Background

The lower Yangtze River basin holds a pivotal role in Chinese history. As previous genetic research in this region has primarily focused on modern population datasets, the limited availability of ancient human genomes has hindered our capacity to reconstruct detailed ancient population histories and evaluate the genetic impact of Yellow River-related groups. 

Results

Here, we present the first set of ancient human genomes from the lower Yangtze River basin, comprising eight individuals from the Song to Qing Dynasties (960–1921 CE). We observed a high degree of genetic homogeneity in most samples, suggesting long-term regional genetic stability. Seven individuals were estimated to derive 69.3–100% of their ancestry from ancient Yellow River-related populations, while the remainder can be attributed to a southern East Asian substrate. Contemporary Han Chinese residing in the lower Yangtze basin can be modelled as direct genetic descendants of historical individuals from this area. Notably, one Qing Dynasty sample reveals a genetic link to the Eastern Mediterranean.

Conclusions

Our findings illustrate enduring genetic continuity in the lower Yangtze River basin throughout historical times. These findings underscore the region’s role as a genetic bridge between northern and southern East Asia, retaining local rice-farming ancestry while being shaped by southward expansions of Yellow River-related ancestry.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12915-025-02343-3.

Keywords: The lower Yangtze River basin, Historical Jiangsu, Population history, Rice farming-related ancestry

Ancient genomic data production

To clarify the genetic components of the lower Yangtze River basin, we collected 67 individuals from seven archaeological sites dating from the Han Dynasty to the Qing Dynasty in Jiangsu province, east of the lower Yangtze River basin. We applied the single-strand deoxyribonucleic acid (ssDNA) library preparation procedure and in-solution capture strategy for all samples. We identified aDNA damage pattern in our libraries and estimated modern contamination of all individuals using schmutzi (Additional file 1: Figure S1 and Additional file 2: Table S1). We retained the libraries with contamination rates below 3% and more than 50,000 single nucleotide polymorphisms (SNPs) targeted in the 1240 k panel. To further assess contamination, we utilized ANGSD and ComtamLD. All fragments were included in the analysis for libraries with contamination rates under 3%. For samples with contamination rates over 3%, we restricted the downstream analysis to only DNA fragments presenting aDNA damage patterns. Libraries with fewer than 50,000 SNPs targeted in the 1240 k panel after filtering were excluded from the study. We obtained eight individuals with endogenous DNA ranging from 3.14 to 70.29% and 130,884 to 948,990 SNPs targeted in the 1240 k panel (Tables1 and2). After trimming six bp from each end of all reads, we performed a kinship analysis and confirmed no relationships in our samples (Additional file 3: Figure S2). All individuals were included in the subsequent population genetic analysis.

All individuals who passed our filter came from three sites in northern Jiangsu Province (Kongwangshan, Lianyungang gym, and Nainaimiaodong), a relatively mountainous area that may benefit DNA preservation (Fig. 1A). These eight individuals in Jiangsu, ranging from the Song to the Qing Dynasties, provided genomic data for studying historical population structure in the lower Yangtze River basin.

Genetic stability and diversity in the lower Yangtze River basin

Due to the complex demographic history of the Lower Yangtze River basin, we aim to explore how population dynamics have shifted in this region over time. We first divided our individuals into three groups based on time (Jiangsu_Song, Jiangsu_Ming and Jiangsu_Qing) and performed principal component analysis (PCA) based on the “Human Origins” dataset (see also Methods and materials). Compared with ancient individuals from the Yellow River region, the samples from the Yangtze River basin were slightly shifted along PC1 towards the ancient southern populations (Fig. 1B). The clustering pattern of ancient Jiangsu individuals reveals their close genetic relationship with ancient Yellow River-related populations, suggests a minor contribution from southern ancestry, and indicates enduring genetic stability throughout this region’s complex demographic history.

We observed genetic homogeneity and similar genetic profiles among our individuals, except for KWSM199 from the Qing Dynasty (Fig. 2A, B). Outgroup-f₃ statistics also support a close genetic affinity of Jiangsu individuals from other dynasties (Fig. 2C, Additional file 4: Table S2A). Furthermore, their affinity was confirmed with non-significant results in thef₄(Mbuti, ancient East Asians; Jiangsu_ancient, Jiangsu_ancient), except for Nagqu1.1 k, who may have experienced continuous genetic exchanges with ancient populations from the Central Plain since the Yuan Dynasty (Additional file 5: Table S3A).

KWSM199 showed heterogeneity compared to most Jiangsu individuals (p-value < 0.05 in pairwise-qpWave analysis). Therefore, we labelled this sample as Jiangsu_Qing_o. Meanwhile, we identified that this outlier harbored a Western Eurasian-related component in ADMIXTURE analysis (Fig. 2B and Additional file 6: Figure S3). Next, we appliedf₄ analysis to quantify the affinity between Jiangsu_Qing_o and Western European populations. Slight signals inf₄(Mbuti, all ancient populations; Jiangsu_Qing_o, YR_MN/YR_LBIA) suggested that Jiangsu_Qing_o shares more alleles with some Eastern Mediterranean populations (Additional file 5: Table S3B). A suggestive result from the subsequentf₄(Mbuti, related populations; Jiangsu_Qing_o, Jiangsu_Qing) indicates a potential genetic connection between Jiangsu_Qing_o and Eastern Mediterranean-related populations (Additional file 5: Table S3C). In our admixture modelling, we used Egypt_Ptolemaic, which shares more alleles with Jiangsu_Qing_o, as a potential source. Our results indicate that Jiangsu_Qing_o can be modeled as a mixture of Yellow River-related populations and southern East Asians, with a minor contribution from Eastern Mediterranean-related lineage (~ 7.6%) (Fig. 3).

Southern China ancestry retained in historical Jiangsu individuals

Our Jiangsu individuals have shifted towards southern East Asians, indicating increased southern-related components compared to Yellow River-related populations (Fig. 1B). To investigate which population contributed to the formation of the historical Jiangsu population, we usedf₄-statistics in the form off₄(Mbuti, aEA; YR, Jiangsu_HE). As expected, significant signals confirmed the connection between Jiangsu_HE and southern East Asian populations (Z-score > 3) (Additional file 5: Table S3D). Interestingly, Tanshishan and Xitoucun, two ancient rice-farming populations in Fujian from about 4500 years ago, exhibited closer genetic affinity to our ancient Jiangsu individuals (Z-score > 3). Due to the lack of ancient DNA from the lower Yangtze River basin and considering that rice was first domesticated there, we used Tanshishan and Xitoucun as proxies for the unknown ancient populations in the lower Yangtze River basin.

We found that these two populations (i.e., Tanshishan and Xitoucun) are genetically closer to other southern ancient populations in Guangxi (i.e., GaoHuaHua, BaBanQinCen, etc.) (Additional file 5: Table S3E), suggesting the affinity between historical Jiangsu individuals and Tanshishan/Xitoucun may reflect a broad southern-related ancestry. However, we successfully modelled all historical Jiangsu individuals with the outgroup, including GaoHuaHua and BaBanQinCen: Jiangsu_Song can be 1-way modelled as YR_LBIA, while Jiangsu_Ming and Jiangsu_Qing can be modelled as a mixture of YR-LBIA (69.3–80.2%) and Tanshishan (19.8–30.7%), proving that historical Jiangsu individuals retained the local rice-farming ancestry despite the expansion of Yellow River-related populations (Fig. 3 and Additional file 7: Table S4A).

To investigate the admixture time between Yellow River-related populations and local ancestries in Jiangsu, we perform DATES analysis using YR_LBIA and Tanshishan as proxies for the northern and local population sources, respectively. Our results suggest that the genetic contribution from Yellow River-related ancestries might date to approximately 4400 years ago (Additional file 8: Table S5). Previous studies have proposed that the rice farming population from southern China may have migrated northward to the Yellow River basin between the Middle and Late Neolithic periods [15]. This is consistent with our findings, which suggest that the northern ancestries admixed with ancient Jiangsu populations around the same time. Therefore, these results support a hypothesis of a bidirectional expansion: ancient populations from the lower Yangtze River basin moved northward into the Yellow River basin, while Yellow River-related populations also expanded southward into the Yangtze River basin during the Middle to Late Neolithic periods.

Genetic contribution to modern Han Chinese

The Taiping Rebellion drastically reduced the population in the lower Yangtze River basin, especially in Jiangsu, leaving the question of what role historical Jiangsu populations played in modern Chinese populations. All Jiangsu individuals clustered with modern Han Chinese in PCA analysis (Fig. 4B). Unsurprisingly, results from the outgroup-f₃ analysis show that ancient Jiangsu individuals share the highest genetic affinity with modern Han Chinese (Fig. 4C and Additional file 4: Table S2B). This genetic connection is further supported by significant negative results from the subsequentf₄ analysis in the form off₄(Mbuti, Jiangsu_HE; modern_Han, modern_EA) (Additional file 5: Table S3F).

To quantify the genetic contribution of ancient Jiangsu individuals to modern Han Chinese from the lower Yangtze River basin and surrounding regions, we usedqpAdm to model the admixture proportions. We found that modern Han Chinese in the lower Yangtze River basin (i.e., Jiangsu, Zhejiang, and Shanghai) can be modelled as direct descendants of Jiangsu_Qing. In contrast, Han_Shandong is homogeneous to Jiangsu_Song, who have more northern ancestry than Jiangsu_Qing (Additional file 7: Table S4B).

Archaeological information

The individuals sampled in this study are from archaeological sites in Jiangsu province; all individuals are dated based on archaeological evidence.

Ancient DNA extraction and library preparation

We screened 67 individuals from seven archaeological sites in Jiangsu province. All samples were processed in the dedicated ancient DNA clean room at the Institute of Anthropology, Xiamen University. Human remains were first cleaned with 75% ethanol and 10% sodium hypochlorite solution to eliminate dust and external DNA contaminants, followed by 30 min of exposure to ultraviolet light. We used dental drills to obtain powder from teeth and the petrous parts of the temporal bones. We modified DNA extraction using Rohland’s protocol [19]: 1 ml lysis buffer containing 0.5 mM EDTA and 0.25 mg/mL Proteinase K to digest powder in a shaker at 37 ℃, 300 rpm. The DNA solution was purified with a MinElute kit (Qiagen, Germany) following the manufacturer’s manual. We applied a single-strand library preparation procedure to prepare libraries for all samples [20]. We utilized an in-solution DNA hybridization capture technique to enrich mitochondrial and nuclear DNA [21]. Enriched library sequencing was performed using the Element platform with a customized sequencing primer: ACACTCTTTCCCTACACGACGCTCTTCC.

Sequence data processing

We used AdapterRemoval v2.3.15 to trim adaptors and merge paired reads into a single sequence [22]. Merged reads at least 30 bp in length were mapped onto the human reference genome hs37d5 using the BWA-aln/samse algorithm implemented in BWA v0.7.176, with the parameters—l 1024 and—n 0.01[23]. PCR duplicates were removed using dedupe v0.12.3 [24]. Each end of the reads was trimmed using trimBam implemented in BamUtil v1.0.14 [25]. We further filtered the alignment quality using SAMtools with the parameters -q30 and -Q30[26].

Authentication of ancient DNA

We used pmdtools to analyze the deamination pattern of aDNA in all libraries [27]. We eliminated contamination rates using Schmutzi and ANGSD for libraries from all individuals and males, respectively [28,29]. Specifically, Schmutzi was used with the –uselength option for genotyping and contamination estimation, and the contDeam.pl script was applied to remove deaminated bases. For ANGSD, we used the -doCounts 1 option to count reads and -minMapQ 30 and -minQ 30 for filtering low-quality reads. Additionally, we estimated contamination using the contamination tool with the -a option for inputting the counts file and the HapMap reference for comparison. Two libraries, 21SMM158SE_1 and 21LHKM106RL, showed contamination rates above 3%. We filtered reads with a characteristic aDNA damage signature for downstream analysis using pmdtools [24] for these libraries.

Sex determination

Sex determination was based on coverage of X and Y chromosomes and autosomes [30].

Kinship detection

We used READ to detect degrees of kinship among individuals in this study [31], with normalization based on the median of all pairwise genetic distances.

Data merging

We merged our genomic data with previously published datasets using mergeit in EIGENSOFT. Two datasets were used in this study [32,33]: the HumanOrigins dataset was used for principal component analysis, and the 1240 k dataset was used in admixture analysis, outgroup-f₃,f₄,qpWave,qpAdm analysis, and DATES [34].

Principal components analysis (PCA)

We performed principal components analysis based on the HumanOrigins dataset using smartpca v16000 in EIGENSOFT with default settings, except that lsqproject was set to YES [35]. All historical Jiangsu individuals were projected onto the PCA, which was calculated with modern populations.

ADMIXTURE analysis

We pruned our dataset for linkage disequilibrium using plink v1.90 with parameters -indep-pairwise 200 25 0.4 [36–38]. Then, we performed unsupervised admixture analysis using ADMIXTURE v.1.3.0 [39].

f-statistic analysis

We calculated outgroup-f₃ usingqp3pop v651 from the ADMIXTOOLS with the parameter inbreed: YES. We calculated thef₄ analysis usingqpDstat v980 from the ADMIXTOOLS with the f4-mode: YES [34,37].

Admixture modelling

We modelled our studied populations as an admixture of the source populations usingqpAdm v810 from ADMIXTOOLS. We used settings as follows: allsnp: YES, inbreed: YES [37].

Calculating admixture dates

We applied the DATES v753 to eliminate admixture dates with the following settings: binsize, 0.001; maxdis, 0.5; runmode, 1; Seed, 77; mincount, 1; jackknife, YES; qbin, 10; lovalfit, 0.45. We assumed 29 years per generation to calculate the dates [40].

Ethics approval and consent to participate

All procedures performed in studies involving human participants were approved by the local archaeological institutions and Xiamen University (XDYX202412K88) reviewed and approved the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

• 1.Zhang J, Jiang L, Yu L, Huan X, Zhou L, Wang C, et al. Rice’s trajectory from wild to domesticated in East Asia. Science. 2024;384(6698):901–6. [DOI] [PubMed] [Google Scholar]
• 2.Jia X, Zhao D, Storozum MJ, Shi H, Bai G, Liu Z. The “2.8 ka BP cold event” indirectly influenced the agricultural exploitation during the late Zhou Dynasty in the coastal areas of the Jianghuai Region. Front Plant Sci. 2022;13: 902534. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.He K, Lu H, Zhang J, Wang C, Huan X. Prehistoric evolution of the dualistic structure mixed rice and millet farming in China. The Holocene. 2017;27(12):1885–98. [Google Scholar]
• 4.Tang Y, Marston JM, Fang X. Early millet cultivation, subsistence diversity, and wild plant use at Neolithic Anle, Lower Yangtze. China. Holocene. 2022;32(10):1003–14. [Google Scholar]
• 5.Hayashi M. The iconography of gods and beasts: ancient Chinese deities [in Chinese]. Beijing: Sanlian Press; 2009. p. 259. [Google Scholar]
• 6.Li X. Emerging from the age of doubt in ancient history [in Chinese]: Liaoning University Press; 1994.
• 7.Zhang GZ, Xu PF. The formation of Chinese civilization [in Chinese]. Beijing: New World Press; 2004. [Google Scholar]
• 8.Ge J, Cao S, Wu S. A concise history of Chinese migration [in Chinese]: Fujian People's Publishing House; 1993. p. 246–7.
• 9.Hu A. An essay on ethnic migrations after the disasters of the Yongjia Period in the Jin Dynasty [in Chinese]. J Anhui Univ (Philos Soc Sci Ed). 2010;34(05):100–11. [Google Scholar]
• 10.Mo X, Gao S. Tongzhi Era administrative offices of the two counties in Upper Jiang [in Chinese]: Jiangsu Ancient Books Publishing House; 1874.
• 11.Ge J, Cao S, Wu S. A concise history of Chinese migration [in Chinese]: Fujian People's Publishing House; 1993. p. 459–64.
• 12.Tao L, Yuan H, Zhu K, Liu X, Guo J, Min R. Ancient genomes reveal millet farming-related demic diffusion from the Yellow River into southwest China. Curr Biol. 2023; 33(22):4995-5002.e7. [DOI] [PubMed] [Google Scholar]
• 13.Wang T, Wang W, Xie G, Li Z, Fan X, Yang Q. Human population history at the crossroads of East and Southeast Asia since 11,000 years ago. Cell. 2021;184(14):3829–41 e21. [DOI] [PubMed]
• 14.Yang MA, Fan X, Sun B, Chen C, Lang J, Ko YC, et al. Ancient DNA indicates human population shifts and admixture in northern and southern China. Science. 2020;369(6501):282–8. [DOI] [PubMed] [Google Scholar]
• 15.Ning C, Li T, Wang K, Zhang F, Li T, Wu X, et al. Ancient genomes from northern China suggest links between subsistence changes and human migration. Nat Commun. 2020;11(1):2700. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Huang D, Sima J, Zeng D, Wang L. Brief report on the clearing of Song Dynasty tombs in Dafun Village. Jintan Cult Relics East. 2021;04:25–32 Chinese. [Google Scholar]
• 17.Li H, Liang W, Du P, Chen Y, Jia X. Agricultural economy investigation in Huaibei area of Jiangsu Province during the Sui and Tang Dynasties: an agricultural archaeological research from the Kongwangshan Cemetery in Lianyungang. Agric Hist China. 2022;41(06):50–62 Chinese. [Google Scholar]
• 18.Zhang D, Jia X, et al. Species identification and application potential of dendrochronology to Tang and Song dynasties coffins in northern Jiangsu Province. Quat Sci. 2024;44(4):1008–20 Chinese. [Google Scholar]
• 19.Rohland N, Glocke I, Aximu-Petri A, Meyer M. Extraction of highly degraded DNA from ancient bones, teeth and sediments for high-throughput sequencing. Nat Protoc. 2018;13(11):2447–61. [DOI] [PubMed] [Google Scholar]
• 20.Gansauge MT, Aximu-Petri A, Nagel S, Meyer M. Manual and automated preparation of single-stranded DNA libraries for the sequencing of DNA from ancient biological remains and other sources of highly degraded DNA. Nat Protoc. 2020;15(8):2279–300. [DOI] [PubMed] [Google Scholar]
• 21.Rohland N, Mallick S, Mah M, Maier R, Patterson N, Reich D. Three assays for in-solution enrichment of ancient human DNA at more than a million SNPs. Genome Res. 2022;32(11–12):2068–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.Schubert M, Lindgreen S, Orlando L. AdapterRemoval v2: rapid adapter trimming, identification, and read merging. BMC Res Notes. 2016;9:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 23.Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25(14):1754–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 24.Peltzer A, Jager G, Herbig A, Seitz A, Kniep C, Krause J, et al. EAGER: efficient ancient genome reconstruction. Genome Biol. 2016;17: 60. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Jun G, Wing MK, Abecasis GR, Kang HM. An efficient and scalable analysis framework for variant extraction and refinement from population-scale DNA sequence data. Genome Res. 2015;25(6):918–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 26.Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):giab008. [DOI] [PMC free article] [PubMed]
• 27.Skoglund P, Northoff BH, Shunkov MV, Derevianko AP, Paabo S, Krause J, et al. Separating endogenous ancient DNA from modern day contamination in a Siberian Neandertal. Proc Natl Acad Sci U S A. 2014;111(6):2229–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Korneliussen TS, Albrechtsen A, Nielsen R. ANGSD: analysis of next generation sequencing data. BMC Bioinformatics. 2014;15(1):356. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Renaud G, Slon V, Duggan AT, Kelso J. Schmutzi: estimation of contamination and endogenous mitochondrial consensus calling for ancient DNA. Genome Biol. 2015;16:224. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 30.Fu Q, Posth C, Hajdinjak M, Petr M, Mallick S, Fernandes D, et al. The genetic history of Ice Age Europe. Nature. 2016;534(7606):200–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 31.Monroy Kuhn JM, Jakobsson M, Gunther T. Estimating genetic kin relationships in prehistoric populations. PLoS One. 2018;13(4): e0195491. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 32.Mallick S, Micco A, Mah M, Ringbauer H, Lazaridis I, Olalde I, et al. The Allen Ancient DNA resource (AADR) a curated compendium of ancient human genomes. Sci Data. 2024;11(1): 182. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.Mallick S, Reich D. A curated compendium of ancient human genomes. 2023. Harvard Dataverse. 10.7910/DVN/FFIDCW.
• 34.Patterson N, Moorjani P, Luo Y, Mallick S, Rohland N, Zhan Y, et al. Ancient admixture in human history. Genetics. 2012;192(3):1065–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 35.Patterson N, Price AL, Reich D. Population structure and eigenanalysis. PLoS Genet. 2006;2(12): e190. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 36.Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015;4:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Peter BM. Admixture, population structure, and F-statistics. Genetics. 2016;202(4):1485–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 38.Lawson DJ, van Dorp L, Falush D. A tutorial on how not to over-interpret STRUCTURE and ADMIXTURE bar plots. Nat Commun. 2018;9(1):3258. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 39.Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009;19(9):1655–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 40.Chintalapati M, Patterson N, Moorjani P. The spatiotemporal patterns of major human admixture events during the European Holocene. eLife. 2022;11. [DOI] [PMC free article] [PubMed]

Supplementary Materials

Data Availability Statement

The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive in the National Genomics Data Center, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA-Human: HRA008768), which is publicly accessible at https://ngdc.cncb.ac.cn/gsa-human.