Researchers find genetic signatures of ‘ghost’ and ‘super-archaic’ lineages
Researchers using new computational techniques have identified new complexity in the genetic ancestry of Homo sapiens. It has long been established that our ancestors interbred with Neanderthals and Denisovans, but the work indicated that at least two other archaic hominin species contributed to our ancestry.
The new techniques involve ancestral recombination graphs (ARGs), which provide a complete record of all the coalescence and recombination events and specifies a complete genealogy at each position in the genome. Regions whose ancestry extends unusually far back in time can be used to identify genetic contributions from extinct human populations, even in the absence of ancient DNA as we have for Neanderthals and Denisovans. TRACE (TRacing Archaic Contributions via ARG Estimation) is a method developed by the researchers to identify genetic signatures of archaic ancestry in modern human genomes using ARGs constructed solely from present-day genomes.
Some of the regions with older ancestry matched known Neanderthal and Denisovan sequences, but there were also many ancient regions containing sequences that matched neither.
These sequences represented two distinct lineages, with different introgression times. The first was a ‘ghost’ lineage that diverged from Homo sapiens more than 500,000 years ago. The ghost lineage is present in all present-day populations, both African and non-African. This implies that interbreeding occurred before modern humans dispersed in large numbers from Africa around 65,000 years ago.
Ghost ancestry persists in so-called Neanderthal and Denisovan ancestry deserts, regions of the genome where these ancestries are absent. This finding challenges the view that the ‘deserts’ were either due to positive selection for Homo sapiens-specific alleles or strong purifying selection against all archaic introgressed alleles. Instead, the persistence of ghost ancestry in these regions suggests purifying selection restricted to Neanderthal- and Denisovan-introgressed alleles. The Neanderthal and Denisovan genes must have been selected against for reasons unrelated to their divergence time from Homo sapiens.
The second lineage was a ‘super-archaic’ lineage that diverged 1.77 million years ago, long before the common ancestor of modern humans, Neanderthals, and Denisovans. It shows up in modern populations alongside Denisovan sequences, suggesting that it entered the modern genome via interbreeding with Denisovans.
The researchers were unable to determine the number, timing, and geographic locations of the ghost introgression events. The results supported at least one pre-Out-of-Africa event, but other episodes may have occurred among structured African populations. The origin and geographic extent of the super-archaic lineage also remain unknown, including whether the super-archaic ancestry entered modern humans solely via interbreeding with Denisovans or also through direct interbreeding.
With this in mind, associating the archaic lineages with known hominin species will be speculative at best. In Africa, early Homo sapiens likely overlapped with Homo naledi, the small-bodied hominins discovered in the Rising Star cave system in South Africa in 2013. The divergence date of 1.77 million years ago for the ‘super-archaics’ fits Homo erectus, and they may well have overlapped with Denisovans in East Asia. Arguably, we should not be thinking in terms of discrete ‘species’ at all. Instead, it might be better to think in terms of populations that remained isolated from one another for hundreds of thousands of years before coming into contact and interbreeding. The ‘super-archaic’ and ‘ghost’ lineages would simply represent populations that eventually became extinct.
References
Zhang, Y., Biddanda, A., Johnson, S. A., O’Dushlaine, C. & Moorjani, P. Recovering signatures of archaic hominin introgression using ancestral recombination graphs. Science 393, eaef8874 (2026).