Levi Waldron and international colleagues recently published an article on the control of gene expression. The work was published in the journal PLoS Computational Biology.
Gene expression is a dynamic program by which the information stored in the genome is rendered functional by production and degradation of two types of macromolecules, RNAs and proteins. Messenger RNAs (mRNAs) are templates for proteins; therefore there is an expectation for correspondence between the quantities of mRNAs and proteins. Genome-wide studies indicate a marked discrepancy between them, when considering their steady-state levels or their variations across different conditions. The research team employed linear regression approaches with paired mRNA/protein datasets to develop a model predicting the protein level of a gene from both the mRNA level and the protein levels of RNA binding proteins inferred to bind the mRNA untranslated regions. The results of their analyses restricted the utility of RNA binding proteins to improve accuracy of predicted protein abundance to a small fraction of the total modelled genes, and identified a novel association of the improvement induced by RNA binding proteins with the presence of upstream translation sites. This finding suggests a new avenue of experimental studies aimed at exploring the hypothesis that RNA binding proteins could influence protein abundance by changing the preference for certain translation initiation sites.
According to Professor Waldron, “One of the most vexing problems in cancer biology is why patients with apparently similar tumors can have such different responses to treatment, and understanding where these differences come from. Cancer arises from malfunctioning of our genome, and genomic technologies can give broad looks at aspects of these individual cancer genomes. However we tend to look at molecules that are easier to measure (DNA and RNA), but are not directly responsible for cellular function. Proteins are the tools and building blocks of cells, but they are much harder to measure in high throughput. For this reason, RNA abundance is often used as a proxy for protein abundance. This work quantifies the disconnect between RNA and protein measurements, and develops models to improve the estimation of protein levels from high throughput RNA measurements.”
This study was part of Professor Waldron’s collaborative research as a Fulbright scholar at the University of Trento in Italy last year. He will be hosting a visiting student from there for a month this winter.
Re A, Waldron L, Quattrone A. Control of Gene Expression by RNA Binding Protein Action on Alternative Translation Initiation Sites. PLoS Comput Biol. 2016 Dec 6; 12(12):e1005198. doi: 10.1371/journal.pcbi.1005198. eCollection 2016.