NPR writes that despite ethical discussions over the matter, generating "designer babies" is not that easy to do.
It notes that when researchers first set out to sequence the human genome, they'd hoped to easily pinpoint what variants in which genes led to particular diseases, but reality turned out to be messier: gene variants often seem to have small effects on disease risk. For instance, Stanford University's Jonathan Pritchard and his colleagues present in Cell this week a model that suggests that the heritability of most complex traits is due to the interaction of many weak trans-eQTL SNPs in genes distantly related to the trait at hand.
NPR notes that not everyone agrees with Pritchard's model: the European Bioinformatics Institute's Ewan Birney tells it that the model seems to suggest genetics is a "mush," though he hopes that with more knowledge, it will make more sense.
Still, he adds that "[i]f anybody thinks we can understand how to change genomes to improve things, they don't have an appreciation for the lack of knowledge that we have."