In eukaryotic cells, the cell type can dictate the three-dimensional organization of the genome within the nucleus. The organization of chromatin is directly tied to transcriptional regulation, and its disruption often leads to disease. A team from the Baylor College of Medicine and Rice University have shown that this chromatin architecture can be predicted de novo using epigenetic marks within the genome.
"We don't understand exactly how the genome gets marked, but we can measure it through ChIP-sequencing, which has become a fairly straightforward experiment," said co-author Peter Wolynes. "In the same way that we can view genetic code (the DNA), we can also measure these marks directly in many different cells. They've become the next layer of sequence on the genome."
The team obtains ChIP-Seq profiles from the Encyclopedia of DNA Elements (ENCODE) for a lymphoblast cell line that probes 84 different DNA-binding proteins and 11 modifications of histones. These profiles are designated into different loci and then mapped onto a neural network that uncovers the relationship between compartment annotations and epigenetic markings.
The network, called MEGABASE (Maximum Entropy Genomic Annotation from Biomarkers Associated with Structural Ensembles), produces an output sequence of chromatin types. Another program, MiChroM, then uses an energy landscape algorithm to predict the 3-D structures of the chromatin. After training the model on odd-numbered chromosomes, the team has been able to predict the sequences of chromatin types and the subsequent 3D conformational ensembles for the even chromosomes.
"Our findings support the idea that compartmentalization in chromosomes arises from the phase separation of different chromatin types in the nucleus, like the separation of oil and water," said co-author Ryan R. Cheng said.
The team’s simulations using MEGABASE and MiChroM show that segments of chromatin belonging to the same structural type tend to segregate and form liquid droplets. These rearrange dynamically by splitting and fusing, and can explain the emergence of compartmentalization in genomes.
When the team reduces the original dataset to just the 11 histone markings, the output results are only marginally different. They conclude that the chromosome form can be predicted using just histone protein modifications. "There's a well-defined code that connects the histone markings to the structure," said lead author Michele Di Pierre. "It's well-conserved, so it's likely that it has a function."
The team published their findings today in the Proceedings of the National Academy of Sciences (PNAS). "This paper says we can take one-dimensional information about histones and use it with our big-data tools to predict three-dimensional structure," Wolynes said.
Image: Process of microphase separation explains compartmentalization in chromosomes. Chromatin characterized by homogeneous epigenetic markings undergoes a process similar to phase separation under the action of the proteome present in the nucleus. Image courtesy of José N. Onuchic and PNAS.