sakurai, H. & Ohashi, M. S. Automated sample combustion and CO2 collection system with IRMS for 14 C AMS in Yamagata University, Japan. Radiocarbon 56, 327–331 (2014).
Mizuno, F. et al. A study of 8,300-year-old Jomon human remains in Japan using complete mitogenome sequences obtained by next-generation sequencing. Ann. Hum. Biol. 47, 555–559 (2020).
Mizuno, F. et al. Population dynamics in the Japanese Archipelago since the pleistocene revealed by the complete mitochondrial genome sequences. Sci. Rep. 11, 12018 (2021).
Rohland, N. et al. Three assays for in-solution enrichment of ancient human DNA at more than a million SNPs. Genome Res. 32, 2068–2078 (2022).
Schubert, M., Lindgreen, S. & Orlando, L. AdapterRemoval v2: rapid adapter trimming, identification, and read merging. BMC Res. Notes. 9, 1–7 (2016).
Andrews, S. FastQC: a quality control tool for high throughput sequence data. Babraham Bioinformatics. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (2010).
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
Li, H. et al. The sequence alignment/map format and samtools. Bioinformatics 25, 2078–2079 (2009).
Jónsson, H., Ginolhac, A., Schubert, M., Johnson, P. L. & Orlando, L. mapDamage2. 0: fast approximate bayesian estimates of ancient DNA damage parameters. Bioinformatics 29, 1682–1684 (2013).
Jun, G., Wing, M. K., Abecasis, G. R. & Kang, H. M. An efficient and scalable analysis framework for variant extraction and refinement from population-scale DNA sequence data. Genome Res. 25, 918–925 (2015).
Korneliussen, T. S., Albrechtsen, A. & Nielsen, R. ANGSD: analysis of next generation sequencing data. BMC Bioinform. 15, 1–13 (2014).
Zhang, F. et al. Ancestry-agnostic Estimation of DNA sample contamination from sequence reads. Genome Res. 30, 185–194 (2020).
Okonechnikov, K., Conesa, A. & García-Alcalde, F. Qualimap 2: advanced multi-sample quality control for high-throughput sequencing data. Bioinformatics 32, 292–294 (2016).
McKenna, A. et al. The genome analysis toolkit: a mapreduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20, 1297–1303 (2010).
Skoglund, P., Storå, J., Götherström, A. & Jakobsson, M. Accurate sex identification of ancient human remains using DNA shotgun sequencing. J. Archaeol. Sci. 40, 4477–4482 (2013). (2013).
Schönherr, S., Weissensteiner, H., Kronenberg, F. & Forer, L. Haplogrep 3-an interactive haplogroup classification and analysis platform. Nucleic Acids Res. 51, W263–W268 (2023).
Chen, H., Lu, Y., Lu, D. & Xu, S. Y-LineageTracker: a high-throughput analysis framework for Y-chromosomal next-generation sequencing data. BMC Bioinform. 22, 1–15 (2021).
Mallick, S. et al. The Allen ancient DNA resource (AADR) a curated compendium of ancient human genomes. Sci. Data. 11, 182 (2024).
Bergström, A. et al. Insights into human genetic variation and population history from 929 diverse genomes. Science 367, eaay5012 (2020).
Mallick, S. et al. The Simons genome diversity project: 300 genomes from 142 diverse populations. Nature 538, 201–206 (2016).
Chang, C. C. et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4, s13742–s13015 (2015).
Raghavan, M. et al. Upper palaeolithic Siberian genome reveals dual ancestry of native Americans. Nature 505, 87–91 (2014).
Fu, Q. et al. The genetic history of ice age Europe. Nature 534, 200–205 (2016).
Yang, M. A. et al. 40,000-year-old individual from Asia provides insight into early population structure in Eurasia. Curr. Biol. 27, 3202–3208 (2017).
Moreno-Mayar, J. V. et al. Terminal pleistocene Alaskan genome reveals first founding population of native Americans. Nature 553, 203–207 (2018).
Sikora, M. et al. The population history of Northeastern Siberia since the pleistocene. Nature 570, 182–188 (2019).
Ning, C. et al. Ancient genomes from Northern China suggest links between subsistence changes and human migration. Nat. Commun. 11, 2700 (2020).
Yang, M. A. et al. Ancient DNA indicates human population shifts and admixture in Northern and Southern China. Science 369, 282–288 (2020).
Wang, C. C. et al. Genomic insights into the formation of human populations in East Asia. Nature 591, 413–419 (2021).
Gelabert, P. et al. Northeastern Asian and Jomon-related genetic structure in the three kingdoms period of Gimhae, Korea. Curr. Biol. 32, 3232–3244 (2022).
Lee, D. N. et al. Genomic detection of a secondary family burial in a single Jar coffin in early medieval Korea. Am. J. Biol. Anthropol. 179, 585–597 (2022).
Dimitromanolakis, A., Paterson, A. D. & Sun, L. Fast and accurate shared segment detection and relatedness Estimation in un-phased genetic data via TRUFFLE. Am. J. Hum. Genet. 105, 78–88 (2019).
Patterson, N., Price, A. L. & Reich, D. Population structure and eigenanalysis. PLoS Genet. 2, e190 (2006).
1000 Genomes Project Consortium. A global reference for human genetic variation. Nature 526, 68 (2015).
Purcell, S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 81, 559–575 (2007).
Harney, É., Patterson, N., Reich, D. & Wakeley, J. Assessing the performance of qpadm: a statistical tool for studying population admixture. Genetics 217, iyaa045 (2021).
Acknowledgements
We would like to thank Michiko Hayashi for her valuable experimental contributions, and Twist Bioscience (Japan office) for their technical assistance during the experiments.
Funding
This work was supported in part by grants from JSPS KAKENHI (Grant Numbers 24H01582 and 19H05737 to FM; 23H04838 to IN; 18H02514, 19H05341, 21H00336, 21H04779, 21H04983, 22H00421, and 23H04840 to JO) and AMED (Grant Number JP20km0405211 to JO).
Author information
Authors and Affiliations
Contributions
T.M. and M.M. provided the archaeological materials and associated information. F.T conducted the radiocarbon dating analyses. F.M. and S.U. extracted DNA from the skeletal remains. F.M. performed whole-genome sequencing. K.K. supported the experimental procedures and advised on the handling of ancient DNA. J.K., F.M., I.N., and S.U. organized and curated the archaeological and genetic data. J.K. performed statistical analyses. J.K. drafted most of the manuscript. F.T. wrote the radiocarbon dating section. F.M. wrote the sequencing analysis section. J.O. provided revisions and editorial inputs. F.M. and J.O. conceived of and supervised the study. All the authors have read and approved the final version of the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics approval
The analysis of human remains from the Shomura site was approved by the Board of Education of Iki City, Nagasaki Prefecture, Japan. The analysis of human remains from the Neshiko site was approved by the Board of Education of Hirado City, Nagasaki Prefecture, Japan.
Informed consent
Informed consent was not applicable, as the study involved ancient human remains.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Kim, J., Mizuno, F., Matsushita, T. et al. Ancient genomes reveal early-stage admixture and genetic diversity in the Northwestern Kyushu Yayoi. Sci Rep 16, 4833 (2026). https://doi.org/10.1038/s41598-026-34996-7
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41598-026-34996-7
