@fionacassiopea.bsky.social

Want to explore your own data with CroCoNet? Check out the R package and the associated tutorial at hellmann-lab.github.io/CroCoNet/.
Huge thanks to all authors and everyone who helped shape CroCoNet!

We found that human- and great-ape-specific transposable elements from the LTR7 family are enriched near POU5F1 target genes, suggesting that these insertions may have contributed to the rewiring of the network. This example showcases that CroCoNet can successfully detect shifts in gene regulation.

We applied CroCoNet to a scRNA-seq dataset of primate early neural differentiation. The module of the pluripotency factor POU5F1 (OCT4) turned out to be one of the most diverged among all network modules, even though POU5F1 itself has a conserved sequence and expression pattern.

To address this challenge, CroCoNet relies on network variability between replicates of the same species and compares cross-species variability to this baseline. This way it can detect meaningful changes, pointing us to the most conserved and diverged modules of the network.

Studying how gene regulatory networks are rewired across species helps us understand phenotypic evolution. But networks inferred from transcriptomic data are notoriously noisy, and cross-species comparisons are particularly sensitive to this noise, as it can mask true evolutionary differences.

We applied CroCoNet to a scRNA-seq dataset of primate early neural differentiation. The module of the pluripotency factor POU5F1 (OCT4) turned out to be one of the most diverged among all network modules, even though POU5F1 itself has a conserved sequence and expression pattern.

To address this challenge, CroCoNet relies on network variability between replicates of the same species and compares cross-species variability to this baseline. This way it can detect meaningful changes, pointing us to the most conserved and diverged modules of the network.

Studying how gene regulatory networks are rewired across species helps us understand phenotypic evolution. But networks inferred from transcriptomic data are notoriously noisy, and cross-species comparisons are particularly sensitive to this noise, as it can mask true evolutionary differences.

For those who want to try it in the lab:
we also published a step-by-step, easy-to-follow protocol for prime-seq2 🧪
👉 dx.doi.org/10.17504/pro...

@pascalstuempfl.bsky.social