Manual pellet press

The energy (kcal/g) of each test meal was determined using bomb calorimetry. Each test meal was prepared identical as if it were to be served to a participant. Meals were homogenized using a food-grade commercial blender (Model HBH450, Hamilton Beach Commercial, Glen Allen, VA) on high setting for 1 min and then passed through a 2 mm sieve (Advantech 2.00 mm USA standard testing sieve No. 10, New Berlin, WI). Aliquots were weighed and then lyophilized at −55 °C using a Uni-Trap Model 10–100 (The Virtis Company, Gardiner, NY). Lyophilized samples were ground and passed through a 1 mm screen before being pelleted into duplicate ~1.00 g pellets using a manual pellet press (Parr Instrument Co., Moline, IL). Pellets were placed within a model 1108 oxygen bomb (Parr Instrument Co.) and bombed in a 6200 Isoperibol calorimeter (Parr Instrument Co.). Each test meal was analyzed in duplicate with an accepted CV of less than 2 % between duplicate samples. Calorimetry standard 1.00 g benzoic acid pellets (Parr Instrument Co.) were used to calibrate each bomb before the analysis of test meal pellets.

Following miniaturization, functionality as a platform for performing chemical assays was confirmed by performing a glucose assay on a miniaturized microPAD with a sample zone, a reagent zone, a test zone and a waste zone all connected in series by a straight channel (Fig. S1)^{41 (link)}. The reagents for the assay were deposited onto the reagent zone using a reagent pencil, which was fabricated by pressing a mixture of 66.6% w/w polyethylene glycol (M_n 2000 g/mol), 22.2% w/w graphite powder, 0.75% w/w glucose oxidase (GOx, 266 U/mg), 0.52% w/w horseradish peroxidase (HRP, 293 U/mg), and 10.0% w/w 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) into the shape of a cylindrical pellet with a diameter of 3.2 mm using a manual pellet press (Parr Instrument Company)^{41 (link),42} . Glucose solutions (3.5 µL) prepared in 1X PBS with concentrations of 0, 0.3, 0.6, 0.9 and 1.2 mM were applied to the sample zone of the device and a colorimetric readout was generated in the test zone. The intensity of the color produced in the test zones was measured via digital image colorimetry (DIC)^{43 (link)}, where the mean color intensity in the red channel of the test zones was measured using a smartphone (Samsung Galaxy Note 4) and the Color Grab application^{9 (link)}.