Benjamin Peter

@jaisonjseq That is the expectation if Denisovan ancestry is mostly neutral, and if there's little gene flow with people that have different amounts of Denisovan ancestry.

@benmpeter Amazing work! How do you explain the similar proportions of Denisovan ancestry in modern and ancient samples with a time difference of over 25,000 years?

If you need more exciting news about Denisovans, check out our paper on Denisovan ancestry in modern humans through time, out today in Current Biology: cell.com/current-biolog…! #Denisovan #ancientDNA

Also check out the dispatch by Shigeki Nakagome: cell.com/current-biolog…, and (of course) our new preprint on the genome of a 200,000 year old Denisovan biorxiv.org/content/10.110…

The most surprising insight is that the Jomon, early inhabitants of the Japanese Archipelago, have much less Denisovan ancestry than all other East Asians - thus the Jomon (partially) descend from a lineage that predates the gene flow between modern-humans and Denisovans

@moorjani_priya @benmpeter @IasiLeonardo @M_Chintalapati @SkovLaurits @MatejaHajdi @arevsumer This is a really cool study and I'm glad to see it come out!

My summary of the new papers about Neanderthal interbreeding

The first paper is a systematic survey of Neandertal ancestry through time; a joint effort with @moorjani_priya ’s lab at UC Berkeley, co-led by @IasiLeonardo and @M_Chintalapati. Leo has a great summary of our main findings here:

Today is a very big day for our research group, with two of my students, @arevsumer and @IasiLeonardo publishing papers on the same day! science.org/doi/10.1126/sc… nature.com/articles/s4158… #Neanderthals #Geneflow

@reject_resubmit Welcome Mark! Really excited to have you as a future colleague!

I am absolutely thrilled to share that in July 2025 I'll be moving across the lake to the University of Rochester to join the Department of Biology as an Assistant Professor!

@MPI_EVA_Leipzig @MatejaHajdi @maxplanckpress Congrats @MatejaHajdi !

Great news! 🥳 Led by the amazing @MatejaHajdi, a new Max Planck Research Group for Hominin Palaeogenomics started @MPI_EVA_Leipzig. 😊😍 Using cutting-edge #aDNA methods, its goal is to shed light on past encounters between hominin groups. @maxplanckpress eva.mpg.de/press/news/art…

9) We searched for candidates of adaptive selection using an outlier approach. We found that many previously published candidate regions related to skin pigmentation, metabolism, and immunity were immediately selected, with some regions becoming adaptive at a later stages.

8) We examined the distribution of Neandertal ancestry over time and found that the ancestry landscape, including the Neandertal deserts, were formed quickly after the gene flow supporting theoretical predictions. We infer the initial Neandertal gene flow may have been >5%.

7) Our data supports a single extended pulse of Neandertal gene flow that took place around 47,000 years ago, with a duration of around 6,000 years.

6) Digging deeper, we infer the timing of Neandertal gene flow per individual by studying the ancestry covariance across the genome.

5) The Early out of Africa individuals including Oase1, Ust’ Ishim and the Bacho Kiros have the highest unique Neandertal ancestry and significantly different matching profile to sequenced Neandertals than later modern humans- a possible indication of diversified gene flow.

4) By exploring the colocalization of Neandertal ancestry segments and contrasting it to genome-wide allele sharing, we uncovered that most (>90%) Neandertal ancestry is shared among individuals and follows the overall population structure.

3) We inferred Neandertal segments in the individuals using admixfrog (github.com/BenjaminPeter/…), and used the ancestry covariance for dating (github.com/MoorjaniLab/Pr…).

2) We use genomic data from 59 ancient modern human individuals ranging between 45,000–2,200 yBP with 275 diverse present-day individuals from worldwide populations forming 16 clusters for our analysis.

1) Previous studies characterized the Neandertal ancestry segments in present-day populations and found a non-homogeneous distribution. But how did this distribution look like in the past?