Ayta Negritos: Highest Levels of Denisovan DNA

new

Admin
Administrator

Posts: 81,617

Ayta Negritos: Highest Levels of Denisovan DNA Aug 21, 2021 5:44:19 GMT

Quote

Post by Admin on Aug 21, 2021 5:44:19 GMT

Materials and Methods
Ethics Statements, Sample Collection, and Genotyping
This study was approved by the Ministry of Health Malaysia under National Medical Research Registry MNDR ID #09—23-3913, JAKOA (Department of Orang Asli Development, Government of Malaysia) and Monash University Human Research Ethics Committee.

Following consultation with JAKOA officers in the various districts in different states, courtesy visits were made to OA community elders and the rationale of the study and the procedure of sample collection explained. Once they had agreed and informed their communities, field visits were carried out. Individuals who provided informed consent and also answered questionnaires were included.

Peripheral blood samples were collected from 169 individuals belonging to Negrito (Jehai, Bateq, Kintaq, and Mendriq subgroups), Senoi (MahMeri and CheWong subgroup), and Proto-Malay (Seletar, Jakun, and Temuan subgroups) groups (fig. 1). Genotyping was performed using Illumina Human Omni 2.5 array (Illumina Inc., San Diego, CA).

Fig. 1.—
Geographical location of Orang Asli communities recruited in this study.

Quality Control and Data Integration
Quality controls were applied to the data obtained from each OA community separately to exclude problematic samples and single nucleotide polymorphisms (SNPs). All SNPs that failed the Hardy–Weinberg exact (HWE) test (P < 10−6) and displayed missing rates >0.05 across all samples in each population were removed. Additionally, samples with call rate <0.99 were excluded. Gender concordance was examined using PLINK v1.07 (Purcell et al. 2007) and samples with inconsistency between genotype results and questionnaire-reported sex were excluded. In order to avoid analysis of close relatives, unknown relatedness was measured between all pairs of individuals within each population using PLINK’s (v1.07) Identity-by-Descent estimation, PI_Hat. An upper cut-off threshold of 0.375 was set to exclude first-degree relatedness within each population. Finally, a principal component analysis (PCA) using EIGENSOFT v3.0 (Patterson et al. 2006) was performed to remove outliers from each population across first ten eigenvectors. In the final stage, all OA populations were merged into one data set and pruned for SNPs that failed HWE (P < 10−6) test and missing rates more than 0.05 across all samples.

The OA genotype data were merged with data from Human Genome Diversity Project (HGDP) (Li et al. 2008), 89 Malay individuals from Singapore Genome Variation Project (SGVP) (Teo et al. 2009) and Onge and Jarawa Negritos from Andaman islands were genotyped using Illumina Human 1.2M (SNP population data courtesy of P. Majumder and A. Basu). After merging data sets (supplementary table S1, Supplementary Material online), a total of 291,096 overlapping autosomal SNPs remained for downstream analysis.

Population Structure Analysis
PCA was used to identify population structure across indigenous Malaysians. PCA analysis was performed on genotyped data of OA combined with Andamanese Negritos, Oceanians, South and East Asian populations in the HGDP, and Malays from SGVP using EIGENSOFT v3.0. To balance sample sizes across our populations, 30 Malay individuals were randomly sampled from SGVP data set (which contains 89 individuals). SNPs with r2 > 0.5 were pruned out in order to avoid the effects of excessive LD between SNPs. After this pruning a total of 204,426 SNPs remained for analysis. Pairwise Fst distance between populations in same data set were calculated using EIGENSOFT v3.0, and a Neighbor-net tree was constructed by SplitsTree v4 software (Huson and Bryant 2006). ADMIXTURE v1.22, a clustering algorithm, was used on pruned SNPs to estimate the ancestral population clustering (Alexander et al. 2009).

PLINK v1.07 was used to estimate ROH in selected populations. PLINK takes 5,000 kb (50 SNPs) sliding windows across the genome and allows for 1 heterozygous and 5 missing calls in each window. To minimize the effects of LD on ROH, minimum ROH length was set to be 500 kb because it is unusual for LD to extend beyond 500 kb. LD decay for each population was calculated as r2 using PLINK. Pairwise LD between all possible SNPs was calculated and mean LD was measured in bins of 5 kb.

TreeMix v1.12 (Pickrell and Pritchard 2012) was used to explore the population relationships and migration events. Same data set described above was used to estimate the Maximum Likelihood tree with Yoruba as outgroup. We used blocks of 200 SNPs (-k 200) to account for LD and migration edges added sequentially until the model explained 99% of variances. We estimated the D statistics using ADMIXTOOLS (Patterson et al. 2012) to examine gene flow between OAs and surrounding populations. Divergence time between OA and EA was estimated using 399,971 shared SNPs between our data and HapMap 3 (The International HapMap 2005). Effective population size (Ne) and divergence time between OAs and Yoruba in Ibadan (YRI), Han Chinese in Beijing (CHB), and Japanese in Tokyo (JPT) samples were estimated according to the method suggested by McEvoy et al. (2011). To estimate LD, pairwise LD was calculated as r2 using PLINK v1.07. In order to minimize the effects of small sample size, all individuals were pooled together in their respective OA groups. Admixture time between OAs and EA was estimated by rolloff package using 399,971 SNPs by HapMap3 and OAs.

Admin
Administrator

Posts: 81,617

Ayta Negritos: Highest Levels of Denisovan DNA Aug 22, 2021 20:00:18 GMT

Quote

Post by Admin on Aug 22, 2021 20:00:18 GMT

Results
To understand population structure across Negritos, other OA subgroups, and their relationship with neighboring populations in Asia and Oceania, a PCA was performed (fig. 2 and supplementary fig. S1, Supplementary Material online). As presented in figure 2A, the first component, which captures 32% of total variation, clearly distinguishes South Asian populations from those in the East. From PC2, the Onge and Jarawa, both Negrito subgroups, clustered together and were distinct from other populations. However, they appeared closest to Papuans and Melanesians. The Malaysian Negrito subgroups, while clustering closer to East Asians, showed a tendency toward other Negrito subgroups in Oceania and Andaman islands. The rest of OAs such as Senoi and Proto-Malays as well as Singaporean Malays were located between Malaysian Negritos and East Asian clusters indicating that these groups might be admixed between these two populations. However, both Senoi and Proto-Malay groups lay closer to East Asians on PC4 suggesting that all these populations may have a common origin.

Fig. 2.—

Like PCA analysis, the results of Neighbor-net tree showed that OAs are closest to EA populations. As evident in supplementary figure S2, Supplementary Material online, all four subgroups of Negritos formed a clade, while Senoi and Proto-Malay were positioned at various points between these two clades. The long branches observed in Bateq, Jehai, Kintaq, CheWong, Seletar, and MahMeri suggest strong drift in each of these populations. Interestingly, Seletar located between Malaysian Negritos and Oceanians. The tree also indicated genetic affinity between Andamanese and Oceanians.

In order to determine critical ancestral components that may have shaped the genetic architecture among the OAs, we applied ADMIXTURE analysis. The results of ADMIXTURE from K = 2 to K = 12 are shown in figure 3. Each individual is represented as a vertical bar and their corresponding ancestry components are shown by different colors. Different colors indicate different ancestry lineages. As presented, K = 2 separated Central-South Asia (red) and EA (yellow) and the latter appears to be the major component in all OA groups. From K = 3, Andamanese component (pink) appeared. This component also presented considerably in Oceanians and in lesser extent in Malaysian Negritos. At higher K = 4 and K = 5, Negrito (dark green) and Oceanian (dark blue) components appeared respectively. The best model which had the lowest cross validation error suggests nine major ancestral groups which gave rise to the 40 distinct populations included in our study. At K = 9, all Negrito subgroups showed similar ancestral patterns. However, we observed small portions of other ancestral components (shown in yellow and purple) in some Negrito individuals (especially Mendriqs).

Fig. 3.—
ADMIXTURE analysis of Orang Asli, Andamanese, South Asian, and East Asian ethnic groups from HGDP and Singaporean Malay.

Results of ADMIXTURE at K = 9 also showed that two Senoi subgroups had different ancestral patterns. The purple colored ancestry component is highest in MahMeri, but also present in the Proto-Malay and Malay. The CheWongs appear to have MahMeri, Negrito, and East Asian components. At K = 11, CheWong appeared distinct.

Different patterns of ancestry were identified in Proto-Malays. At K = 9, Jakun and Temuan had similar ancestral components, but there was a unique substantial component (shown in light blue) only present in the Seletar from K = 6. The ADMIXTURE results further support the uniqueness of OAs.

To understand the relationship between our populations and examine the gene flow between them, we used TreeMix (fig. 4 and supplementary fig. S3, Supplementary Material online). Using Yoruba as root, the graph that best fits our data (99.4% of variances) inferred six migration events. The tree topology was consistent with geographical distribution of populations and with previously shown Neighbor-net tree. Andamanese and Oceanians grouped together in a deep clade, while all OA groups formed a distinct cluster. Focusing on migration events, a migration (migration weight 0.37) directed from root Onge and Jarawa toward Malaysian Negrito root. The resulting tree also highlighted another migration (0.39) from the root of Bateq and Jehai to CheWong.

Admin
Administrator

Posts: 81,617

Ayta Negritos: Highest Levels of Denisovan DNA Aug 22, 2021 22:51:52 GMT

Quote

Post by Admin on Aug 22, 2021 22:51:52 GMT

Fig. 4.—
Treemix tree of Orang Asli subgroups, Negrito groups of Andaman Islands, and South and East Asian populations from HGDP.

To further investigate gene flow between OAs and other populations, we used D statistics (table 1 and supplementary tables S2 and S3, Supplementary Material online). The computed D statistics demonstrated significant gene flow between Andamanese and Malaysian Negritos but there was no significant gene flow detected between Andamanese and other OA groups. This suggests that an earlier gene flow occurred before other OA groups arrived in Peninsular Malaysia. The D statistics supported admixture between different OA groups, as gene flows between Negrito/Senoi, Negrito/ Proto-Malays, and Senoi/Proto-Malays were evident. We also traced admixture in AA-speaking OAs and those of Mainland SEA and Lahu and Dai, ethnic groups from South China.

Table 1
Computed D Statistic Results Showing Gene Flow between Negrito and Other Populations in SEA

Group D Score Z Scorea Group D score Z score
D (Jehai, Yoruba; Han, X)
D (Jehai, Yoruba; Japanese, X)
Temuan −1.16 × 10−02 −10.578 Temuan −1.82 × 10−02 −15.328
Jakun −1.44 × 10−02 −11.125 Jakun −2.10 × 10−02 −15.438
Seletar −2.00 × 10−03 −1.397 Seletar −8.80 × 10−03 −5.923
MahMeri −9.30 × 10−03 −6.931 MahMeri −1.60 × 10−02 −11.455
CheWong −3.47 × 10−02 −21.149 CheWong −4.11 × 10−02 −24.359
Malay −6.00 × 10−04 −0.744 Malay −7.40 × 10−03 −7.999
Cambodian −2.40 × 10−03 −2.464 Cambodian −9.20 × 10−03 −8.482
Lahu −6.10 × 10−03 −5.612 Lahu −1.30 × 10−02 −10.602
Dai −7.80 × 10−03 −8.537 Dai −1.46 × 10−02 −13.701
D (Bateq, Yoruba; Han, X)
D (Bateq, Yoruba; Japanese, X)
Temuan −1.02 × 10−02 −9.277 Temuan −1.56 × 10−02 −12.926
Jakun −1.44 × 10−02 −10.388 Jakun −1.98 × 10−02 −13.604
Seletar −1.80 × 10−03 −1.25 Seletar −7.30 × 10−03 −4.733
MahMeri −7.90 × 10−03 −5.84 MahMeri −1.33 × 10−02 −9.404
CheWong −3.63 × 10−02 −19.949 CheWong −4.14 × 10−02 −22.355
Malay 2.00 × 10−04 0.206 Malay −5.40 × 10−03 −5.496
Cambodian −1.70 × 10−03 −1.635 Cambodian −7.20 × 10−03 −6.351
Lahu −4.60 × 10−03 −4.032 Lahu −1.02 × 10−02 −8.124
Dai −6.70 × 10−03 −7.09 Dai −1.23 × 10−02 −11.212
D (Kintaq, Yoruba; Han, X)
D (Kintaq, Yoruba; Japanese, X)
Temuan −1.02 × 10−02 −9.504 Temuan −1.65 × 10−02 −14.175
Jakun −1.26 × 10−02 −9.693 Jakun −1.88 × 10−02 −13.683
Seletar −1.80 × 10−03 −1.213 Seletar −8.10 × 10−03 −5.271
MahMeri −8.20 × 10−03 −6.342 MahMeri −1.44 × 10−02 −10.695
CheWong −3.40 × 10−02 −20.837 CheWong −4.00 × 10−02 −23.709
Malay 0.00 −0.056 Malay −6.40 × 10−03 −6.985
Cambodian −2.00 × 10−03 −2.089 Cambodian −8.30 × 10−03 −7.879
Lahu −5.20 × 10−03 −4.788 Lahu −1.16 × 10−02 −9.756
Dai −7.20 × 10−03 −7.749 Dai −1.36 × 10−02 −12.934
Open in a separate window
aAbsolute Z score >3 shows significant gene flow between populations.

Focusing on OAs in Malaysia, we determined inheritance of parental genome components, and calculated ROH in all OA groups against Malay from Singapore. Figure 5A shows the distribution of ROH in these populations. As expected, all Negrito groups generally showed long and high ROH compared with other OA groups. This is indicative of small population size or consanguinity. Interestingly, Seletar had the longest ROH among all OA groups which may reflect higher levels of autozygosity.

Fig. 5.—

(A) Runs of homozygosis in Orang Aslis and Malay from SGVP and (B) pattern of linkage disequilibrium decay in Orang Asli groups and SGVP Malay.

To further examine the genetic isolation and admixture between OA groups, we calculated pairwise LD between all autosomal SNPs. LD is the nonrandom association of two SNPs and its decay can be affected by factors like drift, admixture, and inbreeding. Figure 5B shows the LD decay in OA subgroups and Singaporean Malays. LD in all OA groups was markedly higher even for long pairwise SNPs distances.

We estimated the divergence time (T) of OA groups and Africans to be around 67 KYA assuming generation time of 25 years which is a good agreement with other reported estimations of EA and African divergence previously (McEvoy et al. 2011; Pugach et al. 2013). Our results inferred earlier divergence of Negritos from EA in 14–15 KYA which predate those of Senoi (10–11 KYA) and Proto-Malay (8–9 KYA) (table 2). Admixture time estimation between OA groups using “rolloff” showed that the admixture date between Negrito and Senoi to be around 40 generations which was older than Negrito/Proto-Malay and Senoi/Proto-Malay admixture which occurred around 20 generations before the present.

Table 2
Divergence Time (KYA) Estimation between OA Groups and YRI, CHB, and JPT

YRI CHB JPT
Negrito 66.8 14.5 14.6
Senoi 67.5 10 11
Proto-Malay 66.9 8.2 9.2
YRI — 72 72

Admin
Administrator

Posts: 81,617

Ayta Negritos: Highest Levels of Denisovan DNA Aug 23, 2021 1:42:55 GMT

Quote

Post by Admin on Aug 23, 2021 1:42:55 GMT

Discussion
Despite the rich ethnic diversity present in SEA, the region has been underrepresented in large-scale international genome data sets such as HAPMAP and 1000 Genome Project (Lu and Xu 2013). Diverse linguistic, morphological, and anthropological characteristics found in minor ethnic groups of Malaysia, known as OA, offered a promising opportunity to understand the populations of East Asia and SEA.

Our investigation has contributed substantially more data and provided more comprehensive insight into the population structure of diverse indigenous groups and their prehistoric links to other populations in mainland SEA and East Asia. Apparently, the OAs are genetically closer to EA populations compared with those in South Asia or Oceania. However, our results provided evidences supporting genetic affinity between Malaysian and Andamanese Negritos. Our results are entirely consistent with other SNP studies suggesting link between Andamanes, Malaysian Negritos, and Melanesians (Reich et al. 2011; Chaubey and Endicott 2013).

On a finer scale, Malaysia Negrito subgroups were clearly different from EA populations. This distinct pattern may have resulted from genetic drift. It is also conceivable that they had longer periods of isolation from other inhabitants in the region, as indicated by Fst and LD decay. The ancestral component (dark green) “belonging” to Malaysian Negritos was also spread among Southeast Asian and Southern Chinese populations. However, although Negritos predominantly shared this ancestral component, the Mendriq shared more portions of other ancestral components with East Asians and Senoi. This suggests more recent gene flow between them and their neighboring populations, most likely Malays. A similar observation was reported in Jehai, a Negrito subgroup using a less SNP (Jinam et al. 2013).

The Senoi and Proto-Malay were closely related to EA, either because they share relatively recent common ancestors or because of recent gene flow. However, different patterns emerged in Seletar and CheWong. The corresponding ancestral component of Seletar, a subgroup of Proto-Malay, emerged at K = 6 in ADMIXTURE and Neighbor-net tree showed an affinity to the Oceanian. Anthropological information regarding origins of the Seletar is scarce and anecdotal. There is a paucity of information about this community. It is plausible that Seletar might have experienced a recent bottleneck as suggested by the long stretches of LD in their genome. The low levels of mtDNA diversity (Jinam et al. 2012) also provide support for the likelihood of a bottleneck in this population. ADMIXTURE and TreeMix results from CheWong suggest that they are intermediate between Negritos and Senois. Because CheWong appeared distinct at K = 11, it can be inferred that their ancestors experienced one or possibly more admixture events in the past, and later became isolated from founding populations. The argument for CheWongs to be admixed is supported by several factors. First, the cultural practices of CheWong are more similar to other Senoi rather than Negritos, while their language is northern Aslian, similar Negrito dialects. Physically, they appear to have intermediate phenotypes between Negrito and Senoi. The genetic evidence presented here for the first time may reduce disagreement among various anthropologists who study tribes in SEA (Benjamin 2013).

The extent of ROH which are identical segments of an individual’s genome inherited from each parent may be indicative of historical events such as bottlenecks, isolation, and consanguinity within populations. Our findings of markedly longer ROH in Negritos, who are the smallest OA group and fast dwindling, may be due to their small population size and isolation after an early divergence. Given that marriages between siblings and cousins are generally prohibited in current Negrito communities, inbreeding is unlikely to have occurred, although we cannot discount this entirely (Benjamin 2013). They traditionally live in small groups composed of few families; so maintaining a small population over time may have resulted in enriched ROH among them. This parallels some African forager communities that have same lifestyle as hunter-gatherer Negritos (Petersen et al. 2013; Patin et al. 2014).

The longest ROH observed in Seletar may best be explained by the occurrence of a population bottleneck. In contrast, other Proto-Malay groups had shorter and fewer ROH compared with Seletar reflecting their larger outbred communities. LD in Negritos was generally higher compared with other OA groups, a likely consequence of their isolation. The LD patterns from our results are similar to those reported for other isolated groups in Africa and Europe (Gross et al. 2011; Esko et al. 2013; Patin et al. 2014).

The Negrito divergence time is consistent with archeological findings regarding the advent of Hoabinhian culture in Mainland SEA (Bellwood 2007). The genetic evidence supports the view that Malaysian Negritos are descendants of Hoabinhian hunter-gatherers who occupied northern parts of Peninsular Malaysia during late Pleistocene. These hunter-gatherers later interacted with Senoi agriculturists during early Holocene era. It may have been these agriculturists who may have introduced AA-based Aslian languages to Negritos. This time frame also coincides with the Early Train migrations from north to south approximately 10–30 KYA (Jinam et al. 2012). However, our time estimation on LD decay can be affected by any bottleneck experienced by these groups. It has been shown that bottlenecks may result in overestimations of LD in populations which consequently result in underestimation of Ne and divergence time. Nevertheless, there are some challenges associated with our investigation. The ascertainment bias that may be present may affect LD estimation. The considerable difference between Negrito/Senoi and Negrito/Proto-Malay admixture date may suggest that the migration of Senoi ancestors to the Malaysian peninsular occurred earlier than those of Proto-Malays. The latter are believed to be a part of Out-of-Taiwan Austronesian expansion. However, our admixture time estimation seems to be much earlier than archeological reports. In the absence of better analytical methods, our analysis relied on rolloff which may reflect only the most recent admixture event, rather than anything earlier.

To circumvent inaccuracy and further refine divergence times, we performed D statistics to trace ancient admixture within different OA groups and between OAs and other populations in EA. Interestingly, we report gene flow between AA-speaking OAs and Mainland Southeast Asia (MSEA) and Southern Chinese populations. Existence of Negrito ancestral components in some MSEA has been reported by previous studies (HUGO Pan-Asia SNP Consortium 2009).

In summary, we have demonstrated that the current OA while related, are genetically distinct. The Negritos are very different both phenotypically and genetically. The detailed results we have obtained lead us to speculate that their ancestors contributed significant ancestral genetic components probably during the late Pleistocene to the populations of East Asia and SEA. The continuum in divergence times from Negritos to Senois to Proto Malays coupled with the language transitions provide support to a narrative of at least three major human migrations starting with Out of Africa, then the Early Train followed by Out-of-Taiwan Austronesian expansion.

Admin
Administrator

Posts: 81,617

Ayta Negritos: Highest Levels of Denisovan DNA Aug 24, 2021 19:58:51 GMT

Quote

Post by Admin on Aug 24, 2021 19:58:51 GMT

Multiple migrations to the Philippines during the last 50,000 years

PNAS March 30, 2021 118 (13) e2026132118; doi.org/10.1073/pnas.2026132118

Significance
A key link to understand human history in Island Southeast Asia is the Philippine archipelago and its poorly investigated genetic diversity. We analyzed the most comprehensive set of population-genomic data for the Philippines: 1,028 individuals covering 115 indigenous communities. We demonstrate that the Philippines were populated by at least five waves of human migration. The Cordillerans migrated into the Philippines prior to the arrival of rice agriculture, where some remain as the least admixed East Asians carrying an ancestry shared by all Austronesian-speaking populations, thereby challenging an exclusive out-of-Taiwan model of joint farming–language–people dispersal. Altogether, our findings portray the Philippines as a crucial gateway, with a multilayered history, that ultimately changed the genetic landscape of the Asia-Pacific region.

Abstract
Island Southeast Asia has recently produced several surprises regarding human history, but the region’s complex demography remains poorly understood. Here, we report ∼2.3 million genotypes from 1,028 individuals representing 115 indigenous Philippine populations and genome-sequence data from two ∼8,000-y-old individuals from Liangdao in the Taiwan Strait. We show that the Philippine islands were populated by at least five waves of human migration: initially by Northern and Southern Negritos (distantly related to Australian and Papuan groups), followed by Manobo, Sama, Papuan, and Cordilleran-related populations. The ancestors of Cordillerans diverged from indigenous peoples of Taiwan at least ∼8,000 y ago, prior to the arrival of paddy field rice agriculture in the Philippines ∼2,500 y ago, where some of their descendants remain to be the least admixed East Asian groups carrying an ancestry shared by all Austronesian-speaking populations. These observations contradict an exclusive “out-of-Taiwan” model of farming–language–people dispersal within the last four millennia for the Philippines and Island Southeast Asia. Sama-related ethnic groups of southwestern Philippines additionally experienced some minimal South Asian gene flow starting ∼1,000 y ago. Lastly, only a few lowlanders, accounting for <1% of all individuals, presented a low level of West Eurasian admixture, indicating a limited genetic legacy of Spanish colonization in the Philippines. Altogether, our findings reveal a multilayered history of the Philippines, which served as a crucial gateway for the movement of people that ultimately changed the genetic landscape of the Asia-Pacific region.

Fig. 1.
Overview of Philippine population structure. (A) Location of indigenous cultural communities included in the study. The populations are color-coded to represent ethnic group clusters. (B) PCA with worldwide populations, with sample size matched for each regional group. (C) PCA restricted to Philippine ethnic groups. (D) Clustering of Asia-Pacific populations (315,692 SNPs) with an Inset Graph showing clustering of 115 Philippine populations (2.3 million SNPs), assuming K number of genetic ancestry components.

The Philippines is an archipelago of 7,641 islands situated in Island Southeast Asia (ISEA) at the crossroads of past human migrations in the Asia-Pacific region. Until the end of the Last Glacial Period (∼11.7 thousand years ago [kya]), the Philippine islands were mostly contiguous as one large landmass, separated from Sundaland by the Mindoro Strait and the Sibutu Passage (SI Appendix, section 2). Hominins have inhabited the Philippine islands since at least 67 kya (1, 2). While the ancestors of the ethnic groups that self-identify as Negritos are widely regarded as the first human inhabitants (3, 4), their precise relationships to archaic humans, to other early Asian groups, and to subsequent colonizations remain poorly explored and disputable.

Earlier investigations using uniparental markers and/or autosomal data have attempted to resolve the various potential migration events into ISEA (5⇓⇓⇓⇓⇓–11), without arriving at a clear consensus. The lack of resolution may be due to insufficient representation of the diverse ethnic groups of the Philippines and the limited density of the genomic data used. To address these issues, we collected and analyzed the most comprehensive set of population-genomic data thus far for the Philippines: 1,028 individuals representing 115 distinct cultural communities from all geographical regions (Fig. 1A and SI Appendix, Fig. S1A), genotyped for ∼2.5 million single-nucleotide polymorphisms (SNPs). To better understand the history of migrations from the East Asian mainland during the early Holocene, we produced genomic data from two ancient individuals from Liangdao in the Taiwan Strait, with calibrated radiocarbon dates of 8,060 to 8,320 before present (B.P.) and 7,510 to 7,590 B.P., respectively (Fig. 1A).

Our analyses indicate that the Philippines was populated by at least five major human migrations: Northern and Southern Negrito branches of a Basal Australasian group, who likely admixed independently with local Denisovans within the Philippines, plus Papuan-related groups, as well as Manobo, Sama, and Cordilleran branches of Basal East Asians. Cordillerans, who remain the least admixed branch of Basal East Asians, likely entered the Philippines prior to established dates for the agricultural transition and carried with them a genetic ancestry that is widespread among all Austronesian (AN)-speaking populations. This complex demographic history underscores the importance of the Philippines as a migration gateway that profoundly influenced the genetic makeup of populations in the Asia-Pacific region.