An increasing body of archaeological and genomic evidence has hinted at a complex settlement process of the Americas by humans. This is especially true for South America, where unexpected ancestral signals have raised perplexing scenarios for the early migrations into different regions of the continent. Here, we present ancient human genomes from the archaeologically rich Northeast Brazil and compare them to ancient and present-day genomic data. We find a distinct relationship between ancient genomes from Northeast Brazil, Lagoa Santa, Uruguay and Panama, representing evidence for ancient migration routes along South America's Atlantic coast. To further add to the existing complexity, we also detect greater Denisovan than Neanderthal ancestry in ancient Uruguay and Panama individuals. Moreover, we find a strong Australasian signal in an ancient genome from Panama. This work sheds light on the deep demographic history of eastern South America and presents a starting point for future fine-scale investigations on the regional level.

The prehistory of the people of Uruguay is greatly complicated by the dramatic and severe effects of European contact, as with most of the Americas. After the series of military campaigns that exterminated the last remnants of nomadic peoples, Uruguayan official history masked and diluted the former Indigenous ethnic diversity into the narrative of a singular people that all but died out. Here, we present the first whole genome sequences of the Indigenous people of the region before the arrival of Europeans, from an archaeological site in eastern Uruguay that dates from 2,000 years before present. We find a surprising connection to ancient individuals from Panama and eastern Brazil, but not to modern Amazonians. This result may be indicative of a migration route into South America that may have occurred along the Atlantic coast. We also find a distinct ancestry previously undetected in South America. Though this work begins to piece together some of the demographic nuance of the region, the sequencing of ancient individuals from across Uruguay is needed to better understand the ancient prehistory and genetic diversity that existed before European contact, thereby helping to rebuild the history of the Indigenous population of what is now Uruguay.

Despite the extreme and varying environmental conditions prevalent in the Arabian Peninsula, it has experienced several waves of human migrations following the out-of-Africa diaspora. Eventually, the inhabitants of the peninsula region adapted to the hot and dry environment. The adaptation and natural selection that shaped the extant human populations of the Arabian Peninsula region have been scarcely studied. In an attempt to explore natural selection in the region, we analyzed 662,750 variants in 583 Kuwaiti individuals. We searched for regions in the genome that display signatures of positive selection in the Kuwaiti population using an integrative approach in a conservative manner. We highlight a haplotype overlapping TNKS that showed strong signals of positive selection based on the results of the multiple selection tests conducted (integrated Haplotype Score, Cross Population Extended Haplotype Homozygosity, Population Branch Statistics, and log-likelihood ratio scores). Notably, the TNKS haplotype under selection potentially conferred a fitness advantage to the Kuwaiti ancestors for surviving in the harsh environment while posing a major health risk to present-day Kuwaitis.

Natural selection leaves detectable patterns of altered spatial diversity within genomes, and identifying affected regions is crucial for understanding species evolution. Recently, machine learning approaches applied to raw population genomic data have been developed to uncover these adaptive signatures. Convolutional neural networks (CNNs) are particularly effective for this task, as they handle large data arrays while maintaining element correlations. However, shallow CNNs may miss complex patterns due to their limited capacity, while deep CNNs can capture these patterns but require extensive data and computational power. Transfer learning addresses these challenges by utilizing a deep CNN pretrained on a large dataset as a feature extraction tool for downstream classification and evolutionary parameter prediction. This approach reduces extensive training data generation requirements and computational needs while maintaining high performance. In this study, we developed TrIdent, a tool that uses transfer learning to enhance detection of adaptive genomic regions from image representations of multilocus variation. We evaluated TrIdent across various genetic, demographic, and adaptive settings, in addition to unphased data and other confounding factors. TrIdent demonstrated improved detection of adaptive regions compared to recent methods using similar data representations. We further explored model interpretability through class activation maps and adapted TrIdent to infer selection parameters for identified adaptive candidates. Using whole-genome haplotype data from European and African populations, TrIdent effectively recapitulated known sweep candidates and identified novel cancer, and other disease-associated genes as potential sweeps.

Mutually beneficial partnerships between genomics researchers and North American Indigenous Nations are rare yet becoming more common. Here, we present one such partnership that provides insight into the peopling of the Americas and furnishes another line of evidence that can be used to further treaty and Indigenous rights. We show that the genomics of sampled individuals from the Blackfoot Confederacy belong to a previously undescribed ancient lineage that diverged from other genomic lineages in the Americas in Late Pleistocene times. Using multiple complementary forms of knowledge, we provide a scenario for Blackfoot population history that fits with oral tradition and provides a plausible model for the evolutionary process of the peopling of the Americas.

Identifying genomic regions shaped by natural selection is a central goal in evolutionary ge-nomics. Existing machine learning methods for this task are typically trained using simulated genomic data labeled according to specific evolutionary scenarios. While effective in controlled settings, these models are limited by their reliance on explicit class labels. They can only detect the specific processes they were trained to recognize, making it difficult to interpret predictions for regions influenced by other evolutionary forces. This limitation is especially problematic when analyzing empirical genomes shaped by a mixture of adaptive, demographic, and ecolog-ical factors. One-vs.-rest strategies offer a potential alternative, but suffer from the inherent complexity of modeling all other evolutionary and demographic processes as a catch-all “rest” class. Here, we explore positive-unlabeled learning as a more flexible framework for detection of adaptive events. Positive-unlabeled learning is a semi-supervised approach that permits iden-tification of samples of a target class using only positive labels and an unlabeled background, without requiring explicit modeling of negative samples. To assess the utility of this approach, we focus on a binary classification setting for detecting selective sweeps (positive samples) aris-ing from positive natural selection against a mixed background composed of both unlabeled sweeps and neutrally-evolving regions. To accomplish this goal, we introduce PULSe , a method that employs only a set of labeled sweep observations for training while treating all remaining data as unlabeled. By avoiding assumptions about the composition of the background, PULSe enables robust sweep discovery in realistic genomic landscapes. We systematically evaluate its performance across a range of demographic, adaptive, and confounding contexts, including domain shift arising from misspecified demographic models, and find that PULSe delivers high performance and generalizability. To demonstrate a practical application of PULSe , we ana-lyzed European and Bengali in Bangladesh human genomes, treating the empirical genomes as unlabeled sets, and recapitulating several previously-identified sweep candidates. Our results show that PULSe provides a powerful and versatile alternative for detecting adaptive regions, with the potential to generalize across a range of genomic landscapes.