Julien JOSEPH

I am very pleased to share with the mastodon community the first part of my PhD work:

High prevalence of Prdm9-independent recombination hotspots in placental #mammals

https://www.biorxiv.org/content/10.1101/2023.11.17.567540v1

This work was done in collaboration with @djivanprentout Alexandre Laverré, Théo Tricou and @duret_lbbe. (1/8)

#Recombination #PopGen #Evolution #gBGC #PRDM9

Nov 20, 2023 at 10:21amWeb
Show threadJulien JOSEPHNov 20, 2023
In many euraryotes, #recombination events are not evenly distributed across the genome and tend to be concentrated in so-called recombination hotspots. In humans and mice, nearly all recombination hotspots are determined by the protein PRDM9. In species lacking PRDM9, recombiation tends to occur in promoter-like features such as CpG islands (default hotspots). In this study, we have shown that these default hotspots are also active in the presence of PRDM9 in many placental mammals. (2/8)
Show threadJulien JOSEPHNov 20, 2023
As linkage disequilibrium maps are costly to build for a large number of species, we used another proxy for recombination rate that rely on the signature left in base composition by recombination-associated gene conversion events (GC-biased gene conversion #gBGC). This proxy is based on the equilibrium GC content, i.e. the GC content that the sequence would reach if the substitution patterns occurring in a branch were to continue for a long period of time. (3/8)
Show threadJulien JOSEPHNov 20, 2023
We first show that hotspots of a PRDM9 KO mouse (default hotspots) are also used in dogs that naturally lack PRDM9. We can see that in dogs, the signal of equilibrium GC content is highly congruent with that of linkage disequilibrium, as observed in a wider range of organisms including several birds, snakes and humans. Notably, we confirm that recombination activity in dogs is strongly associated with DNA hypomethylation. (4/8)
Show threadJulien JOSEPHNov 20, 2023
Using Nagylaki's derivation of the fixation probability of a mutation in the presence of biased gene conversion, we then developed an estimator that can directly translate the equilibrium GC content into a relative recombination rate that has occured in the terminal branch of a given species. (5/8)
Show threadJulien JOSEPHNov 20, 2023
Using this estimator, we can see that recombination rates at loci orthologous to the strong DSB hotspots of a mouse KO for PRDM9 (mouse default hotspots) are quite high in most mammals, and particularly canids who have lost the gene. Of note, the northern elephant seal, the ring-tailed lemur and the Daurian ground squirrel display recombination rates at these loci that are comparable to those of canids. (6/8)
Show threadJulien JOSEPHNov 20, 2023
We controlled for many confounders and concluded that the signal was very unlikely to be driven by other biological processes affecting GC content evolution or methodological artefacts, details are in the preprint. (7/8)
Show threadJulien JOSEPHNov 20, 2023
These results show that PRDM9 may not be the sole determinant of recombination hotspots in mammals. These results also raise the question of why the ancestral mechanism for directing recombination has been maintained and why there is variability in its use in mammals. If you think that these results might be of interest to some of your colleagues, please don't hesitate to share the preprint, I would be very happy to receive feedback. (8/8)
Show threadBenoit NabholzNov 20, 2023
@Julien_JOSEPH very nice. Congrats Julien!
Show threadJulien JOSEPHMay 29, 2024

This is now published:

https://www.pnas.org/doi/full/10.1073/pnas.2401973121

It is not freely available until 6 month (immediate open access is too expensive), but feel free to send me a message if you want the pdf!

Show threadBenoit NabholzJun 3, 2024
@Julien_JOSEPH congratulation Julien!