Rice Biophysicists Model Genome Mechanics

Rice University simulations help explain how genomes take form of 3-D chromosomes

The Minimal Chromosome Model (MiChroM) developed at Rice University predicts chromosome conformations by using the idea that chromatin — which consists of DNA and hundreds of proteins that interact with the genetic material — can be subdivided into types based on their biochemical interactions. Chromatin types, which are distinct from DNA sequence, are partially epigenetically controlled and change during cell differentiation, thus constituting a link among epigenetics, chromosomal organization and cell development. A) The matching conformation of a human chromosome as it appears from Hi-C experiments (lower triangular part of image) and from MiChroM simulations (upper diagonal part). B) A pictorial representation of MiChroM, in which different colors indicate different chromatin types. C) MiChroM output of the 3-D conformation of a human chromosome. (Credit: Illustration by Michele Di Pierro/Rice University)

Rice University scientists trying to solve the ultimate puzzle — the architecture of the human genome — have snapped another piece into place.

Researchers at Rice’s Center for Theoretical Biological Physics have developed a model to explain one part of the mechanism, the folding of chromosomes during a cell’s interphase. Their work offers the possibility of predicting the three-dimensional organization of entire genomes from limited one-dimensional data.

The researchers have used experimental information about one human chromosome to create their Minimum Chromatin Model (MiChroM) and have shown that their model generates accurate 3-D structures for all other chromosomes in the cell.

The new computational tool will help researchers understand how genome architecture contributes to cell development and differentiation.

The paper in the Proceedings of the National Academy of Sciences this week suggests that a true model of chromosome architecture should include not only the code embedded in one’s DNA but also the entire complex of molecules in the cell nucleus, collectively known as chromatin, as they all influence the 3-D arrangement of the genome.

According to the Rice researchers, all of these factors can be recapitulated by subdividing chromatin into just a few types based on their biochemical interactions. They say this simplifies the model and suggests the existence of a hidden code in the genome.

“Chromosomes are very long polymers,” said Rice postdoctoral researcher Michele Di Pierro, co-lead author with former postdoctoral researcher Bin Zhang, now an assistant professor at the Massachusetts Institute of Technology. “The way they’re compressed in a very small space is cell-specific: A lung cell will be different from a brain cell or a liver cell. Part of the difference between these cells is stored in the way the chromosome is folded inside the nucleus.

“So even though we have the same DNA in every cell, the information about different folds in different cells, which is important to cell development and differentiation, is somewhere else. It’s known that this information is partially contained in epigenetics and not in DNA,” Di Pierro said.

“Chromatin types are not simply DNA sequences,” said biophysicist and co-author José Onuchic. “Types are determined by the DNA, the histones and their biochemical modifications, and all the proteins in the cell nucleus. All these factors are part of what we mean by epigenetics and all have an impact on chromosome organization and cell development.”

The Rice team used data drawn from Hi-C experiments, which identify contacts formed between faraway parts of chromosomes as they fold and loop inside the cell nucleus. Erez Lieberman Aiden, a researcher at Rice and at Baylor College of Medicine and a co-author on the new study, led the team that originally created Hi-C.

More recently, scientists in his laboratory reported the highest-resolution Hi-C map ever generated, a dataset 1,000 times the size of the human genome. The Rice researchers, led by Onuchic and biophysicist Peter Wolynes, used this immense dataset to see if chromosomes simulated with MiChroM matched the real ones. They did.

MiChroM “explains the physics of the system,” Di Pierro said. “It’s remarkably efficient at predicting a lot of the known behaviors and effects, well beyond what’s built into the model. This is a good indication that the physics is right.

“It’s still not clear how epigenetics leads to different folds in different cells,” he said. “But here we’re beginning to establish a link between the biochemical modifications through epigenetics and the structure.”

The researchers applied principles similar to those they used to pioneer protein folding, in which the sequence of amino acids in a protein defines its energy landscape; this in turn prescribes how it will fold. In the current research, they show the sequence of chromatin types determines genome folding.

“We’re making a tool that allows us to predict chromosome conformation from a limited set of information,” Di Pierro said. “This is part of a process in which we’re investing a lot of effort.”

The Rice team’s goal is to simulate the mechanisms of the human genome through all of its phases. Previous papers have studied mechanistic details of the process that takes place as a cell moves from interphase, in which it spends most of its time, to the dramatic event of mitosis.

“The new paper is a step into the normal life of a cell and allows us to study how its 3-D organization affects its function,” Wolynes said. “The breakthrough that’s been achieved in this paper will soon give us the mathematical tools to study chromosomes without needing any structural experimental data.”

The National Science Foundation, the Cancer Prevention and Research Institute of Texas and the Welch Foundation supported the research.