Magnus Nordborg: “Pangenomes” and the path towards an unbiased characterization of genetic polymorphism

  • Datum: 25.11.2026
  • Uhrzeit: 11:00 - 12:00
  • Ort: Max Planck Instititute for Plant Breeding Research
  • Raum: Lecture hall
  • Gastgeber: André Marques
person

Our current view of genetic polymorphism is heavily biased by technology that relies on mapping short sequences to a reference genome. The technology to overcome this problem now exists: it has become feasible to independently and completely sequence genomes from large samples of individuals. However, making sense of such whole-genome polymorphism data is far from easy. I will provide an overview of the challenges involved as well as the tremendous opportunities for new insights, drawing in particular from our experience with generating a collection of over 450 complete or near-complete genomes of the model plant Arabidopsis thailiana.

As expected, these efforts have revealed how incomplete our understanding of polymorphism is. Not only are SNP-data biased and full or errors, we now know that, on a base-pair level, structural variants make up many times more of the differences between two genomes, and we can also gain insight into variation in genome size (in A. thaliana mostly due to centromeric and rDNA tandem-repeat regions). We can also investigate variation in gene content, and, at least in A. thaliana, we find little evidence for the kind of pan-genome that characterizes many microbial species, with a set of essential “core” genes complemented with a large number of non-essential ones that are maintained only in the species pan-genome. Instead, we find a large number of lowly-expressed and epigenetically silenced “genes” of unknown and dubious importance, at least half of which appear to be related to transposable element activity.

Finally, our view of the mobilome is revolutionized by whole-genome polymorphism data. Taking advantage of the fact that mobile elements by definition generate structural variation, we can simply cluster sequences involved in structural variation in order to identify mobile element families without any prior notion of what they look like, allowing us to produce a mobile element annotation automatically for any species simply by sequencing it. The resulting annotation can then be compared with classical annotation methods. Mobile element polymorphism data also allow us to ask questions about bursts of transposition and selective effects.


Zur Redakteursansicht