A mathematical model could make a new epoch in stem cell treatments

“The big question in the field is which changes first, the epigenome or the transcriptome,” said Joshua Welch, Ph.D., an assistant professor in the Department of Computational Medicine and Bioinformatics at the U-M Medical School.

Published in Nature Biotechnology on October 13, the study demonstrates a mathematical model may be used to estimate the timing.

Why are they moving so slowly?

Of course, there are many reasons why scientists are so lingering, but basically, it's about the properties of DNA and RNA.

The genome adjusts the shape of chromatin, the tightly wound bundle that houses the DNA, in response to cues from the environment that tell it what to express and when to express it. This means that the genome doesn't only carry out the instructions written in its DNA. The epigenome is the name given to these molecular signals.

Researchers were unable to observe gene expression in a single cell until recently. They can today because of single-cell sequencing technology. The cell is destroyed when it is measured, thus it is still difficult to visualize the timing of changes.

“To address this, we developed an approach based on models in basic physics,” explained Welch, “treating the cells like they are masses moving through space and we are trying to estimate their velocity.”

The model, named "MultiVelo", predicts the direction and speed of the molecular changes the cells are undergoing.

“Our model can tell us which things are changing first–epigenome or gene expression--and how long it takes for the first to ramp up the second,” said Welch.

Four types of stem cells were used

They found two ways the epigenome and transcriptome can be out of sync and were able to validate the strategy using four different stem cell types the brain, blood, and skin.

So-called cellular atlases, which are being created utilizing single-cell sequencing to view the many cell types and gene expression in diverse body systems, are given an additional, crucial layer of understanding by the technology.

“One of the big problems in the field is the artificially differentiated cells created in the lab never quite make it to full replicas of their real-life counterparts,” said Welch. “I think the biggest potential for this model is better understanding what are the epigenetic barriers to fully converting the cells into whatever target you want them to be.”

Abstract:

Multi-omic single-cell datasets, in which multiple molecular modalities are profiled within the same cell, offer an opportunity to understand the temporal relationship between epigenome and transcriptome. To realize this potential, we developed MultiVelo, a differential equation model of gene expression that extends the RNA velocity framework to incorporate epigenomic data. MultiVelo uses a probabilistic latent variable model to estimate the switch time and rate parameters of chromatin accessibility and gene expression and improves the accuracy of cell fate prediction compared to velocity estimates from RNA only. Application to multi-omic single-cell datasets from brain, skin, and blood cells reveals two distinct classes of genes distinguished by whether chromatin closes before or after transcription ceases. We also find four types of cell states: two states in which epigenome and transcriptome are coupled and two distinct decoupled states. Finally, we identify time lags between transcription factor expression and binding site accessibility and between disease-associated SNP accessibility and expression of the linked genes. MultiVelo is available on PyPI, Bioconda, and GitHub.