豫西地区,仰韶村遗址龙山时代人群,全文共12例古人样品,大部分是四千多年的样品(只有一例5600年),常染呈现出比已发表的仰韶人数据(2019年吉大论文,郑州荥阳市距今五千年出头的汪沟人)还要“北方”(准确来说是还要黄土高原)的常染水平,令人震惊。



共获得6个父系遗传标记和10个线粒体遗传标记,3例C南支,2例O系和1例N1a,测序结果平平无奇▼▼▼



龙山时代(时间概念)是严文明老先生于1981年提出的,认为距今5000~4000年,囊括了大汶口晚期和龙山文化期,最新年代标准已经修正为距今4700~3800年。
雄风同志尽快出来走两步,解释解释,你口中的大汶口中晚期考古学因素陆续“强势扩张”,西进的同时,为何遗传却没有同步扩张???你的古山东成分哪去了?
全文半个字没提到山东狩猎采集者(博山等)对仰韶人群存在显著替换,越来越多的迹象表明,所谓的大汶口文化的对外影响力,很可能仅仅是以文化因素/技术传播为主的,并未在遗传层面有太大的影响,相反,山东一直在被皖北地区人群遗传影响。
根据原文构拟,仰韶村龙山期人群的常染可以模拟为由83-89%汪沟成分和11-17%古东北亚成分(ANA)的混合,几乎不含南方稻作农人血统。
仰韶村居民,大部分祖源成分都与五年前发表的汪沟古人高度一致,呈现出强烈的类汪沟常染格局,而且还在此基础上额外与古藏缅人群、古东北亚人群存在遗传亲和力,朝着更西、更北偏移,她就是不往东偏。
注意,这是豫西地区,年代不是六七千年,而是相当晚近的四千余年,已经处于龙山时代晚期,即将踏入夏纪年,在淮河流域的人口开始广泛南方化(稻农北上)的大背景下,豫西当地居民的常染水平仍能北得令人难以置信,实在是太可怕了。。。
•
以下是原文章介绍▼▼▼

近日,河南省文物考古研究院、厦门大学等单位在《Science Bulletin》(《科学通报》英文版)上在线发表了题为 “Ancient genomic time transect unravels the population dynamics of Neolithic middle Yellow River farmers” 的研究论文,通过对仰韶文化命名地——仰韶村遗址的仰韶文化和龙山文化时期古人基因组进行研究,揭示黄河中游地区新石器时代人群动态历史。
•
1921年,中国政府矿政顾问瑞典地质学家安特生和中国地质学家袁复礼等人,对中国河南省渑池县的仰韶村遗址进行了第一次发掘,发现大量新石器时代文化遗存,将其命名为仰韶文化,标志着中国现代考古学的诞生。2021年,O近O总O记致信祝贺仰韶文化发现和中国现代考古学诞生100周年。尽管仰韶村遗址在中国考古学史和仰韶文化研究史上占据十分重要的地位,但我们对仰韶村遗址古代人群的来源历史及其对中华民族的贡献知之甚少。仰韶村遗址的古人遗传学数据研究将有助于我们理解新石器时代黄河中游地区的人群历史。河南省文物考古研究院和厦门大学的研究人员取样提取和测序了仰韶村遗址的古人DNA,首次发表了仰韶村遗址8个属于新石器时代仰韶文化和龙山文化的个体的古基因组数据。研究发现仰韶村遗址仰韶文化与龙山文化相关人群之间具有高度的遗传连续性,并对中华民族的形成有着重要的遗传贡献。比如,各地汉族人群有57%-92%的遗传成分来自于以仰韶村遗址古人为代表的新石器时代黄河中游地区古代人群,而这一比例在西藏藏族人群中也高达70%-80%,华南地区的苗瑶和壮侗语人群也有大量血统来自新石器时代黄河中游地区古人。

▲▲▲仰韶村遗址样本与东亚地区古今人群的主成分分析图
•
先前的研究表明,龙山文化的河南禹州瓦店、漯河郝家台和淮阳平粮台遗址的古代人群遗传同质,并且相比于黄河中游地区的仰韶文化人群,瓦店、郝家台和平粮台遗址的人群受到来自东亚南方的遗传影响。然而,研究人员发现这一南方基因流并没有影响到邻近的、同时期的仰韶村遗址龙山文化古人。最后,研究发现一些来自仰韶村遗址的龙山文化个体受到了古东北亚相关的基因流影响。仰韶村遗址龙山时期人群既可以被模拟为仰韶文化相关血统和古东北亚血统的混合,又与陕北石峁遗址人群相对遗传同质。古东北亚相关成分在黄河中游地区的出现可能与龙山时期河套地区对黄河流域的文化影响相关。此外,部分仰韶村遗址龙山文化个体和1个来自瓦店遗址的个体可以被模拟为仰韶文化人群的直接后代。目前尚不清楚为什么这些个体没有像其他同时期个体那样受到东亚南部和古东北亚相关的基因流的影响。

▲▲▲仰韶村遗址样本的遗传组成分析图
•
该研究填补了新石器时代黄河中游人群古基因组数据的部分空白,对理解黄河中游新石器时代人群历史和中华文明的起源和传播有着重要作用。
•
河南省文物考古研究院李世伟、厦门大学生命科学学院博士生王睿和马昊为论文的共同第一作者,河南省文物考古研究院魏兴涛研究员和厦门大学人类学研究所王传超教授为论文的共同通讯作者。该研究得到了河南省哲学社会科学规划项目、河南省兴文化工程研究项目、国家杰出青年基金项目、国家社科基金重大项目、国家自然科学基金面上项目等的资助。
•
仰韶村论文中最核心、重要的结论部分▼▼▼
Next, we focused on elucidating the genetic relationship between YR_Yangshaocun_Longshan
and YR_Longshan_Published.
As shown in PCA, we quantified the genetic heterogeneity between YR_Yangshaocun_Longshan and YR_Longshan_Published, which could be explained by the significant positive values of f4 (Yoruba, Taiwan_Hanben_IA;
YR_Yangshaocun_Longshan, YR_Longshan_Published) and the significant negative values of f4
(Yoruba, ancient Tibetans/Ancient Northeast Asians-related ancestry;
YR_Yangshaocun_Longshan, YR_Longshan_Published) based on the 1240k dataset.
Pairwise qpWave analysis based on the 1240k dataset also rejected the model in which YR_Longshan_Published was genetically cladal with YR_Yangshaocun_Longshan.
This finding reflected the population structure within Late Longshan culture-related people.
The top signals in outgroup-f3 statistics in the form f3 (Yangshao, X; Yoruba) based on the 1240k dataset suggested the strong genetic drift sharing between Yangshao and Longshan culture-related populations.
It highlighted the genetic continuity of
Yangshao-related ancestry in middle YR. We further used Yangshao as one possible genetic source to model the ancestry composition of Longshan-related populations. YR_Longshan_Published themselves required the second source related to southern East Asians (~10%, represented by Taiwan_Hanben_IA) in the qpAdm modeling.
Pairwise qpWave modeling based on the base outgroup set rejected the model in which YR_Yangshaocun_Longshan was genetically cladal with Yangshao.
F4 symmetry tests suggested that
YR_Yangshaocun_Longshan shared extra affinity with ancient Tibetans and Ancient Northeast Asians (ANA)-related ancestry compared with Yangshao.
Ancient Tibetan-related lineage derived its genome from YR, ANA, and ~10% unsampled deeply diverged
lineages, as documented in Ref. .
Hence, we chose ANA-related people as the possible second ancestry source for YR_Yangshaocun_Longshan.
The qpAdm modeling suggested that
YR_Yangshaocun_Longshan could be successfully modeled as ANA (~11%–17%) and Yangshao(~83%-89%)-related ancestry.
Moreover, we found that two of
seven individuals from YR_Yangshaocun_Longshan (i.e., K1-2 and K1-6) shared slightly more
alleles (i.e., 2<|Z-scores|<3) with ANA-related people than Yangshao and could be best modeled as the 2-way Yangshao+ANA; other YR_Yangshaocun_Longshan individuals were well-fitted by 1-way Yangshao.
The individuals labeled as “YR_Longshan_Published_o” were indistinguishable from the Yangshao group, with no need for additional SEA/ANA-related ancestry in qpAdm modeling.
接下来,我们重点阐明仰韶村龙山期人与已发表龙山古人(即郝家台、平粮台、瓦店三组数据,以下简称淮系龙山期人)之间的遗传关系。
如主成分分析(PCA)所示,我们量化了仰韶村龙山期人群和淮河龙山期人群之间的遗传异质性,这可以用f4的显著正值(Yoruba,Taiwan_Hanben_IA;仰韶村龙山期人,淮河龙山期人)和f4的显著负值(Yoruba,古代藏族人/古代东北亚人相关血统;仰韶村龙山期人,淮河龙山期人)来解释。基于1240k的数据集,对仰韶村龙山期人,淮河龙山期人进行了分析。基于1240k数据集的成对qpWave分析完全不支持仰韶村龙山期人与淮河龙山期人具有遗传亲缘关系这一模型。
这一发现反映了龙山时代晚期相关人群的人口结构。基于1240k数据集的f3(汪沟,X;约鲁巴)形式的outgroup-f3统计中的顶部信号表明,汪沟和淮系龙山期人之间存在强烈的共享遗传漂变信号,暗示了基因的连续性。本研究进一步利用汪沟作为可能的遗传来源,对龙山相关种群的祖先组成进行建模。
淮系龙山期人本身在qpAdm模型中需要与东亚南部相关的第二层祖源来源(约10%,以Taiwan_Hanben_IA表示)存在才能被更好的拟合。基于成对qpWave模型否定了仰韶村龙山期人与汪沟在遗传上具有亲缘关系的模型。F4对称性测试表明,与汪沟相比,仰韶村龙山期人与古代藏族、古代东北亚人(即ANA,古老黑龙江)相关的祖先有着额外的遗传亲和力。古代藏族相关谱系的基因组来源于汪沟、ANA和约10%的未采样的(青藏高原幽灵未知古老东亚)深度分化谱系,如参考文献。因此,我们认为ANA相关祖源作为仰韶村龙山期人体内最为可能的第二种祖先成分来源。qpAdm 建模表明,淮系龙山期人可以被成功地建模为 ANA(~11%–17%)和汪沟(~83%-89%)相关祖先的混合。此外,我们发现来自仰韶村龙山期人的 7 个个体中的 2 个(即 K1-2 和 K1-6)与 ANA 相关人群共享的等位基因(即 2<|Z-scores|<3)比汪沟略多!可以被更好地建模为汪沟+ANA的双向混合;其余5个仰韶村龙山期个体均符合由汪沟直接演变而成。被标记为的“YR_Longshan_Published_o”的那个个体(即淮系龙山期人诸多样品的其中一个体)与仰韶类群(即汪沟)没有区别,在qpAdm建模中不需要额外的SEA/ANA相关祖
先。(SEA即东亚南方人群血统)
•
To sum up, we observed the existence of previously unappreciated genetic diversity within people of the Late Longshan cultural period that could be ascribed to two demographic processes.
First, SEA-related populations who relied on sedentary wet rice agriculture migrated northward, resulting in the relatively genetically homogenous populations from Wadian, Haojiatai, and
Pingliangtai sites (i.e.,YR_Longshan_Published).
This was reflected by the increased
reliance on rice farming in the Longshan period; however, this gene flow did not reach the neighboring Yangshaocun site.
Further archaeobotanical study will be needed to directly test whether Yangshaocun also intensified the reliance on rice farming.
It should be noted that ancient DNA associated with rice farmers in the Yangtze River Basin, which is not available
currently, would be necessary to refine the admixture model.
Second, we found that the gene flow
related to ANA contributed to at least some Late Longshan culture-related individuals in Yangshaocun.
This admixture pattern between YR and ANA was observed throughout the Yellow River in the Late Neolithic Age, such as Qijia culture-related from the upper reaches of YR (Upper_YR_LN, Fig. 2b) and Shimao in the Yellow River Curve region (Shimao_LN, Fig. 2b).
It also indicated that ANA once spread further south than previously observed.
YR_Yangshaocun_Longshan could also be modeled as 1-way Shimao_LN (the Late Neolithic population from Shimao site, northern Shannxi), mirroring that the occurrence of ANA ancestry likely coincided with the cultural influence of the Hetao (河套) area on the
Central Plains during the middle and late Longshan period.
Besides, we found one individual
from Wadian whose ancestry was entirely from the Yangshao group.
According to the radiocarbon dating, this individual was ~300 years younger than other Longshan individuals.
Some YR_Yangshaocun_Longshan individuals could also be best described as the direct descendants of Yangshao-related lineage.
However, it is now unclear why these individuals did not receive gene flows from SEA/ANA as other contemporaneous individuals did.
Further sampling from the middle YR may shed more light on it.
综上所述,我们观察到龙山时代晚期人群中存在以前未被重视的遗传多样性,这可以归因于两个人口统计过程。
首先,依赖固定水田农耕(稻作定居农业)的东亚南方血统相关人口向北迁移,导致瓦店、郝家台、平粮台遗址(即YR_Longshan_Published,淮系龙山期人群)所在地区的人群遗传相对同质化,开始向南方血缘偏离,考古上也反映在龙山时期中原地区对稻作农业的依赖增加;然而,该基因流并未到达邻近的仰韶村遗址所代表的豫西地区。需要进一步的植物考古学研究来直接检验仰韶村是否也强化了对水稻种植的依赖。值得注意的是,目前尚无法获得与长江流域稻农相关的古代DNA,对于完善混合模型来说,这是必要的。其次,我们发现与ANA(古东北亚)相关的基因流至少对仰韶村的一些龙山文化晚期相关个体有贡献。汪沟和ANA的这种双向混合模式在新石器时代晚期的整个黄河流域地区中都能观察到,例如黄河上游的齐家文化相关的(Upper_YR_LN,青海民和县距今3900年喇家古人)和黄河几字形地区的石峁文化古人。这表明 ANA 向更南地区的传播比之前观察到的要深远。仰韶村龙山期人同样可以被模拟为由Shimao_LN(来自陕西北部石峁遗址的新石器时代晚期人群)单向贡献而来,反映了该遗传学事件的发生很可能与考古学上龙山中晚期河套地区对中原地区的文化影响相吻合。
此外,我们还发现了一个来自瓦店遗址的个体,其祖源贡献近乎完全来自仰韶族(即汪沟)。根据放射性碳测年,该个体比其他龙山个体年轻约300岁。同时,一些仰韶村龙山期人群同样可以被模拟为仰韶相关血统(汪沟)的直系后裔,然而,目前还不清楚为什么这些人(龙山时代豫西人群)没有像其他同时代的人群那样受到来自 SEA 或 ANA 的基因流影响。
•
Outgroup-f3 statistics suggested the genetic affinity between Neolithic YR-related ancestry and present-day East Asians, especially Han Chinese and Tibeto-Burman-speaking populations. Here, we reanalyzed the available SNP array data of ethnically and linguistically
diverse populations in China and quantified the genetic contribution of middle YR-related ancestry to the present-day Chinese people.
Late Longshan culture-related populations were either the mixture between Yangshao and other lineages (i.e., YR_Longshan_Published was the mixture between Yangshao and southern East Asians; Yangshaocun_Longshan population was the mixture between Yangshao and ANA) or the direct descendants of Yangshao (i.e., YR_Longshan_Published_o and some individuals from YR_Yangshaocun_Longshan).
Therefore, to our knowledge, Yangshao-related ancestry could be the oldest available genomes and the
relatively unadmixed ancestry form in the middle Yellow River. Here, we applied the Yangshao group in the middle YR as one of the sources for present-day Chinese using qpAdm.
Fitting present-day Chinese ethnic groups into the qpAdm modeling, we found that Neolithic Yangshao culture-related lineage significantly impacted the genetic landscape of today’s China.
These models fitted well even when other northern East Asian lineages, such as Shandong Hunter-gatherers, were included in the outgroup set, suggesting the target group could be adequately explained without the contribution of these groups.
Previous studies reported that Han Chinese could be modeled as the mixture between YS and Southern East Asian-related ancestry and formed a geographically structured genetic cline.
Nearly all ethnolinguistic groups in southern China speaking Tai-Kadai (TK), Hmong-Mien (HM), Austronesian (AN), and Austro-Asiatic (AA) were compatible with a two-way mixture of YS and
its corresponding linguistic proxy (i.e., population with limited admixture signals from other lineages).
We here choose Li as the proto-TK proxy, Hmong as the proto-HM proxy, Htin as the proto-AA proxy, and Amis as the proto-AN proxy).
Tungusic- and Mongolic-speaking populations in northeast China could be modeled as a mixture among ANA, YS, and Western Eurasians.
Highland Tibetans maintained a 5,100 years ago Zongri-related genetic profile composed of YS, ANA, and unsampled deeply diverged lineage; lowland Tibetans from southwest China received additional southern Chinese-related gene flow.
In Xinjiang, Turkic-speaking Uyghur and Kyrgyz could be modeled as a mixture between ANA, YS, and Western Eurasians. Hexi Corridor indigenous ethnic groups harbored a high proportion of YS and limited Western Eurasian-related
ancestry.
These results suggested that Yangshao-related lineage was one of the most important biological roots of present-day Chinese people, highlighting the importance of YR in the genetic
formation of the Chinese nation.
外群f3统计数据表明,新石器时代YR相关祖先(以下简称类汪沟祖源)与当今东亚人,特别是汉族和藏缅语人群之间存在极其密切的遗传关系。在这里,我们重新分析了种族和语言方面的可用SNP阵列数据并量化了汪沟相关祖先对当今中国人的遗传贡献。龙山文化晚期相关人群,要么是汪沟与其他谱系的混合(淮系龙山期人是汪沟与某种南东亚人的
混合;仰韶村龙山期人是汪沟与古东北亚的混合),要么是仰韶文化的直系后裔(即
YR_Longshan_Published_o瓦店个体和来自仰韶村龙山期的部分人)。因此,据我们所知,
类汪沟祖源,极有可能是黄河中游最古老的基因组和相对单一、未混合的祖先形式。本文以黄河中下游的仰韶群作为现代汉语的起源地之一。将今天的中国民族纳入qpAdm模型,我们发现新石器时代仰韶文化相关的谱系(汪沟祖源)对今天中国的遗传景观产生了重大影响。即使将山东狩猎采集者(即古山东人群,博山/小高/小荆山)等其他东北亚方血统纳入外群,该模型(类汪沟的现代中国人群的贡献)依然不受影响,这表明即使没有这些群体(古山东/后李)的贡献依然可以充分解释目标群体。之前的研究表明,汉族人可以被建模为仰韶(汪沟)和东亚南相关祖先的混合体,并形成了一个地理结构的遗传谱系。中国南方几乎所有讲仡台语(TK)、苗瑶语(HM)、南岛语(AN)和南亚语(AA)的语言民族群体都与仰韶(汪沟)及其相应语言代理的双向混合兼容(即来自其他谱
系的混合信号有限的群体)。我们在这里选择黎族血统作为原始仡台语人群的遗传代表,苗族作为原始苗瑶遗传代表,Htin(老挝山区猎民)作为原始南亚遗传代表,阿美族作为原始AN代表。中国东北地区讲通古斯语和蒙古语的人口可以被建模为ANA、汪沟和西欧亚人的混合体。5100年前,高原藏族人保持着由汪沟、ANA和未取样的深度分化谱系(高原未知古老东亚)组成的宗日相关遗传谱;来自中国西南部的低地藏人获得了额外的与中国南方相关的基因流。在新疆,讲突厥语的维吾尔语和吉尔吉斯语可以被建模为ANA、汪沟和西欧亚洲人的混合体。河西走廊土著民族拥有较高比例的汪沟和极其有限的西欧亚相关祖先。这些结果表明,仰韶相关谱系(类汪沟祖源)是当今中国人最重要的生物学根源之一,,凸显了仰韶人在中华民族遗传形成中的重要性。




仰韶村遗址龙山时代人群这种分析结果,有两种可能的解释 :
1,豫西地区龙山期人群,依旧保持着当地仰韶期人群的遗传面貌,这是一种当地既有的、延续至少数千年历史的常染遗传格局。本身就比郑州附近的人群(汪沟)更北。
2,豫西地区龙山期人群,原本血缘格局同郑州的人群(汪沟)差不多,但遭受晋陕北部龙山时代人群(黄土高原山地居民)的遗传冲刷,高地龙山人在大肆扩张,带来了更多的遗传多样性,改变了豫西既有的常染格局。。
•
两种解释,可能性都不小。无论哪种,都以无可辩驳的事实证明,整个龙山时代(乃至更早的仰韶时代)豫西当地并未遭受以山东狩猎采集者为代表的海岱细石器常染(古山东人群)的显著影响,而是延续了当地自身的传统(或更北黄土高原移民的传统)。而山东地区的常染面貌出现了严重的断层,原先的古山东人群(类后李),被来自苏皖地区的新移民类汪沟祖源成分逐步替换。同时所谓的「良渚>石峁」一类的远程移民更是不切实际,石峁类人群的遗传格局,甚至可能比仰韶还要北。
按照仰韶村论文最终结论,龙山时代淮系人口的最佳拟合方案是大量汪沟祖源与少量南东亚祖源的2-way model,结合先前的西夏侯、焦家等可知,与河南荥阳市距今五千年出头的汪沟人关联性极强的一支北方东亚血缘大类(泛汪沟祖源)所产生的各种遗传亚群在新石器时代同时对西夏侯、仰韶村、红山、尉迟寺等诸多古今人口同时产生了堪称颠覆性的遗传贡献,而这种贡献的力度绝对是其它祖源(如古山东)完全没法比的。仰韶村所在的豫西乃至整个晋陕豫地区在仰韶时代的血缘主干,有可能仅仅这种血缘大类扩张的一个终端,事实上,自称大汶口论文、焦家论文陆续出炉后,很多事情已经昭然若揭。并不存在什么遗传学上的古黄河成分,硬要说也只有类汪沟成分,但这是一个很大的概念,类汪沟可能有N种,不同的类汪沟成分,在长江以北不同地区遍地开花,有的在山东,有的在河南,有的在山西河北,但最终毫无疑问的,只有龙山时代淮河流域的一支或几支明显南方化的类汪沟常染人群取得了繁衍上的成功,他们在龙山时代及之后对整个北方各地的移民填充,逐渐消弥了各地的常染不均性,拉进了整个北方内部的遗传联系,是奠定北方汉族祖源真正的最核心的贡献者。虽然仰韶村论文认为,这是一种纯粹的、原生的黄河中游古老祖先成分,但其仍然有可能在更为南方(如江淮丘陵/大别山)的区域起源。
与现代北方汉族祖源最为密切相关的古代人口枢纽其实恰恰是龙山时代豫皖平原地区(彼时刚好因为干冷化湿地消失大量成陆形成耕地)以郝家台、平粮台、瓦店为代表的一批淮系人口(及当地后续的王油坊/造律台/新砦/王湾3等),并非古山东人群,这部分人口与古山东人的差距是非常明显的。这批淮系人口的血缘主体同时与汪沟、仰韶村、西夏侯存在更亲近的遗传关系(古山东除外)。目前暂时无法明确淮系人口的“类汪沟”主干成分究竟是来源于鲁南汶泗地区向外扩张,还是当地苏皖北部群体既有的一类土著类汪沟成分(后者可能性更大)。即使该类祖源(有人称之为尉迟寺成分)系鲁南汶泗地区西夏侯类人群向豫东皖北扩张后产生的,西夏侯类祖源也是更早时期从苏皖迁入山东的移民成果,无论如何追溯,不不直接源起于山东本地,这是确凿无疑的。



共获得6个父系遗传标记和10个线粒体遗传标记,3例C南支,2例O系和1例N1a,测序结果平平无奇▼▼▼



龙山时代(时间概念)是严文明老先生于1981年提出的,认为距今5000~4000年,囊括了大汶口晚期和龙山文化期,最新年代标准已经修正为距今4700~3800年。
雄风同志尽快出来走两步,解释解释,你口中的大汶口中晚期考古学因素陆续“强势扩张”,西进的同时,为何遗传却没有同步扩张???你的古山东成分哪去了?
全文半个字没提到山东狩猎采集者(博山等)对仰韶人群存在显著替换,越来越多的迹象表明,所谓的大汶口文化的对外影响力,很可能仅仅是以文化因素/技术传播为主的,并未在遗传层面有太大的影响,相反,山东一直在被皖北地区人群遗传影响。
根据原文构拟,仰韶村龙山期人群的常染可以模拟为由83-89%汪沟成分和11-17%古东北亚成分(ANA)的混合,几乎不含南方稻作农人血统。
仰韶村居民,大部分祖源成分都与五年前发表的汪沟古人高度一致,呈现出强烈的类汪沟常染格局,而且还在此基础上额外与古藏缅人群、古东北亚人群存在遗传亲和力,朝着更西、更北偏移,她就是不往东偏。
注意,这是豫西地区,年代不是六七千年,而是相当晚近的四千余年,已经处于龙山时代晚期,即将踏入夏纪年,在淮河流域的人口开始广泛南方化(稻农北上)的大背景下,豫西当地居民的常染水平仍能北得令人难以置信,实在是太可怕了。。。
•
以下是原文章介绍▼▼▼

近日,河南省文物考古研究院、厦门大学等单位在《Science Bulletin》(《科学通报》英文版)上在线发表了题为 “Ancient genomic time transect unravels the population dynamics of Neolithic middle Yellow River farmers” 的研究论文,通过对仰韶文化命名地——仰韶村遗址的仰韶文化和龙山文化时期古人基因组进行研究,揭示黄河中游地区新石器时代人群动态历史。
•
1921年,中国政府矿政顾问瑞典地质学家安特生和中国地质学家袁复礼等人,对中国河南省渑池县的仰韶村遗址进行了第一次发掘,发现大量新石器时代文化遗存,将其命名为仰韶文化,标志着中国现代考古学的诞生。2021年,O近O总O记致信祝贺仰韶文化发现和中国现代考古学诞生100周年。尽管仰韶村遗址在中国考古学史和仰韶文化研究史上占据十分重要的地位,但我们对仰韶村遗址古代人群的来源历史及其对中华民族的贡献知之甚少。仰韶村遗址的古人遗传学数据研究将有助于我们理解新石器时代黄河中游地区的人群历史。河南省文物考古研究院和厦门大学的研究人员取样提取和测序了仰韶村遗址的古人DNA,首次发表了仰韶村遗址8个属于新石器时代仰韶文化和龙山文化的个体的古基因组数据。研究发现仰韶村遗址仰韶文化与龙山文化相关人群之间具有高度的遗传连续性,并对中华民族的形成有着重要的遗传贡献。比如,各地汉族人群有57%-92%的遗传成分来自于以仰韶村遗址古人为代表的新石器时代黄河中游地区古代人群,而这一比例在西藏藏族人群中也高达70%-80%,华南地区的苗瑶和壮侗语人群也有大量血统来自新石器时代黄河中游地区古人。

▲▲▲仰韶村遗址样本与东亚地区古今人群的主成分分析图
•
先前的研究表明,龙山文化的河南禹州瓦店、漯河郝家台和淮阳平粮台遗址的古代人群遗传同质,并且相比于黄河中游地区的仰韶文化人群,瓦店、郝家台和平粮台遗址的人群受到来自东亚南方的遗传影响。然而,研究人员发现这一南方基因流并没有影响到邻近的、同时期的仰韶村遗址龙山文化古人。最后,研究发现一些来自仰韶村遗址的龙山文化个体受到了古东北亚相关的基因流影响。仰韶村遗址龙山时期人群既可以被模拟为仰韶文化相关血统和古东北亚血统的混合,又与陕北石峁遗址人群相对遗传同质。古东北亚相关成分在黄河中游地区的出现可能与龙山时期河套地区对黄河流域的文化影响相关。此外,部分仰韶村遗址龙山文化个体和1个来自瓦店遗址的个体可以被模拟为仰韶文化人群的直接后代。目前尚不清楚为什么这些个体没有像其他同时期个体那样受到东亚南部和古东北亚相关的基因流的影响。

▲▲▲仰韶村遗址样本的遗传组成分析图
•
该研究填补了新石器时代黄河中游人群古基因组数据的部分空白,对理解黄河中游新石器时代人群历史和中华文明的起源和传播有着重要作用。
•
河南省文物考古研究院李世伟、厦门大学生命科学学院博士生王睿和马昊为论文的共同第一作者,河南省文物考古研究院魏兴涛研究员和厦门大学人类学研究所王传超教授为论文的共同通讯作者。该研究得到了河南省哲学社会科学规划项目、河南省兴文化工程研究项目、国家杰出青年基金项目、国家社科基金重大项目、国家自然科学基金面上项目等的资助。
•
仰韶村论文中最核心、重要的结论部分▼▼▼
Next, we focused on elucidating the genetic relationship between YR_Yangshaocun_Longshan
and YR_Longshan_Published.
As shown in PCA, we quantified the genetic heterogeneity between YR_Yangshaocun_Longshan and YR_Longshan_Published, which could be explained by the significant positive values of f4 (Yoruba, Taiwan_Hanben_IA;
YR_Yangshaocun_Longshan, YR_Longshan_Published) and the significant negative values of f4
(Yoruba, ancient Tibetans/Ancient Northeast Asians-related ancestry;
YR_Yangshaocun_Longshan, YR_Longshan_Published) based on the 1240k dataset.
Pairwise qpWave analysis based on the 1240k dataset also rejected the model in which YR_Longshan_Published was genetically cladal with YR_Yangshaocun_Longshan.
This finding reflected the population structure within Late Longshan culture-related people.
The top signals in outgroup-f3 statistics in the form f3 (Yangshao, X; Yoruba) based on the 1240k dataset suggested the strong genetic drift sharing between Yangshao and Longshan culture-related populations.
It highlighted the genetic continuity of
Yangshao-related ancestry in middle YR. We further used Yangshao as one possible genetic source to model the ancestry composition of Longshan-related populations. YR_Longshan_Published themselves required the second source related to southern East Asians (~10%, represented by Taiwan_Hanben_IA) in the qpAdm modeling.
Pairwise qpWave modeling based on the base outgroup set rejected the model in which YR_Yangshaocun_Longshan was genetically cladal with Yangshao.
F4 symmetry tests suggested that
YR_Yangshaocun_Longshan shared extra affinity with ancient Tibetans and Ancient Northeast Asians (ANA)-related ancestry compared with Yangshao.
Ancient Tibetan-related lineage derived its genome from YR, ANA, and ~10% unsampled deeply diverged
lineages, as documented in Ref. .
Hence, we chose ANA-related people as the possible second ancestry source for YR_Yangshaocun_Longshan.
The qpAdm modeling suggested that
YR_Yangshaocun_Longshan could be successfully modeled as ANA (~11%–17%) and Yangshao(~83%-89%)-related ancestry.
Moreover, we found that two of
seven individuals from YR_Yangshaocun_Longshan (i.e., K1-2 and K1-6) shared slightly more
alleles (i.e., 2<|Z-scores|<3) with ANA-related people than Yangshao and could be best modeled as the 2-way Yangshao+ANA; other YR_Yangshaocun_Longshan individuals were well-fitted by 1-way Yangshao.
The individuals labeled as “YR_Longshan_Published_o” were indistinguishable from the Yangshao group, with no need for additional SEA/ANA-related ancestry in qpAdm modeling.
接下来,我们重点阐明仰韶村龙山期人与已发表龙山古人(即郝家台、平粮台、瓦店三组数据,以下简称淮系龙山期人)之间的遗传关系。
如主成分分析(PCA)所示,我们量化了仰韶村龙山期人群和淮河龙山期人群之间的遗传异质性,这可以用f4的显著正值(Yoruba,Taiwan_Hanben_IA;仰韶村龙山期人,淮河龙山期人)和f4的显著负值(Yoruba,古代藏族人/古代东北亚人相关血统;仰韶村龙山期人,淮河龙山期人)来解释。基于1240k的数据集,对仰韶村龙山期人,淮河龙山期人进行了分析。基于1240k数据集的成对qpWave分析完全不支持仰韶村龙山期人与淮河龙山期人具有遗传亲缘关系这一模型。
这一发现反映了龙山时代晚期相关人群的人口结构。基于1240k数据集的f3(汪沟,X;约鲁巴)形式的outgroup-f3统计中的顶部信号表明,汪沟和淮系龙山期人之间存在强烈的共享遗传漂变信号,暗示了基因的连续性。本研究进一步利用汪沟作为可能的遗传来源,对龙山相关种群的祖先组成进行建模。
淮系龙山期人本身在qpAdm模型中需要与东亚南部相关的第二层祖源来源(约10%,以Taiwan_Hanben_IA表示)存在才能被更好的拟合。基于成对qpWave模型否定了仰韶村龙山期人与汪沟在遗传上具有亲缘关系的模型。F4对称性测试表明,与汪沟相比,仰韶村龙山期人与古代藏族、古代东北亚人(即ANA,古老黑龙江)相关的祖先有着额外的遗传亲和力。古代藏族相关谱系的基因组来源于汪沟、ANA和约10%的未采样的(青藏高原幽灵未知古老东亚)深度分化谱系,如参考文献。因此,我们认为ANA相关祖源作为仰韶村龙山期人体内最为可能的第二种祖先成分来源。qpAdm 建模表明,淮系龙山期人可以被成功地建模为 ANA(~11%–17%)和汪沟(~83%-89%)相关祖先的混合。此外,我们发现来自仰韶村龙山期人的 7 个个体中的 2 个(即 K1-2 和 K1-6)与 ANA 相关人群共享的等位基因(即 2<|Z-scores|<3)比汪沟略多!可以被更好地建模为汪沟+ANA的双向混合;其余5个仰韶村龙山期个体均符合由汪沟直接演变而成。被标记为的“YR_Longshan_Published_o”的那个个体(即淮系龙山期人诸多样品的其中一个体)与仰韶类群(即汪沟)没有区别,在qpAdm建模中不需要额外的SEA/ANA相关祖
先。(SEA即东亚南方人群血统)
•
To sum up, we observed the existence of previously unappreciated genetic diversity within people of the Late Longshan cultural period that could be ascribed to two demographic processes.
First, SEA-related populations who relied on sedentary wet rice agriculture migrated northward, resulting in the relatively genetically homogenous populations from Wadian, Haojiatai, and
Pingliangtai sites (i.e.,YR_Longshan_Published).
This was reflected by the increased
reliance on rice farming in the Longshan period; however, this gene flow did not reach the neighboring Yangshaocun site.
Further archaeobotanical study will be needed to directly test whether Yangshaocun also intensified the reliance on rice farming.
It should be noted that ancient DNA associated with rice farmers in the Yangtze River Basin, which is not available
currently, would be necessary to refine the admixture model.
Second, we found that the gene flow
related to ANA contributed to at least some Late Longshan culture-related individuals in Yangshaocun.
This admixture pattern between YR and ANA was observed throughout the Yellow River in the Late Neolithic Age, such as Qijia culture-related from the upper reaches of YR (Upper_YR_LN, Fig. 2b) and Shimao in the Yellow River Curve region (Shimao_LN, Fig. 2b).
It also indicated that ANA once spread further south than previously observed.
YR_Yangshaocun_Longshan could also be modeled as 1-way Shimao_LN (the Late Neolithic population from Shimao site, northern Shannxi), mirroring that the occurrence of ANA ancestry likely coincided with the cultural influence of the Hetao (河套) area on the
Central Plains during the middle and late Longshan period.
Besides, we found one individual
from Wadian whose ancestry was entirely from the Yangshao group.
According to the radiocarbon dating, this individual was ~300 years younger than other Longshan individuals.
Some YR_Yangshaocun_Longshan individuals could also be best described as the direct descendants of Yangshao-related lineage.
However, it is now unclear why these individuals did not receive gene flows from SEA/ANA as other contemporaneous individuals did.
Further sampling from the middle YR may shed more light on it.
综上所述,我们观察到龙山时代晚期人群中存在以前未被重视的遗传多样性,这可以归因于两个人口统计过程。
首先,依赖固定水田农耕(稻作定居农业)的东亚南方血统相关人口向北迁移,导致瓦店、郝家台、平粮台遗址(即YR_Longshan_Published,淮系龙山期人群)所在地区的人群遗传相对同质化,开始向南方血缘偏离,考古上也反映在龙山时期中原地区对稻作农业的依赖增加;然而,该基因流并未到达邻近的仰韶村遗址所代表的豫西地区。需要进一步的植物考古学研究来直接检验仰韶村是否也强化了对水稻种植的依赖。值得注意的是,目前尚无法获得与长江流域稻农相关的古代DNA,对于完善混合模型来说,这是必要的。其次,我们发现与ANA(古东北亚)相关的基因流至少对仰韶村的一些龙山文化晚期相关个体有贡献。汪沟和ANA的这种双向混合模式在新石器时代晚期的整个黄河流域地区中都能观察到,例如黄河上游的齐家文化相关的(Upper_YR_LN,青海民和县距今3900年喇家古人)和黄河几字形地区的石峁文化古人。这表明 ANA 向更南地区的传播比之前观察到的要深远。仰韶村龙山期人同样可以被模拟为由Shimao_LN(来自陕西北部石峁遗址的新石器时代晚期人群)单向贡献而来,反映了该遗传学事件的发生很可能与考古学上龙山中晚期河套地区对中原地区的文化影响相吻合。
此外,我们还发现了一个来自瓦店遗址的个体,其祖源贡献近乎完全来自仰韶族(即汪沟)。根据放射性碳测年,该个体比其他龙山个体年轻约300岁。同时,一些仰韶村龙山期人群同样可以被模拟为仰韶相关血统(汪沟)的直系后裔,然而,目前还不清楚为什么这些人(龙山时代豫西人群)没有像其他同时代的人群那样受到来自 SEA 或 ANA 的基因流影响。
•
Outgroup-f3 statistics suggested the genetic affinity between Neolithic YR-related ancestry and present-day East Asians, especially Han Chinese and Tibeto-Burman-speaking populations. Here, we reanalyzed the available SNP array data of ethnically and linguistically
diverse populations in China and quantified the genetic contribution of middle YR-related ancestry to the present-day Chinese people.
Late Longshan culture-related populations were either the mixture between Yangshao and other lineages (i.e., YR_Longshan_Published was the mixture between Yangshao and southern East Asians; Yangshaocun_Longshan population was the mixture between Yangshao and ANA) or the direct descendants of Yangshao (i.e., YR_Longshan_Published_o and some individuals from YR_Yangshaocun_Longshan).
Therefore, to our knowledge, Yangshao-related ancestry could be the oldest available genomes and the
relatively unadmixed ancestry form in the middle Yellow River. Here, we applied the Yangshao group in the middle YR as one of the sources for present-day Chinese using qpAdm.
Fitting present-day Chinese ethnic groups into the qpAdm modeling, we found that Neolithic Yangshao culture-related lineage significantly impacted the genetic landscape of today’s China.
These models fitted well even when other northern East Asian lineages, such as Shandong Hunter-gatherers, were included in the outgroup set, suggesting the target group could be adequately explained without the contribution of these groups.
Previous studies reported that Han Chinese could be modeled as the mixture between YS and Southern East Asian-related ancestry and formed a geographically structured genetic cline.
Nearly all ethnolinguistic groups in southern China speaking Tai-Kadai (TK), Hmong-Mien (HM), Austronesian (AN), and Austro-Asiatic (AA) were compatible with a two-way mixture of YS and
its corresponding linguistic proxy (i.e., population with limited admixture signals from other lineages).
We here choose Li as the proto-TK proxy, Hmong as the proto-HM proxy, Htin as the proto-AA proxy, and Amis as the proto-AN proxy).
Tungusic- and Mongolic-speaking populations in northeast China could be modeled as a mixture among ANA, YS, and Western Eurasians.
Highland Tibetans maintained a 5,100 years ago Zongri-related genetic profile composed of YS, ANA, and unsampled deeply diverged lineage; lowland Tibetans from southwest China received additional southern Chinese-related gene flow.
In Xinjiang, Turkic-speaking Uyghur and Kyrgyz could be modeled as a mixture between ANA, YS, and Western Eurasians. Hexi Corridor indigenous ethnic groups harbored a high proportion of YS and limited Western Eurasian-related
ancestry.
These results suggested that Yangshao-related lineage was one of the most important biological roots of present-day Chinese people, highlighting the importance of YR in the genetic
formation of the Chinese nation.
外群f3统计数据表明,新石器时代YR相关祖先(以下简称类汪沟祖源)与当今东亚人,特别是汉族和藏缅语人群之间存在极其密切的遗传关系。在这里,我们重新分析了种族和语言方面的可用SNP阵列数据并量化了汪沟相关祖先对当今中国人的遗传贡献。龙山文化晚期相关人群,要么是汪沟与其他谱系的混合(淮系龙山期人是汪沟与某种南东亚人的
混合;仰韶村龙山期人是汪沟与古东北亚的混合),要么是仰韶文化的直系后裔(即
YR_Longshan_Published_o瓦店个体和来自仰韶村龙山期的部分人)。因此,据我们所知,
类汪沟祖源,极有可能是黄河中游最古老的基因组和相对单一、未混合的祖先形式。本文以黄河中下游的仰韶群作为现代汉语的起源地之一。将今天的中国民族纳入qpAdm模型,我们发现新石器时代仰韶文化相关的谱系(汪沟祖源)对今天中国的遗传景观产生了重大影响。即使将山东狩猎采集者(即古山东人群,博山/小高/小荆山)等其他东北亚方血统纳入外群,该模型(类汪沟的现代中国人群的贡献)依然不受影响,这表明即使没有这些群体(古山东/后李)的贡献依然可以充分解释目标群体。之前的研究表明,汉族人可以被建模为仰韶(汪沟)和东亚南相关祖先的混合体,并形成了一个地理结构的遗传谱系。中国南方几乎所有讲仡台语(TK)、苗瑶语(HM)、南岛语(AN)和南亚语(AA)的语言民族群体都与仰韶(汪沟)及其相应语言代理的双向混合兼容(即来自其他谱
系的混合信号有限的群体)。我们在这里选择黎族血统作为原始仡台语人群的遗传代表,苗族作为原始苗瑶遗传代表,Htin(老挝山区猎民)作为原始南亚遗传代表,阿美族作为原始AN代表。中国东北地区讲通古斯语和蒙古语的人口可以被建模为ANA、汪沟和西欧亚人的混合体。5100年前,高原藏族人保持着由汪沟、ANA和未取样的深度分化谱系(高原未知古老东亚)组成的宗日相关遗传谱;来自中国西南部的低地藏人获得了额外的与中国南方相关的基因流。在新疆,讲突厥语的维吾尔语和吉尔吉斯语可以被建模为ANA、汪沟和西欧亚洲人的混合体。河西走廊土著民族拥有较高比例的汪沟和极其有限的西欧亚相关祖先。这些结果表明,仰韶相关谱系(类汪沟祖源)是当今中国人最重要的生物学根源之一,,凸显了仰韶人在中华民族遗传形成中的重要性。




仰韶村遗址龙山时代人群这种分析结果,有两种可能的解释 :
1,豫西地区龙山期人群,依旧保持着当地仰韶期人群的遗传面貌,这是一种当地既有的、延续至少数千年历史的常染遗传格局。本身就比郑州附近的人群(汪沟)更北。
2,豫西地区龙山期人群,原本血缘格局同郑州的人群(汪沟)差不多,但遭受晋陕北部龙山时代人群(黄土高原山地居民)的遗传冲刷,高地龙山人在大肆扩张,带来了更多的遗传多样性,改变了豫西既有的常染格局。。
•
两种解释,可能性都不小。无论哪种,都以无可辩驳的事实证明,整个龙山时代(乃至更早的仰韶时代)豫西当地并未遭受以山东狩猎采集者为代表的海岱细石器常染(古山东人群)的显著影响,而是延续了当地自身的传统(或更北黄土高原移民的传统)。而山东地区的常染面貌出现了严重的断层,原先的古山东人群(类后李),被来自苏皖地区的新移民类汪沟祖源成分逐步替换。同时所谓的「良渚>石峁」一类的远程移民更是不切实际,石峁类人群的遗传格局,甚至可能比仰韶还要北。
按照仰韶村论文最终结论,龙山时代淮系人口的最佳拟合方案是大量汪沟祖源与少量南东亚祖源的2-way model,结合先前的西夏侯、焦家等可知,与河南荥阳市距今五千年出头的汪沟人关联性极强的一支北方东亚血缘大类(泛汪沟祖源)所产生的各种遗传亚群在新石器时代同时对西夏侯、仰韶村、红山、尉迟寺等诸多古今人口同时产生了堪称颠覆性的遗传贡献,而这种贡献的力度绝对是其它祖源(如古山东)完全没法比的。仰韶村所在的豫西乃至整个晋陕豫地区在仰韶时代的血缘主干,有可能仅仅这种血缘大类扩张的一个终端,事实上,自称大汶口论文、焦家论文陆续出炉后,很多事情已经昭然若揭。并不存在什么遗传学上的古黄河成分,硬要说也只有类汪沟成分,但这是一个很大的概念,类汪沟可能有N种,不同的类汪沟成分,在长江以北不同地区遍地开花,有的在山东,有的在河南,有的在山西河北,但最终毫无疑问的,只有龙山时代淮河流域的一支或几支明显南方化的类汪沟常染人群取得了繁衍上的成功,他们在龙山时代及之后对整个北方各地的移民填充,逐渐消弥了各地的常染不均性,拉进了整个北方内部的遗传联系,是奠定北方汉族祖源真正的最核心的贡献者。虽然仰韶村论文认为,这是一种纯粹的、原生的黄河中游古老祖先成分,但其仍然有可能在更为南方(如江淮丘陵/大别山)的区域起源。
与现代北方汉族祖源最为密切相关的古代人口枢纽其实恰恰是龙山时代豫皖平原地区(彼时刚好因为干冷化湿地消失大量成陆形成耕地)以郝家台、平粮台、瓦店为代表的一批淮系人口(及当地后续的王油坊/造律台/新砦/王湾3等),并非古山东人群,这部分人口与古山东人的差距是非常明显的。这批淮系人口的血缘主体同时与汪沟、仰韶村、西夏侯存在更亲近的遗传关系(古山东除外)。目前暂时无法明确淮系人口的“类汪沟”主干成分究竟是来源于鲁南汶泗地区向外扩张,还是当地苏皖北部群体既有的一类土著类汪沟成分(后者可能性更大)。即使该类祖源(有人称之为尉迟寺成分)系鲁南汶泗地区西夏侯类人群向豫东皖北扩张后产生的,西夏侯类祖源也是更早时期从苏皖迁入山东的移民成果,无论如何追溯,不不直接源起于山东本地,这是确凿无疑的。