Inference on autoregulation in gene expression with variance-to-mean ratio

Area of Science:

• Molecular Biology
• Systems Biology
• Computational Biology

Background:

• Gene regulation is crucial in biology, but gene autoregulation remains understudied.
• Detecting autoregulation biochemically is challenging.
• Existing research links autoregulation to gene expression noise.

Purpose of the Study:

• To develop a robust and simple method for inferring gene autoregulation.
• To overcome limitations of existing biochemical and computational approaches.
• To identify novel instances of gene autoregulation from expression data.

Main Methods:

• Generalizing existing findings using discrete-state continuous-time Markov chains.
• Proposing a method based on comparing the mean and variance of gene expression levels.
• Utilizing non-interventional, one-time gene expression data without parameter estimation.

Main Results:

• A novel, robust method for inferring autoregulation was established.
• The method requires only mean and variance comparisons of gene expression.
• Applied to four datasets, the method identified potential autoregulating genes.

Conclusions:

• The proposed method offers a simplified, non-interventional approach to detect gene autoregulation.
• Identified autoregulations show promise, with some already supported by external evidence.
• This work facilitates further research into the understudied field of gene autoregulation.