🧬 genomics

Genomic impact of the second plague pandemic on three human populations

This study analyzes whole-genome sequencing data from 529 ancient individuals across three northern European sites and finds no evidence that the second plague pandemic drove strong positive selection on protective genetic variants, but instead reveals a significant reduction in long-range immigration and ancestry diversity following the pandemic's arrival, coinciding with the end of the Viking Age, Christianisation, and the onset of the Little Ice Age.

Original authors: Liu, X., Moore, K., Ebenesersdottir, S. S., Arcini, C., Walker, G.-T., Denham, S. D., Slavin, P., Sotofte, M. B., Nielsen, S. D., Ellegaard, M. R., Vagene, A. J., Margaryan, A., Iraeta-Orbegozo, M., M

Published 2026-07-10

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Liu, X., Moore, K., Ebenesersdottir, S. S., Arcini, C., Walker, G.-T., Denham, S. D., Slavin, P., Sotofte, M. B., Nielsen, S. D., Ellegaard, M. R., Vagene, A. J., Margaryan, A., Iraeta-Orbegozo, M., Mylopotamitaki, D., Laffoon, J., Philippsen, B., Rydahl, M. C., Jankauskas, R., Kozakaite, J., Vol, E., Lomes, A., Waldman, S., Carmi, S., Alfredsson, L., Cavalleri, G. L., Chen, H. S., Cheronet, O., Demetz, L., Emeruem, D. N., Fernandes, D., Gelabert, P., Gilbert, E., Hansen, T. F., Hovig, E., Kockum, I., Llanos-Lizcano, A., Oberreiter, V., Olsson, T., Pinhasi, R., Praxmarer, E., Skar, B., Werge,

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine the Black Death not just as a terrifying ghost story from the 1300s, but as a massive, invisible filter that swept across Europe. For years, historians and scientists have wondered: Did this filter change the very DNA of the people who survived? Did it act like a giant sieve, keeping only the "super-healthy" genes and washing away the rest?

A team of researchers decided to play detective with time travel. They gathered the genetic blueprints (DNA) from 529 ancient skeletons found in three cities: Lund in Sweden, Trondheim in Norway, and Vilnius in Lithuania. They split these people into two groups: those who lived before the plague arrived (pre-BD) and those who lived after (post-BD). Think of it as comparing the genetic "menu" of a city before and after a massive storm.

The Big Surprise: No Genetic "Superpower" Found

The team ran a high-speed scan of the entire genome, looking for any specific genetic variant that suddenly became super common after the plague. They were hoping to find a "hero gene" that helped people survive the Yersinia pestis bacteria.

But here's the twist: They found nothing.

Despite having a much larger and clearer dataset than any previous study (529 people with high-quality DNA), they could not find a single genetic variant that showed strong evidence of being "selected" by the plague. They checked the specific genes other scientists had pointed to before (like ERAP2 and CCR5), and none of them showed the dramatic changes in frequency that would prove the plague was the cause.

The authors are very clear on this: they didn't just miss it; their study was powerful enough to spot big changes if they existed. It's like searching a library with a flashlight that can see every single book, only to find that the "survival section" hasn't been reorganized at all. They suggest that maybe the plague killed people so quickly that there wasn't enough time for genetics to shift, or perhaps the medieval population just didn't have the specific "super-protective" genes to begin with.

The Real Story: The City Got Smaller and Quieter

If the plague didn't change the genes inside the people, did it change the people themselves? Yes, absolutely.

The study found a massive shift in where these people came from. Before the plague, these cities were like bustling international airports. The DNA showed that a huge chunk of the population (up to 52% in Lund) were "outliers"—people whose ancestry traced back to faraway places like the British Isles, Germany, France, and even Iceland. It was a cosmopolitan mix, likely driven by the excitement of the Viking Age, the spread of Christianity, and trade.

But after the plague hit, the "airport" closed its long-haul gates.

The researchers observed a sharp drop in this long-distance diversity. In the post-plague era, the population became much more "local." In Lund, the number of people with foreign ancestry dropped from 52% to just 9%. The cities became more insular. It's as if the vibrant, noisy market square suddenly emptied out, leaving only the locals behind.

Why Did This Happen?

The paper suggests this wasn't just the plague doing the work. It was a "perfect storm" of history. The plague arrived right around the same time the Viking Age ended, the climate got colder (the Little Ice Age), and the social structures were shifting. The massive death toll likely broke the trade networks and the religious connections that had been pulling people from all over Europe into these cities.

The authors note that while the plague was a huge part of the story, it was likely the combination of these events that stopped the long-distance travelers from coming. The cities didn't just lose people; they lost their variety.

The Bottom Line

So, did the Black Death rewrite the human genetic code to make us stronger? No. The data suggests that the plague didn't leave a mark on our DNA in the way we hoped.

Did it change the makeup of our cities? Yes, dramatically. It turned diverse, international hubs into more closed-off, local communities. The paper proves that while the plague didn't select for a "survival gene," it definitely selected for a quieter, less connected world. The authors are confident in these numbers because they looked at a massive amount of data, but they admit that to find those tiny genetic shifts, we might need even more skeletons to dig up in the future. For now, the story of the Black Death is less about genetic superpowers and more about how a tragedy can shrink the world around us.

Technical Summary: Genomic Impact of the Second Plague Pandemic on Three Human Populations

Problem Statement
The second plague pandemic (early 14th–early 19th centuries), caused by Yersinia pestis, resulted in massive demographic collapse across Eurasia, with mortality rates estimated at 40–60% in many European regions. While historical records document profound socio-economic shifts, the genetic consequences of this pandemic remain debated. Previous studies have proposed that the pandemic drove strong positive selection on specific immune-related loci (e.g., ERAP2, CCR5, TLR genes) and altered migration patterns. However, prior ancient DNA (aDNA) investigations have been limited by small sample sizes, single-site focus, or reliance on targeted capture of specific genomic regions rather than whole-genome sequencing. This study addresses the need for a larger, higher-coverage, and geographically diverse dataset to rigorously test for plague-driven selection and to quantify demographic changes in ancestry composition before and after the pandemic's arrival.

Methodology
The authors analyzed a dataset of 529 high-quality ancient whole-genomes (mean depth 8.8x) from three northern European urban sites: Lund (Sweden, N=337), Trondheim (Norway, N=135), and Vilnius (Lithuania, N=57). Individuals were temporally stratified into "pre-Black Death" (pre-BD) and "post-Black Death" (post-BD) groups based on archaeological context and radiocarbon dating.

The analytical pipeline included:

Quality Control and Imputation: After filtering for low coverage and contamination, genotypes were imputed using GLIMPSE2 against a reference panel of 50,129 contemporary individuals (deCODE 50k), resulting in ~99.6 million loci.
Selection Scans: Genome-wide association studies (GWAS) were performed using linear mixed models (GEMMA) to detect allele frequency differences between pre-BD and post-BD groups. Covariates included sex and the first seven principal components (PCs) from a West Eurasian PCA to control for population structure. A meta-analysis combined results from Lund and Trondheim to increase power.
Replication Checks: The study specifically tested allele frequencies at loci previously reported as under selection (e.g., ERAP2, TICAM2, NFATC1, CTLA4, RIPK2, CCR5-delta32, and TLR cluster variants).
Ancestry and Demography Analysis:
- PCA and Dispersion: Ancient individuals were projected onto a PCA of 9,052 present-day West Eurasians. Multivariate dispersion (generalized variance) and Mahalanobis distances ( $D_M$ ) were calculated to assess ancestry diversity and proximity to modern regional proxies.
- Identity-by-Descent (IBD): IBD sharing (>6 cM for Scandinavia, >7 cM for Lithuania) was calculated between ancient individuals and large present-day reference panels (OmniExpress for Scandinavia; UK Biobank for Baltic/Eastern Europe) to identify ancestry outliers and local vs. non-local origins.
- Outlier Classification: A two-step framework combining PCA-based $D_M$ and IBD sharing was used to classify individuals as "outliers" (non-local ancestry) and further subdivide locals into "local" and "non-local" within Scandinavia.

Key Results

Absence of Strong Positive Selection: The study found no genome-wide significant evidence for strong positive selection driven by Y. pestis exposure.
- No variants met the functionally informed genome-wide significance thresholds in the meta-analysis of Lund and Trondheim.
- Previously reported candidate loci (including ERAP2) showed no significant allele frequency shifts between pre-BD and post-BD groups.
- Power analysis indicated the study was well-powered (>80%) to detect frequency changes of the magnitude reported in previous studies (e.g., for ERAP2), suggesting that if such strong selection occurred, it would have been detected. The authors note that subtle frequency changes (<0.12–0.26) or selection on rare variants would require even larger sample sizes to detect.
Reduction in Ancestry Diversity: In contrast to the lack of selection signals, the study identified a marked decline in the diversity of ancestry at all three sites following the arrival of the Black Death.
- PCA Dispersion: Pre-BD individuals exhibited significantly greater dispersion in PCA space compared to post-BD individuals (Box's M test: $P < 2.2 \times 10^{-16}$ for Lund).
- Outlier Decline: The proportion of ancestry outliers (individuals with non-local ancestry) dropped significantly from pre-BD to post-BD. In Lund, outliers decreased from 52% to 9% ( $P = 1.37 \times 10^{-13}$ ); in Vilnius, from 47% to 5% ( $P = 3.71 \times 10^{-4}$ ).
- Specific Ancestry Shifts: Pre-BD Lund showed high diversity with significant contributions from the British Isles, Ireland, Germany, and the Benelux region. Post-BD, the population became more homogeneous, clustering closer to present-day Scandinavian populations. In Trondheim, pre-BD outliers included a notable number of individuals with Icelandic ancestry, which declined post-BD.
- Local vs. Non-Local: While long-range immigration (outliers) decreased, the proportion of non-local Scandinavians (migration within the region) remained statistically unchanged between periods.

Significance and Claims
The paper claims that while the second plague pandemic was a demographic catastrophe, it did not leave a detectable signature of strong positive selection on the human genome in these northern European populations, challenging the hypothesis that the pandemic acted as a primary driver of rapid adaptive evolution in immune genes.

Instead, the study highlights that the pandemic coincided with a broader socio-historical transition toward more insular societies. The authors attribute the observed reduction in long-range migration and ancestry diversity to a confluence of factors:

The end of the Viking Age.
The onset of the Little Ice Age.
The process of Christianisation, which initially attracted diverse missionaries and emissaries (contributing to pre-BD diversity) but may have subsequently stabilized local populations.
The pandemic itself, which likely disrupted the commercial, ecclesiastical, and political networks that facilitated long-distance mobility.

The authors conclude that the genetic impact of the pandemic in these regions was primarily demographic (reducing migration and diversity) rather than selective (altering allele frequencies of protective variants). They emphasize that detecting subtle selection signals would require substantially larger sample sizes or sampling strategies focusing on individuals who died during the acute phase of the pandemic versus survivors.

More like this