Back

Fetal signatures in the 3D genome of iPSC-derived neurons: implications for disease modeling

Zagirova, D. R.; Kononkova, A. D.; Morozov, K. V.; Molodova, M. N.; Vaulin, N. S.; Dudkovskaia, A. V.; Dozorova, P. I.; Efimova, O. I.; Tvorogova, A. V.; Ulianov, K. A.; Khaitovich, P. E.; Razin, S. V.; Lagarkova, M. A.; Ulianov, S. V.; Khrameeva, E. E.

2025-08-03 genomics
10.1101/2025.08.03.667702 bioRxiv
Show abstract

Induced pluripotent stem cells (iPSCs) have revolutionized neuroscience, providing an approach to generate patient-specific neurons for modeling of neurological diseases. However, it remains unclear how closely iPSC-derived neurons replicate the chromatin architecture of authentic brain neurons. Here, we collected and uniformly processed more than 300 human and mouse neuronal Hi-C datasets and examined chromatin features ranging from chromatin compartments and topologically associating domains (TADs) to chromatin loops, Polycomb-mediated contacts, and frequently interacting regions (FIREs). We found that iPSC-derived neurons largely retain chromatin state of undifferentiated cells and resemble fetal rather than mature neurons. iPSC-derived neurons exhibit unusually strong compartmentalization, an enrichment of developmental genes at TAD borders, and a marked reduction of long-range repressive Polycomb-mediated contacts that typically silence early fetal programs. Conversely, these cells robustly capture contacts involving GWAS SNPs relevant to neurological diseases. Thus, while iPSC-derived neurons may not fully replicate 3D genome architecture of adult neurons, they effectively model interactions linking disease-associated SNPs to target genes. Our study offers a valuable Hi-C resource for the community and provides a detailed comparison of chromatin architecture throughout neuronal maturation, underscoring its importance for validating neuronal models and providing a robust framework for future studies.

Published in Genome Research (predicted rank #7) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.