The title of the paper by a team from places as widespread as the University of Chicago, Cairo University and the Argonne National Laboratory gives the answer away, so I won’t repeat it here. The researchers mined genetic databases, including 2000 genomes and a large number of gene tags to predict which genes the most widespread in nature.
The answer is a bit unexpected, although there is a subtlety in the definition of the problem: it’s not the gene that results in the most protein, but the one best at “spreading its DNA around”.
