Proteins do not act alone: they form a shifting web of partnerships that determines how a cell behaves. In a study published in Nature Genetics, Adrian Serohijos and Stephen Michnick of Université de Montréal's Department of Biochemistry and Molecular Medicine, working with doctoral researcher Savandara Besse, former visiting professor Tatsuya Sakguchi and the laboratory of Julie Hussin, set out to chart how genetic mutations ripple through this network, a question the genomics era had largely left open.
The team studied 350 strains of baker's yeast, exposing them to several drugs including antifungals, an antidiabetic and an antipsychotic, and measured nearly one million protein-protein interactions. The most striking result was how often changes in protein partnerships arose from mutations in parts of the genome with no direct link to the proteins concerned. Some of the most influential variants sat in noncoding DNA, the regulatory regions once dismissed as junk, showing that mutations can act at a distance.
The researchers also found that protein relationships vary in strength and that certain proteins act as hubs, meaning a well-chosen subset of interactions can sketch a reliable portrait of the whole cell. The implications reach well beyond yeast: understanding how DNA variation propagates through protein networks could help explain why patients respond so differently to the same medication. Serohijos now plans to repeat the experiment in human cells, a step towards genuinely personalised medicine.