Using gene networks to make sense of single cell biology

What is it about?

Genes encode proteins and proteins dictate cell function. Gene activity or how much of a gene is expressed in a particular cell determines what that cell can do in its life cycle. Recent advances in genomics now allow us to look at gene expression profiles from single cells, as opposed to examining samples made of thousands or millions of cells in bulk. By measuring gene expression in single cells we will be capable of identifying and understanding human cells and molecular states within tissues. However, the technology is still evolving and gene expression measurements from single cells have considerable technical and biological noise. A number of computational methods are recently proposed to dampen that noise.

Featured Image

Why is it important?

We have implemented a method for smoothing the noisy gene expression values obtained from single-cell gene expression experiments. Our method leverages a plethora of previous experiments that illuminate the nature of gene interactions. These take the form of gene interaction networks, which we use to smooth the gene expression values in single-cell experiments. If genes are connected in the network, they are likely to be co-regulated. Using this information gene expression values influence each other to change their measured values. This way we can more easily see the meaningful differences between individual cells. In turn, this helps us move forward with achieving the full potential of single cell expression profiles, which is to understand differences between cell types and molecular states within tissues.

Perspectives