*2.2. Extraction*

The dried flower of *Clitoria ternatea* L. was purchased from a local herbal drug store, Bangkok, Thailand. The extraction was performed according to a previous study [20]. The content of the phenolic compounds and the total anthocyanins in the CTE was 53.08 ± 0.08 mg gallic acid equivalents/g extract and 1.08 ± 0.12 mg delphinidin-3-glucoside equivalents/g extract, respectively.

#### *2.3. Preparation of Flour and Extract*

Flour (0.25 g) was dissolved with 50 mL of boiled water at 100 ◦C and stirred for 10 min. The flour solution was allowed to cool at room temperature for 10 min. Furthermore, the CTE powder was dissolved in 0.1 M phosphate buffer saline (PBS) or 0.2 M sodium acetate buffer. The CTE solution was vortexed for 10 min. The CTE solution was added into the flour solution (final concentration: 0.5%, 1%, and 2% *w*/*v*) and subjected to in vitro digestion.

#### *2.4. Inhibition of Pancreatic α-Amylase*

The activity of pancreatic α-amylase was carried out using a modified procedure of Adisakwattana et al. [19]. Fifty microliters of the flour solution were mixed with 100 μL of CTE in 0.1 M phosphate buffer saline (PBS). Fifty microliters of porcine pancreatic α-amylase (15 U/mL) in 0.1 M PBS, pH 6.9, was then added and the mixture was made to 250 μL with 0.1 M PBS. After incubation at 37 ◦C for 10 min, the reaction was terminated by adding 250 μL of DNS reagen<sup>t</sup> (1% DNS, 0.2% phenol, 0.05% Na2SO3, and 1% NaOH in distilled water) and heated at 100 ◦C for 10 min. Then, 40% potassium sodium tartrate (250 μL) was added to stabilize the color. After cooling at room temperature, the absorbance was measured at 540 nm. The pancreatic α-amylase inhibitory activity was calculated as the percentage inhibition. A control was prepared using the same procedure, replacing the CTE solution with 0.1 M PBS.

#### *2.5.* In Vitro *Starch Digestibility and Predicted Glycemic Index (pGI)*

The in vitro digestion of flour and CTE was performed according to a previous method with some modifications [21,22]. Fifty microliters of the flour solution were mixed with 100 μL of CTE in 0.2 M sodium acetate buffer. The mixture was incubated with 50 μL of porcine pancreatic α-amylase (15 U/mL) and 50 μL of amyloglucosidase (31.25 μg/mL) in 0.2 M sodium acetate buffer, pH 6.0, at 37 ◦C for 180 min. After heating at 100 ◦C for 10 min, for stopping the reaction, the supernatant was measured for the glucose content using a glucose oxidase-peroxidase (GOPOD) kit. The values were plotted a graph and the area under the curve (AUC) was calculated using the trapezoidal rule. The hydrolysis index (HI) was calculated from the percentage of the area under the hydrolysis curve of the sample to the area under the curve of the standard glucose. The predicted glycemic indices (pGI) of the samples were estimated according to the followed equation: pGI = 39.71 + 0.549 HI [4]. A control was prepared using the same procedure, replacing the CTE solution with 0.2 M sodium acetate buffer.

#### *2.6. Estimation of Starch Fraction*

The starch fraction was calculated based on the in vitro starch digestibility of samples. Rapidly digestible starch (RDS) was calculated as the amount of glucose present in the sample at 20 min of the in vitro digestion, whereas slowly digestible starch (SDS) was calculated as the difference between the amount of glucose measured at 120 min and 20 min [21,23]. The undigested starch was calculated as the amount of glucose that was not digested within 120 min. The conversion factor from the glucose to starch was 0.9.
