Search

Context:Africa’s role in the narrative of human evolution is indisputably emphasised in the emergence Homo sapiens. However, once humans disperse beyond Africa, the history of those who stayed remains vastly understudied lacking the proper attention it deserves as the birthplace of both modern and archaic humans. The sequencing of Neanderthal and Denisovan genomes has elucidated evidence of admixture between archaic and modern humans outside of Africa but has not aided efforts in answering whether archaic admixture happened within Africa. Objectives:We review here the state of research for archaic introgression in African populations and discuss recent insights into this topic. Methods:Gathering published sources and recently released preprints, this review reports on the different methods developed for detecting archaic introgression. Particularly we discuss how relevant these are when implemented on African populations and what findings these studies have shown so far. Results:Methods for detecting archaic introgression have been predominantly developed and implemented on non-African populations. Recent preprints present new methods considering African populations. While a number of studies using these methods suggest archaic introgression in Africa, without an African archaic genome to validate these results, such findings remain as putative archaic introgression. Conclusion:In light of the caveats with implementing current archaic introgression detection methods in Africa, we recommend future studies to concentrate on unravelling the complicated demographic history of Africa through means of ancient DNA where possible and through more focused efforts to sequence modern DNA from more representative populations across the African continent.

The advancements of high-throughput genomics have unveiled much about the human genome highlighting the importance of variations between individuals and their contribution to disease. Even though numerous software have been developed to make sense of large genomics datasets, a major short falling of these has been the inability to cope with repetitive regions, specifically to validate structural variants and accordingly assess their role in disease. Here we describe our program STEAK, a massively parallel software designed to detect chimeric reads in high-throughput sequencing data for a broad number of applications such as identifying presence/absence, as well as discovery of transposable elements (TEs), and retroviral integrations. We highlight the capabilities of STEAK by comparing its efficacy in locating HERV-K HML-2 in clinical whole genome projects, target enrichment sequences, and in the 1000 Genomes CEU Trio to the performance of other TE and virus detecting tools. We show that STEAK outperforms other software in terms of computational efficiency, sensitivity, and specificity. We demonstrate that STEAK is a robust tool, which allows analysts to flexibly detect and evaluate TE and retroviral integrations in a diverse range of sequencing projects for both research and clinical purposes.