Flowchart outlining the acquisition of breast milk samples into sets for discovery, interrogation and validation.

<p>Each of these cohorts has been previously described and well-characterized <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0050564#pone.0050564-Maningat1" target="_blank">[7]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0050564#pone.0050564-Mohammad1" target="_blank">[27]</a>, <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0050564#pone.0050564-Mohammad2" target="_blank">[32]</a>. Briefly, in the discovery set lean healthy women received weight-based recombinant human growth hormone for synchronization at 24 hours of study initiation for synchronization. The women continued to collect breast milk via pump every 3 hours for an additional 48 hours, and every 4 hours thereafter for a total of 96 hours studied. Breast milk samples used for the discovery set were collected at hour 24 and hour 96 (2 samples from each of 3 women, for a total of 6 samples). To both confirm and ascribe potential functional significance, two dietary manipulations were additionally employed in the validation set. These two cohorts were part of two crossover studies (whereby each subject serves as her own control following random sequence allocation) examining the effects of diet composition on milk quality, quantity, and carbohydrate expenditure. In the glucose-galactose cohort <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0050564#pone.0050564-Mohammad1" target="_blank">[27]</a>, 7 obese subjects consumed a diet of 15% protein, 50% carbohydrate, and 35% fat under controlled GCRC conditions. After an overnight fast, women were randomized to receive an isocaloric drink every 3 hours comprised of either glucose or galactose, the sugars in which provided them with 60% of their daily estimated energy requirement. In the high carbohydrate-high fat cohort <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0050564#pone.0050564-Mohammad2" target="_blank">[32]</a> 7 lean subjects were randomly assigned to either a high-carbohydrate, low fat diet (60% carbohydrate, 25% fat, and 15% protein), or 2) a low-carbohydrate, high-fat diet (30% carbohydrate, 55% fat, and 15% protein), the caloric value of which was tailored to the woman’s required total energy intake (1.3 times basal metabolic rate). Breastfeeding occurred every 3 hours followed by milk expression via electric pump.