ORIGINAL RESEARCH
Optimized formulation and processing protocol for a
supplementary bean-based composite flour
Catherine T. Ndagire1, John H. Muyonga1, Reddy Manju2 & Dorothy Nakimbugwe1
1Department
2Department
of Food Technology and Nutrition, Makerere University, Kampala, Uganda
of Food Science and Human Nutrition, Iowa State University, Ames, Iowa
Keywords
Antinutrients, common beans, digestibility,
Response Surface Methodology,
supplementary food
Correspondence
Dorothy Nakimbugwe, School of Food
Technology, Nutrition and Bioengineering,
Makerere University, P.O. Box,
7062 Kampala, Uganda.
Tel: 256782246089; Fax: 256414533676;
E-mail: dnakimbugwe@agric.mak.ac.ug
Funding Information
This work was supported by the USAID
funded Pulse CRSP entitled ‘Enhancing the
Nutritional value and Marketability of Beans’
through Research and Strengthening Value
Chain Stakeholders in Uganda and Rwanda.
Received: 11 August 2014; Revised: 1 April
2015; Accepted: 6 April 2015
Food Science & Nutrition 2015; 3(6):
527–538
doi: 10.1002/fsn3.244
Abstract
Protein-energy malnutrition is the most serious nutritional body depletion disorder
among infants and young children in developing countries, attributable to inadequate energy and nutrient intake, partly due to high dietary bulk of weaning
and infant foods. The gruels fed to children are typically of low nutrient and
energy density due to the low flour incorporation rate required for drinking
viscosity. The aim of this study was to develop a nutritious product, based on
common dry beans and other grains, suitable for supplementary feeding. The
optimal processing conditions for desired nutritional and sensory attributes were
determined using Response Surface Methodology. For bean processing, soaking
for 6, 15, or 24 h, germination for 24 or 48 h, and cooking under pressure for
either 10 or 20 min were the independent variables. The processed bean flour’s
total polyphenol, phytic acid and protein content, the sensory acceptability of the
bean-based composite porridge and its protein and starch digestibility were
dependent variables. Based on product acceptability, antinutrients and protein
content, as well as on protein and starch digestibility, the optimum processing
conditions for the bean flour for infant and young child feeding were 24 h of
soaking, 48 h of malting, and 19 min of steaming under pressure. These conditions resulted in a product with the highest desirability. The model equations
developed can be used for predicting the quality of the bean flour and the beanbased composite porridge. Bean optimally processed and incorporated with grain
amaranth and rice flours of a ratio of 40: 30: 30, respectively, resulted into flour
with high energy, mineral, and nutrient density of the final porridge. The composite
is well adaptable to preparation at rural community level. The use of these locally
available grains and feasible processes could make a great contribution to nutrition security in sub-Saharan Africa and other developing countries.
Introduction
In Uganda, infant and childhood malnutrition due to
inadequate energy and nutrient density has been associated with the high viscosity of gruels fed to children
(Kikafunda et al. 2006). Therefore, there is need to
develop less viscous nutrient and energy-dense foods
to supplement infants’ and young children’s diets.
Through blending of common staples and application
of suitable processing procedures, it is possible to develop highly acceptable products with enhanced energy
and nutrient density, from foods commonly grown in
developing countries.
Beans are high in nutrients yet rarely utilized for
porridges to feed young children. On the other hand, the
utilization of nutrients from beans is limited by antinutrients. Most of the porridges served to young children
especially in rural Uganda are made from sole flours yet
blending common staples enhances energy and nutrient
density. The aim of this study was to determine the
optimum formulation and processing conditions for a
bean-based composite flour with reduced levels of total
polyphenol and phytate, high protein and starch digestibility, and a high level of consumer acceptability.
As the common dry bean is highly nutritious, the digestibility of its macronutrients and bioavailability of its
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc. This is an open access article under the terms of
the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
527
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
micronutrients is limited by antinutritional factors such
as trypsin inhibitors, lectins, polyphenols, and phytic acid
(Gibson et al. 2006), whose removal requires appropriate
processing (Uebersax 2006; Esenwah and Ikenebomeh
2008) such as soaking (Adeparusi 2001), germination, and
cooking (Zia-ur and Salariya 2005). Improving sensory
acceptability of supplementary foods is important because
undesirable sensory properties of foods are key dietary
factors affecting energy and nutrient intake of children
(Arimond and Ruel 2004). Cooking generally improves
palatability of foods (Ramakrishna et al. 2006). Blending
(WHO, 2004) and processing, including germination
(Herlache 2007) and starch pregelatinization (Scattergood
and Cunningham 1999) have been used to produce flours
that make high nutrient and energy-dense gruels for infant
and young child diet supplementation.
The aim of the study was also to determine the optimum formulation and optimize processing conditions
for the bean flour. Specific objectives were to reduce
antinutrients (total polyphenol and phytates), improve
protein and starch digestibility of porridge from the developed composite flour, evaluate the pasting properties
of the composite flour and the acceptability of its porridge, and compare the energy and nutrient (protein and
mineral) density of porridge from the bean-based composite flour to those of widely consumed maize and millet
porridges.
Methods
Preprocessing of flours
To process bean flour, common dry bean (var. K131)
was obtained from National Crops Resources Research
Institute (NaCCRI) in Uganda, cleaned by manual sorting,
batch washed three times with potable water and soaked
for 6, 15, and 24 h in water (200% volume for volume).
The soaked beans were drained and spread on a plastic
tray lined with a thick wet cloth then covered with another wet piece of cloth and left to germinate. The germinated beans were rinsed and steamed under pressure
(15 psi at 121°C for 10 or 20 min) using a domestic
pressure cooker. The steamed beans were then spread on
a metallic tray and dried in a fan oven (GALLENKAMP,
Hotbox Oven with Fan, SG93/08/850, UK) at 75°C for
6 h to a final moisture content of 6%. The dried beans
were then milled using a Wondermill grinder (Grain of
Truth Bread Company, Arlington, VA 22922). To process
rice flour, white rice (super brand) was procured from
a retail shop in Kampala city, cleaned by manual sorting,
and milled using the above mill into fine flour. To process
grain amaranth flour, golden colored grain amaranth
(Amaranthus spp.) was procured from farmers in Kamuli
528
district, cleaned by manual sorting and pan-roasted at
250°C for 5 min, to produce a distinct roasted flavor
(Bahika 2008). The roasted amaranth grains were also
milled into fine flour using the above mill.
Determining the optimum incorporation
level of beans, grain amaranth, and rice into
the composite flour
Concept 4 creative software (CREATIVE FORMULATION CONCEPTS, LLC, Annapolis, MD 21401) was
used to determine the optimum level of incorporation
of beans, grain amaranth, and rice flours into the composite to significantly contribute to the protein and
energy requirements of children aged 2–5 years
(1046 kcal/day and 902 kcal/day for male and female
children aged 2 and 3 years and 1742 kcal/day and
1642 kcal/day for male and female children aged 4 and
5 years and, 13 g/day for both male and female children
aged 2 and 3 years and 19 g/day for those aged 4 and
5 years) (Whitney and Rolfes 2002). The common dry
bean with 23.6% protein complimented rice and grain
amaranth of 6.5 and 13.6% protein, respectively (USDA,
2011).
Optimization of the bean flour processing
protocol
The desirability function approach (DFA) was used to
simultaneously optimize the composite porridge’s acceptability, protein and starch digestibility, and the bean
flour’s total polyphenol, phytate, and protein content.
These are very important characteristics of meals fed
to infants and young children. The desirability function
approach method incorporates desires and priorities for
each of the variables and the maximum desirability is
1 (Mepha et al. 2007).
Optimization research design
Each independent variable was varied as shown in Table 1.
A D-Optimal design (Table 2) was used to determine the
influence of three independent variables (soaking time
[X1], germination time [X2], and steaming time [X3]) on
the nutritional quality and consumer acceptability of bean
flour and bean/rice/amaranths composite porridge.
The variables were expressed individually as a function
of the independent variables. The data were fitted to the
following second-order approximating model (eq. 1):
Y = B0 +
k
∑
i−1
Bi Xi +
k
∑
i−1
Bii X2i +
k
∑
Bij Xi Xj + 𝜀
(1)
i−1
i<j
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Table 1. Processing variables and their levels in the D-Optimal design.
Coded variables
Variables
Symbol
Soaking time (h)
Germination time (h)
Steaming time (min)
X1
X2
X3
−1
0
1
6
0
0
15
24
10
24
48
20
Where Y is the response function, ε is the random
error, Bo the center point of the system, Bi, Bii, and Bij
represent the coefficients of the linear, quadratic, and
interactive effects, respectively, and Xi, X2i and XiXj represent the linear, quadratic, and interactive effects of the
independent variables (soaking time, germination time,
and steaming time), respectively.
Optimizing acceptability of composite
porridges
Table 2. D-Optimal design used to optimize levels of soaking, germination, and cooking of beans.
Actual values
Soaking Germination Steaming Soaking Germination Steaming
time
time
time
time
time
time
Run (h)
(h)
(min)
(h)
(h)
(min)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
−1
−1
−1
−1
−1
1
1
1
1
1
1
−1
0
0
0
0
0
0
−1
−1
−1
1
1
−1
−1
1
1
1
0
0
−1
1
0
0
0
0
−1
−1
1
−1
1
−1
1
−1
1
1
0
0
0
0
−1
1
0
0
6
6
6
6
6
24
24
24
24
24
24
6
15
15
15
15
15
15
0
0
0
0
48
48
0
0
48
48
48
24
24
0
48
24
24
24
24
0
Protein digestibility
In vitro protein digestibility was determined using pepsinpancreatin enzyme system (Saunders et al. 1973; Chavan
et al. 2001). Protein content of the sample was determined
with a NITROGEN (N)-ANALYZER (Elementar Americas,
Inc., Mt. Laurel, NJ) before and after digestion and digestibility was calculated using the formula:
%protein digestibility = (A − B)∕(A)
where A is % protein in the sample before digestion and
B is % protein in sample after enzyme digestion.
Porridges were prepared by dissolving 150 g of the beanbased composite flour in 400 mL of cold water, adding
the resulting paste to 450 mL of boiling water, and boiling
for 8 min with constant stirring. Sensory attributes (color,
flavor, thickness, appearance, smell, taste, and texture) of
the bean-based composite flour porridges, from each run
(Table 2) were evaluated using a 9-point hedonic scale (1 =
dislike extremely and 9 = like extremely). The sensory
Coded values
panel comprised of six Ugandan students at Iowa State
University that were familiar with porridge characteristics
and had consented to be part of the study. Protocols for
conducting sensory evaluation in this study were approved
by the Institutional Review Board at Iowa State University.
0
0
20
0
20
0
20
0
20
20
10
10
10
10
0
20
10
10
0
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
Starch digestibility
Starch digestibility was determined using the Megazyme
resistant starch assay K-RSTAR 05/2008 (Megazyme, 2008).
Total polyphenol and phytic acid content
determination
Polyphenol content was measured colorimetrically as
Gallic acid equivalents using the Folin-Ciocalteau colorimetric method (Singleton and Rossi 1965; Memnune
et al. 2009). Phenolic compounds were extracted from
samples using methanol/water/acetic acid solution
(70:30:5), Folin-Ciocalteau reagent was added and
Catechin and Gallic acid equivalents were determined
as absorbance at 765 nm.
Phytic acid content was determined using the anionexchange method as total phosphorus (AOAC, 1999).
Sensory Evaluation of porridges made from
bean-based composite, millet, and maize
flours
Sensory evaluation of the porridge attributes (overall
acceptability, appearance, smell, taste, texture, color,
aftertaste, flavor, and after taste) was done at the School
of Food Technology, Nutrition and Bio-engineering
sensory laboratory. The panel consisted of 75 males
and females aged 18–50 years, familiar with porridge
characteristics. The porridges were rated on a 9-point
Hedonic scale where 9 is like extremely and 1 is dislike
extremely. The panelists who had consented to
529
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
participate in the study also gave additional comments
on the porridges.
Pasting properties of bean-based composite,
millet, and maize flours
Pasting properties of bean-based composite (finely milled
using a WONDERMILL GRINDER, Grain of Truth Bread
Company, Arlington, VA 22922), finely milled millet, and
maize flours (procured from a retail shop in Kampala
city) flours were determined using Rapid Visco Analyzer
(RVA-4, Newport Scientific, Warrie-wood, Australia). Peak
viscosity, breakdown, final viscosity, set back, peak time,
and pasting temperature were read from the pasting profile
with the aid of thermocline for windows software (Newport
Scientific, 1998).
Proximate analysis
The moisture, crude protein, ash, fat, mineral, total carbohydrate, total fat, and gross energy contents of the
bean-based composite flour were determined using standard methods. Moisture content was determined by oven
drying overnight at 98°C (AOAC (Association of Official
Analytical Chemists) 1999); crude protein content with a
NITROGEN (N) -ANALYZER (Elementar Americas, Inc.,
Mt. Laurel, NJ) (USDA, 2009); ash content by igniting
a dried, ground sample in a furnace at 600°C for 2 h to
oxidize all organic matter (AOAC (Association of Official
Analytical Chemists) 1999); gross energy using the bomb
calorimetry method. Fat content was determined using
the Soxhlet method (AOAC (Association of Official
Analytical Chemists) 1999); total carbohydrate using
Colorimetric Quantification of Carbohydrates assay (Wiley
2001); and mineral profile by Hach Method (HACH, 1997)
using the HACH MODEL 21400 DIGESDAHL
APPARATUS, Colorado.
Table 3. Viscosity (cP) and torque (%) of 10% flour rate of the beanbased composite porridge at 55°C using spindle size 63.
RPM
Mean viscosity
Mean percentage torque
0.3
0.6
1.5
3
4222.5 ± 96.0
2228 ± 31.2
1437.5 ± 10.8
EEEE
41.9 ± 0.6
55.48 ± 0.6
80.9 ± 2.6
EEEE
The bean-based composite flour attained this viscosity at 15%, millet
flour at 8% while maize flour at 7% flour (w/v). ± their standard deviation, EEE-error.
child consumption (Mosha and Svanberg 1983) were determined. Energy and nutrient density (protein and minerals: phosphorus, potassium, magnesium, calcium,
sodium, iron, copper, and zinc) of the flours were calculated for flour rates resulting porridge viscosities of
2500–3000 cP.
Data analysis
To optimize the bean flour processing protocol, data were
analyzed by Response Surface Methodology (RSM) procedures using Design-expert statistical software (DX 6.0; StatEase, Inc., MN; 2003). Statistical parameters used to relate
input variables to responses were P-value and R2 of the
models. Data for sensory attributes (appearance, smell, taste,
and consistency/mouth feel) and pasting properties (peak,
breakdown, final, setback viscosities) of the bean-based
composite, millet, and maize porridges and flours, respectively, were compared using the Statistical Package for Social
Scientists 16.0 software program (SPSS software, release 16.0,
SPSS Inc.). Means, standard deviations were determined,
analysis of variance was performed to calculate significant
differences in treatment means and the TUKEY technique
(P < 0.05) was used for separation of means using SPSS.
Results
Nutrient density determination
Porridges of the bean-based composite, millet, and maize
flours were prepared at different concentrations in water
and boiled for 10 min. The porridges were then placed
in a water bath maintained between 54 and 56°C, the
recommended consumption temperature for porridges by
young children (Dawn et al. 1994). The viscosity (centipoises, cP) of the porridges was measured using a
Brookfield Viscometer (Model DVII Rheometer V2.0 RV;
Middleboro, Massachusetts), with spindle number 63 at
a shear rate of 1.5 rpm. These parameters gave the highest torque values compared to other combinations
(Table 3). Flour rates that resulted in porridge viscosities
of 2500–3000 cP, which are suitable for infant and young
530
Different combinations of common dry bean, grain amaranth,
and rice flour exhibited varying protein and energy contents
(Table 4). The formulation with 40, 30, and 30% bean,
grain amaranth, and rice flours was most adequate since it
exhibited relatively balanced protein and energy contents.
The responses of the preprocessed bean flour and the
bean-based composite porridge varied between different
combinations of soaking, germination and steaming time
(Fig. 1, Table 5).
Predictive models
Equation 2 defines the effect of processing on porridge’s
overall acceptability.
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Table 4. Formulations with grain amaranth, bean, and rice flours and
their energy and protein contents as predicted by Concept 4 Creative
Formulation Software.
Rice
Amaranth
Beans
Energy/100 g
Protein (%)
80
70
60
50
10
20
10
30
30
30
10
20
20
20
80
60
10
10
30
20
10
10
20
30
10
20
80
60
40
50
362
363
359
360
366.5
362
329.9
346
353
350
9.2
10.1
11.6
12.3
13.9
14.2
20.9
17.6
15.6
15.92
Starch digestibility = 75.62 + 18.96X2 + 19.87X3 −
9.68X2 X3 − 20.83X23 (R2 = 0.96)
The formulation in bold font was chosen as optimal and used for this
study because its predicted protein and energy content were relatively
balanced.
Acceptability = 7.28 − 0.033X1 + 1.55X3 + 0.164X1 X3 −
1.40X23 (R2 = 0.99)
effects and positive quadratic effect on phytate content
of the bean flour.
Starch digestibility of the bean-based composite porridge
was affected by germination and steaming time of the
bean flour (eq. 5)
(2)
Soaking and steaming time had a negative and positive
linear effect, respectively, on acceptability. Soaking and steaming time had positive interactive effects on acceptability while
steaming time had a significant negative quadratic effect
on acceptability. Steaming time significantly (P < 0.0001)
increased the acceptability of the composite porridge.
Acceptability of the porridges made from cooked beans was
low and did not significantly change with soaking time
with a mean score of 4.40 on a 9-point hedonic scale.
Soaking, germination, and steaming time affected total
polyphenol content of the bean flour (eq. 3).
Total polyphenol = 0.67 − 0.062X1 − 0.047X2 − 0.033X3 −
(3)
0.052X2 X3 + 0.091X12 (R2 = 0.80)
Soaking, germination, and steaming time had a negative
linear effect on the bean flour’s total polyphenol content.
Steaming and germination time had a significant negative
interactive effect while only soaking had a positive significant quadratic effect on the total polyphenol content
of the bean flour. Increased soaking and germination time
significantly reduced polyphenol content of the bean flour
and the model was significant (P = 0.0003).
Equation 4 defines the significant predictive model for
bean flour’s phytates content (P < 0.0001)
(5)
Steaming and germination time of the beans had positive linear effects while germination and steaming time
had negative interactive effect on starch digestibility of
the bean-based composite flour porridge. Only steaming
time had a quadratic effect on starch digestibility of the
bean-based composite flour porridge which was
negative.
The relationship between soaking, germination, and
steaming time of beans and the bean-based composite
porridge’s protein digestibility is represented by equation 6 and resulted in a significant quadratic model
(P < 0.0001)
Protein digestibility = 85.46 + 0.056X1 + 5.59X2 + 4.07X3 −
4.2 × 10−4 X1 X2 + 0.032X1 X3 +
0.51X2 X3 − 0.38X21 − 0.71X22 −
3.63X32 (R2 = 0.98)
(6)
All the three variables (soaking, germination and steaming time of beans) had positive linear effects, only soaking
and germination time had negative interactive and quadratic
effects on the bean-based composite porridge’s protein
digestibility. Positive interactive effects existed between
soaking and steaming time and between germination and
steaming time. Steaming time also had a negative quadratic
effect on protein digestibility.
In equation 7 the predictive model was significant
(P = 0.0001) and showed both linear and quadratic effects of soaking and germination time of beans on its
flour’s protein content.
Protein content = 17 + 0.62X1 + 0.65X2 + 0.76X21 − 0.75X22
(R2 = 0.79)
(7)
Both soaking and germination time had positive linear
effects on protein content while soaking time had a positive quadratic effect and germination had a negative
quadratic effect.
Phytates = 0.33 − 0.098X1 − 0.090X2 − 0.031X1 X2 +
0.048X12 + 0.08X22 . (R2 = 0.92)
(4)
Optimal processing conditions
Soaking and germination time were the only significant
model terms, showing negative linear and interactive
The processing conditions chosen as optimum were those
that resulted in a product with the highest desirability
(0.94) (Table 6). These were: soaking beans for 24 h,
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
531
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Design-Expert® Software
Factor Coding: Actual
Polyphenol
Design points above predicted value
Design points below predicted value
1.01649
(B)
Design-Expert® Software
Factor Coding: Actual
Acceptability
Design points above predicted value
Design points below predicted value
7.66667
0.533
4
8
X1 = A: Soaking time
X2 = C: Steaming time
7
Acceptability
Actual Factor
B: Germination time = 24.00
X1 = B: Germination time
X2 = A: Soaking time
Actual Factor
C: Steaming time = 10.00
Polyphenol
(A)
6
5
4
0.5
0.6
0.7
0.8
0.9
1
1.1
24.00
0.00
8.00
21.00
15.00
10.00
C: Steaming time
5.00
24.00
15.00
32.00
12.00
A: Soaking time
A: Soaking time
9.00
16.00
18.00
24.00
21.00
18.00
15.00
12.00
20.00
0.00 6.00
Design-Expert® Software
Factor Coding: Actual
Phytates
Design points above predicted value
Design points below predicted value
0.78803
(D)
Design-Expert® Software
Factor Coding: Actual
Starch digestibility
Design points above predicted value
Design points below predicted value
86.44
0.230513
0
X1 = B: Germination time
X2 = C: Steaming time
Actual Factor
C: Steaming time = 10.00
Actual Factor
A: Soaking time = 15.00
Phytates
X1 = A: Soaking time
X2 = B: Germination time
0.2
0.3
0.4
0.5
0.6
0.7
0.8
24.00
21.00
18.00
15.00
12.00
A: Soaking time
B: Germination time
6.00 48.00
100
80
Starch digestibility
(C)
40.00
9.00
60
40
20
0
48.00
40.00
15.00
24.00
10.00
16.00
C: Steaming time 5.00
8.00 B: Germination time
9.00
48.00
40.00
32.00
24.00
16.00
20.00
32.00
0.00 0.00
8.00
B: Germination time
6.00 0.00
(E)
Design-Expert® Software
Factor Coding: Actual
Protein digestibility
Design points above predicted value
Design points below predicted value
91.34
(F)
Design-Expert® Software
Factor Coding: Actual
Protein
Design points above predicted value
Design points below predicted value
19.035
71.54
90
85
20
X1 = B: Germination time
X2 = A: Soaking time
19
Actual Factor
C: Steaming time = 10.00
18
80
Protein
Protein digestibility
Actual Factor
A: Soaking time = 15.00
15.54
95
X1 = C: Steaming time
X2 = B: Germination time
75
70
17
16
15
48.00
20.00
40.00
32.00
15.00
10.00
48.00
40.00
18.00
32.00
15.00
16.00
B: Germination time
24.00
21.00
24.00
C: Steaming time
0.00 0.00
24.00
12.00
5.00
8.00
A: Soaking time
16.00
9.00
8.00
B: Germination time
6.00 0.00
Figure 1. Response surface plots showing effect of soaking, germination, and steaming on bean-based composite porridge’s overall acceptability (A),
bean flour’s total polyphenol content (B), bean flour’s phytates content (C), bean-based composite porridge’s starch digestibility (D), bean-based
composite porridge’s protein digestibility (E), and bean flour’s protein content (F).
532
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Table 5. D- Optimal design arrangement and responses.
Variables
Responses
Run
Sk
Gn
St
Tp (%)
Pa (%)
Pt (%)
OA (av.s)
Std (%)
Ptd (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
6
6
6
6
6
24
24
24
24
24
24
6
15
15
15
15
15
15
0
0
0
0
48
48
0
0
48
48
48
24
24
0
48
24
24
24
24
0
0
0
20
0
20
0
20
0
20
20
10
10
10
10
0
20
10
10
0
0.94
0.93
0.86
0.84
0.72
0.64
0.76
0.83
0.531
0.56
0.76
0.78
0.78
0.63
0.56
0.66
0.68
0.67
1.02
0.61
0.63
0.58
0.57
0.48
0.48
0.45
0.24
0.231
0.231
0.29
0.34
0.51
0.27
0.35
0.33
0.38
0.38
0.79
15.54
15.69
15.61
16.80
17.90
17.39
16.10
17.92
18.18
15.56
19.041
17.00
16.48
16.45
16.73
16.95
16.76
17.05
16.62
4.83
4.83
7.33
4.33
7.33
4.00
7.50
4.17
7.671
7.50
7.17
7.17
7.17
7.17
4.33
7.33
7.62
7.33
4.33
1.47
1.16
68.96
61.29
84.54
10.06
55.77
64.91
84.44
86.321
77.36
73.30
70.00
84.91
48.22
75.15
73.87
74.05
0
71.54
72.01
78.53
81.71
90.96
71.88
79.02
82.07
91.341
89.78
84.33
84.12
77.97
90.78
76.44
86.47
86.40
87.01
71.61
Sk, soaking time (h); Gn, germination time (h); St, steaming time (min); Tp, total polyphenol content; Pa, phytic acid; OA, overall acceptability; Std,
starch digestibility; Ptd, protein digestibility.1Desired response value for required product attributes.
Table 6. Optimal solutions.
Number
Soaking
time (hr)
Germination
time (hr)
Steaming
time (min)
Starch
digestibility
(%)
1
2
3
4
24.00
24.00
24.00
24.00
48.00
47.04
42.70
38.77
18.68
18.56
17.77
13.47
87.73
87.60
87.39
89.60
Protein %
Phytates
(%)
Polyphenol
(%)
Acceptability
(mean
scores)
Desirability
18.27
18.30
18.42
18.49
0.22
0.22
0.22
0.23
0.57
0.58
0.61
0.65
7.67
7.67
7.72
7.67
0.9423
0.9410
0.9327
0.9136
germinating for 48 h, and steaming under pressure for
19 min. These processing conditions resulted in bean flour
with low phytate content (0.22%), low total polyphenol
content (0.57%), moderate protein content (18.27%) and
bean-based composite flour with high starch digestibility
(87.73%), high protein digestibility (91.6%), and high
overall acceptability (7.67 on a scale of 1–9).
Proximate compositions (moisture, protein, ash, fat,
gross energy, nonresistant starch, resistant starch, total
starch, total carbohydrates, and minerals (phosphorus,
potassium, magnesium, calcium, sodium, iron, copper,
and zinc)) of the bean-based composite flour were determined and it was found to have substantially good
quantities of energy and nutrients (Table 7).
The mean scores of porridge attributes (color, overall
appearance, smell, taste, flavor, mouth feel, thickness, after
taste, and overall acceptability) on the 9-point Hedonic
scales were compared (Table 8). Overall appearance scores
of the porridges were highest for bean-based composite
and lowest for millet. There was no significant difference
(P > 0.05) between bean-based composite and maize porridges, which in turn were significantly preferred to millet
porridges. Scores for flavor were not significantly different
(P > 0.05) between the bean-based composite porridge
and maize porridge yet it varied significantly between millet
and bean-based composite porridge, with the bean-based
porridge being superior. Overall acceptability scores for
the different porridges were not different (P > 0.05) between the bean-based composite and maize porridges. The
millet porridge’s overall acceptability was significantly different from and inferior to that of both millet and beanbased composite porridges. Overall acceptability for the
bean-based composite porridge was superior to those of
both maize and millet porridges. Mean acceptance scores
for color, smell, taste, texture, after taste and mouth feel
did not differ significantly (P > 0.05) between bean-based
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
533
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Table 7. Nutritional value of the optimal bean-based composite flour.
% composition
Moisture
Protein
Ash
Fat
Nonresistant starch
Resistant starch
Total starch
Total carbohydrates
2900
Porridge viscosity (cP)
Nutrient
3400
5.4
13.2
10.3
10.4
43.7
8.2
52
64.4
2400
BBCP
MLP
1900
MZP
1400
900
Energy
4
9
14
% flour rate
Gross energy
435.8 kcal/100 g (1823.4 kJ/100 g)
Minerals
mg/100 g
Phosphorus
Potassium
Magnesium
Calcium
Sodium
Iron
Copper
Zinc
317
623
157
170
6.8
12.8
2.2
6.3
Figure 2. Variation in porridge viscosities (cP) of the bean-based
composite, millet, and maize porridges at varied flour rates
Table 8. Mean scores and their standard deviations of sensory attributes of the bean-based composite, maize, and millet porridges on a
9-point hedonic scale.
Mean score ± their Standard Deviations1
Attribute
Bean-based
composite porridge
Maize
porridge
Millet
porridge
Overall appearance
Color
Smell
Texture
Taste
Flavor
Mouth feel
After taste
Overall acceptability
7.2 ± 0.8ac
7.3 ± 0.8a
6.7 ± 0.9a
7.2 ± 0.8a
7.3 ± 1.0a
6.7 ± 0.9ac
7.1 ± 0.8a
6.9 ± 0.7a
7.4 ± 0.7a
7.0 ± 0.9a
7.3 ± 1.0a
6.5 ± 0.9a
7.5 ± 0.8a
7.3 ± 0.9a
6.3 ± 1.2a
7.1 ± 1.0a
6.6 ± 1.1a
7.3 ± 0.7a
5.9 ± 1.5ab
6.0 ± 1.5b
5.8 ± 1.4a
5.5 ± 1.5b
5.8 ± 0.9b
5.8 ± 1.0ab
5.7 ± 1.5b
5.3 ± 1.6b
6.2 ± 0.9b
Figure 3. Rapid Visco Analyzer profiles of bean-based composite, millet,
and maize flours.
composite and maize porridge yet were significantly different (P < 0.05) and were superior to those of millet
porridge.
Use of spindle size 63 of the Brookfield viscometer at
1.5 rpm resulted in higher torque values and was therefore
used in the study. The bean-based composite flour
attained viscosity in the range of 2500–3000 at 15%, millet
flour at 8% while maize flour at 7% flour (w/v), (Fig. 2).
On comparison of the bean-based composite, maize
and millet flours’ RVA parameters: peak viscosity, breakdown viscosity, final viscosity, setback viscosity, peak time,
and pasting temperature, the bean-based composite had
the lowest peak, breakdown, final and set back viscosity
(profile COMPOSITE 1 viscosity in Fig. 3, Table 9). These
were significantly lower than those of millet and maize
flour. As the maize flour’s peak viscosity was higher but
not significantly different from that of millet flour
(P > 0.05), its break down, final, and set back viscosities
were higher and significantly different from those of millet
flour (P < 0.05). The results showed that there were significant differences (P < 0.05) in the pasting temperature
of the different flours. The bean-based composite flour
exhibited a significantly higher pasting temperature than
both millet and maize flour. The bean-based composite
flour also exhibited the highest peak time that was not
significantly different from that of millet (P > 0.05). Maize
porridge exhibited a significantly lower peak time
534
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
Figures in a row with the same letter as their first superscript are not
significantly different. Figures in a row with the same letter as their
second superscript are not significantly different (P < 0.05). Rows were
compared to establish how different sensory attributes were between
the three porridges.1Scores 1 – Dislike extremely, 2 – Dislike very much,
3 – Dislike moderately, 4 – Dislike slightly, 5 – Neither like nor dislike,
6 – Like slightly, 7 – Like moderately, 8 – Like very much, 9 – Like
extremely.
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Table 9. Means and standard deviations of pasting properties of the bean-based composite, maize, and millet flours.
Pasting property
Bean-based composite flour
Maize flour
Millet flour
Peak viscosity (cP)
Breakdown viscosity (cP)
Final viscosity (cP)
Set back viscosity (cP)
Peak time (min)
Pasting temperature (°C)
493 ± 19.3a
24.3 ± 3.8a
740.3 ± 17.9a
271.7 ± 2.5a
6.4 ± 0.1a
92.2 ± 0.1a
1986.7 ± 79.7bd
553.3 ± 75.3b
4096.7 ± 138.0b
2663.3 ± 121.6b
5.49 ± 0.04b
76.22 ± 0.5b
1798.7 ± 19.8 cd
336 ± 49.5c
2231.3 ± 38.2c
768.8 ± 43.6c
6.3 ± 0.0a
89.3 ± 0.5c
Figures in a row with the same letter as their first superscript are not significantly different. Figures in a row with the same letter as their second superscript are not significantly different (P < 0.05). Rows were compared to establish how different pasting properties were between the three flours.
Sensory characteristics of a food are key factors in
determination of energy and nutrient intake of infants
and young children as they consume less of foods with
inferior sensory characteristics than those of superior sensory characteristics. Desirable sensory attributes are key
factors for foods as it determines acceptability and consumption. In this study, acceptability of the porridge at
different levels of bean flour processing increased with
increasing steaming time. Cooking improves palatability
of foods (Ramakrishna and Ramakrishna 2006). Improved
palatability of food with cooking could be partly explained
by reduction in phenolic compounds, known to be responsible for the bitterness and astringency of many foods
and beverages (Delcour et al. 1984), by cooking (Habiba
2002). The observed improvement in acceptability with
steaming time could be attributed to reduction in the
levels of polyphenols, which are associated with undesirable beany flavors (Nwosu 2010). Roasting grain amaranth
produced a pleasant flavor (Bahika 2008) playing a big
role as a flavor improver of the porridge.
The optimized processing protocol reduced the levels
of antinutrients in the product. Increased soaking and
germination time led to reduced polyphenol content of
the bean flour. Ramakrishna and Ramakrishna (2006)
reported that malting was a more effective method for
reducing polyphenols than the various cooking treatments.
Decrease in phenolic compounds because of soaking and
germination may be due to decomposition, covalent linkage to structural polymers, or entrapment within the solid
endosperm matrix (Waniska 2000). Reduction in polyphenol content can also be attributed to leaching into
soaking water (Afify et al. 2011). Phytate content of bean
flour decreased with increased soaking and germination
time from 0.79% to 0.23%. The observation agrees with
the findings of (Akpapunam et al. 1996) who reported
a 76 and 59% loss of phytic acid contents of soybean
and bambara groundnut flours, respectively, upon malting
for 120 h. The activity of phytase, an enzyme found in
malt like barley and known to break down phytate increases 7.9-fold during malting (Herlache 2007). The
observed lowered phytate content of soaked and germinated
beans can therefore be attributed to both leaching loss
(Afify et al. 2011) and degradation due to increased phytase
activity.
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
535
Table 10. Nutrient density of 100 mL of different porridges.
Nutrient
Maize
porridge
Millet
porridge
Bean-based
composite
porridge
Energy (kcal/kJ)
Protein (g)
Minerals (mg)
Phosphorus
Potassium
Magnesium
Calcium
Sodium
Iron
Copper
Zinc
25.6/106.9
0.7
29.8
0.9
65.8
2.0
14.7
20.1
8.9
0.5
2.5
0.2
0.02
0.2
22.8
17.9
9.5
1.1
0.3
0.3
0.04
0.2
47.6
93.5
23.6
25.5
1.0
1.9
0.3
1.0
Values calculated for the bean-based porridge are based on chemical
analysis while those for maize and millet porridge are based on nutrient
database values of USDA (2011).
(P < 0.05) than both millet and the bean-based composite
flours. The mean final viscosity values were lowest for
the bean-composite and highest for millet flours.
Energy, protein, and mineral (phosphorus, potassium,
magnesium, calcium, iron, copper, zinc, manganese) densities (as served) of the bean-based composite porridge were
superior to densities of maize and millet porridges except
for sodium that was more in maize porridge than millet
and the bean-based composite porridges (Table 10). Energy
density of the bean-based composite porridge
(65.8/275.2 kcal/kJ/100 mL) was higher than that for millet
and maize porridges (29.8/124.9 and 25.6/106.9 kcal/
kJ/100 mL, respectively). Likewise, the protein density of
the bean-based composite porridge (1.98 g/100 mL) was
higher than that for millet and maize porridges (0.9 and
0.7 g/100 mL, respectively).
Discussion
C. T. Ndagire et al.
Optimized Supplementary Composite Flour
Starch digestibility of the bean-based composite flour
porridge increased with increased steaming and germination time. This may be attributed to action of amylases
during germination and increase in ease with which starch
gets digested after cooking since it breaks down fibrous
cellulose. Alpha- and beta-amylase enzymes present in
malt break down carbohydrates to maltriose, maltose, and
glucose (Herlache 2007). The phytate molecule, containing
six phosphate groups, is highly charged, making it an
excellent chelator and it can form insoluble complexes
with proteins leading to reduced digestibility (Wedad et al.
2008). Magdi (2007) reported that malting significantly
enhanced protein digestibility of Dolichos Lablab Bean.
This was attributed to phytic acid hydrolysis as well as
degradation of protein. Herlache (2007) also reported an
increase in the activity of proteases during germination
of cereals. Increase in protein digestibility can also be
attributed to breakdown of long protein chains by proteases. Protein content of the bean flour apparently increased with increased germination time. Malting has been
shown to yield apparent increases in grain protein content
(Griffith and Castell-Perez 1998). The apparent increase
in protein content resulting from germination has been
linked to shift in dry matter through depletion of carbohydrates during germination (Wedad et al. 2008).
Starch in foods is responsible for their pasting properties thus viscosity. The high flour rate for bean-based
composite porridge is attributable its low starch content
(51.97%), (Table 6). Since millet has lower starch content
(69.88%) (Saunders et al. 1973) than maize (89.3%) (MoraEscobedo et al. 2004), its porridge gave a high flour rate
than of maize porridge. The bean-based composite flour
porridge also had a low viscosity attributable to its low
starch content. Since millet has lower starch content than
maize, its flour gave a lower viscosity than of maize porridge. It is expected that soy-based porridge would have
a much higher flour rate compared to that of the beanbased porridge since soy has much lower starch content
(Birmingham et al. 1978).
Energy and nutrient density of infants and young children’s feeds is a vital factor in their nutritional status as
it determines energy and nutrient intake. The bean-based
composite porridge’s higher nutrient density as compared
to maize and millet (Table 9) is attributable to nutrient
complementation as well as the higher flour rate used.
The higher flour rate was possible because of the lower
starch content. Millet porridge’s nutrient density was higher
than that of maize and this can be attributed to higher
flour rate used for millet porridge than for maize porridge, because of the lower starch content of millet flour.
The bean-based porridge’s higher nutrient density is attributable to its higher flour rate that is attributable to
germination of the beans used for the composite flour
536
during the preprocessing of the bean flour. It can also
be attributed to compositing the bean flour with rice and
grain amaranth flours where different ingredients complimented one another making an energy and nutrientdense formulation. Millet nutrient density was superior
to that of maize porridge and this can be attributed to
higher flour rate exhibited by millet porridge than maize
porridge.
The relatively high viscosity exhibited by maize and
millet starch is indicative that the flours are less suitable
for infant and young child feeding as infant and young
children require less viscous porridges. On the other hand,
the low viscosity exhibited by the bean-based composite
starch is indicative that the flour is suitable for infant
and young child feeding that requires less viscous porridges. The relatively low energy and nutrient densities
exhibited by maize and millet porridges signifies that the
porridges may not be suitable for infant and young child
feeding that requires high energy and nutrient density.
The high energy and nutrient density exhibited by the
bean-based composite porridge shows that the flour is
suitable for infant and young child feeding that requires
high energy and nutrient density.
Conclusions
Soaking, germination, and steaming of bean flour result
in significant improvements in the sensory and nutritional
quality of the flour and increase its suitability for feeding
infants and young children. The optimal processing conditions for dry beans are soaking for 24 h followed by
germination for 48 h and steaming under pressure for
19 min, as they result in high acceptability and protein
and starch digestibility as well as reduction in antinutrients. Surface Response Methodology (SRM) is a suitable
approach to optimizing formulation for complementary
foods and the model equations developed closely predicted
the quality of the bean flour and bean-based composite
porridge incorporating the processed flour. Dry beans,
when optimally processed and composited with other
common grains result in a nutrient-dense complementary
food with the potential to contribute to reduction in
macro and micronutrient malnutrition among infants and
young children.
Acknowledgments
I express my sincere gratitude and appreciation to my
academic supervisors; Dr. Dorothy Nakimbugwe Prof. John
H. Muyonga and Prof. Reddy Manju for their endless
academic support. May God bless them abundantly. This
research was made possible through funding from a USAID
supported Pulse CRSP project.
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
C. T. Ndagire et al.
Conflict of Interest
Optimized Supplementary Composite Flour
Adeparusi, N. E. 2001. Effects of processing on the nutrient
and anti-nutrient of lima beans (Phaseolus lunatus L.)
flour. Nahrung 45:94–96.
Afify, A., H. S. El-Beltagi, S. M. Abd El-Salam, and A. A.
Omran. 2011. Bioavailability of iron, zinc, phytate and
phytase activity during soaking and germination of White
Sorghum varieties. PLoS One 6.
Akpapunam, M. A., S. O. Igbedioh, and I. Aremo. 1996.
Effect of malting time on chemical composition and
functional properties of soybean and bambara groundnut
flours. Int. J. Food Sci. Nutr. 47:27–33.
AOAC (Association of Official Analytical Chemists). 1999.
Official methods of analysis. AOAC International, 16th
ed. AOAC International, Gaithersburg, USA.
Arimond, M., and M. T. Ruel. 2004. Dietary diversity is
associated with child nutritional status: evidence from
11 demographic and health surveys. J. Nutr.
134:2579–2585.
Bahika, R. 2008. Development of an Instant Amaranth
Soup, Suitable for Convalescents. BSc. Project Report,
Makerere University.
Birmingham, V. A., A. Wilson, P. Moon, and H. E. Snyder.
1978. Isolation and characterization of starch from
mature soybeans. Cereal Chem. 55:661–670.
Chavan, U. D., D. B. McKenzie, and F. Shahidi. 2001.
Functional properties of protein isolates from beach
pea (Lathyrus maritimus L.). J. Food Chem.
74:177–187.
Dawn, M. S., J. M. G. Meeks, S. P. Walker, and A.
Ashworth. 1994. Weaning-food viscosity and energy
density: their effects on ad libitum consumption and
energy intakes in Jamaican children. Am. J. Clin. Nutr.
60:465–469.
Delcour, J. A., M. M. Vandenberghe, P. E. Corten, and P.
Dondeyne. 1984. Flavor thresholds of polyphenolics in
water. Am. J. Enol. Viticult. 5:134–136.
Esenwah, C. N., and M. J. Ikenebomeh. 2008. Processing
effects on the nutritional and anti-nutritional contents of
African Locust Bean (Parkia biglobosa Benth.) seed.
Pakistan J. Nutrit. 7:214–217.
Gibson, R. S., L. Perlas, and C. Hotz. 2006. Improving the
bioavailability of nutrients in plant foods at the
household level. Proceed. Nut. Soc. 65:160–168.
Griffith, L. D., and M. E. Castell-Perez. 1998. Effects of
roasting and malting on physicochemical properties
of select cereals and legumes. Cereal Chem.
75:780–784.
Habiba, R. A. 2002. Changes in anti-nutrients, protein
solubility, digestibility, and HCl-extractability of ash and
phosphorus in vegetable peas as affected by cooking
methods. Food J. Chem. 77:187–192.
HACH, 1997. Digesdahl Digestion Apparatus. Available at
http://www.fr.hach-lange.be/shop/action_q/login_
light%3Bregister/DOK_ID/14407/type/pdf/lkz/BE/spkz/fr/
TOKEN/TnNu2zMazb70rhdVUiOZirxVSD4/M/dz2cSA
(Accessed 24 July 2010).
Herlache, J. L. 2007. Power flour (High Diastatic Milled
Barley Malt). Its important and critical role in the care
of weanling infants and the severely malnourished. Perin
Press Inc., Wisconsn.
Kikafunda, J. K., L. Abenakyo, and F. B. Lukwago. 2006.
Nutritional and sensory properties of high energy/nutrient
dense composite flour porridges from germinated maize
and roasted beans for child-weaning in developing
countries: a case for Uganda. Ecol. Food Nutrit.
45:279–294.
Magdi, A. O. 2007. Effect of different processing methods,
on nutrient composition, antinutrional factors, and in
vitro protein digestibility of Dolichos Lablab Bean
(Lablab purpuresus (L) Sweet). Pakistan J. Nutrit.
6:299–303.
Megazyme, 2008. Resistant Starch Assay Procedure.
Megazyme International Ireland Limited. (pdf). Available
at http://www.megazyme.com/booklets/KRSTAR.pdf
(Accessed June 12 2010).
Mepha, H. D., L. Eboh, and S.U. Nwajigwa. 2007. Chemical
composition, functional and baking properties of
wheat-plantain composite flours. Afr. J. Food Agric. Nutr.
Dev. 7:1684–5358.
Mora-Escobedo, R., P. Osovio, M. I. Garcia-Rosas, A.
Bello-Perez, and H. Hernandez-UNzon. 2004. Changes in
selected nutrients and microstructure of white starch
quality maize and common maize during Tortilla
preparation and storage. Food Sci. Technol. Int.
10:79–87.
Mosha, A., and U. Svanberg. 1983. Preparation of weaning
foods with high nutrient density using flour of
germinated cereals. Food Nutrit. Bullet. 5:10–14.
Newport Scientific, 1998. Applications manual for the rapid
viscotm analyzer using thermocline for windows. Newport
Scientific Pty Ltd., Australia 2–26.
Nwosu, J. N. 2010. Effect of soaking, blanching and
cooking on the anti-nutritional properties of Asparagus
Bean (Vigna Sesquipedis) Flour. Nat. Sci. 8:8.
Ramakrishna, V., R. P. Jhansi, and R. P. Ramakrishna.
2006. Anti-nutritional factors during germination in
Indian Bean (Dolichos lablab L.) seeds. World J. Dairy
Food Sci. 1:06–11.
Ramakrishna, V., and R. Ramakrishna. 2006. Storage protein
degradation in germinating Indian bean (Dolichos lablab
L. var lignosus) seeds. Seed Sci. Technol. 34:161–168.
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.
537
Catherine T. Ndagire, John H. Muyonga, Reddy Manju,
and Dorothy Nakimbugwe do not have any conflicting
interests.
References
Optimized Supplementary Composite Flour
C. T. Ndagire et al.
Saunders, R. M., M. A. Connor, A. N. Booth, M. M.
Bickoff, and G. O. Kohler. 1973. Measurement of
digestibility of alfalfa protein concentrate by in-vivo and
in-vitro method. J. Nutr. 103:530–535.
Scattergood, L. K., and R. C. Cunningham. 1999. Evaluation
of a Partially Pregelatinized Starch in Comparison with
Superdisintegrants in a Direct-Compression
Hydrochlorothiazide Formulation. Poster.
Sengul, M., H. Yildiz, N. Gungor, B. Cetin, Z. Eser, and
S. Ercisli. 2009. Total phenolic content, antioxidant and
antimicrobial activities of some medicinal plants. Pak J
Pharm Sci. 22:102–106.
Singleton, V. L., and J. A. Rossi. 1965. Colorimetry of total
phenolics with phosphomolybdic-phosphotungstic acid
reagents. Am. J. Ecol. Viticult. 16:144–158.
Uebersax, M. A. 2006. Dry Edible Beans: Indigenous Staple
and Healthy Cuisine (Report), Published by the Forum
on Public Policy.
USDA (United States Department of Agriculture), 2009.
United States Department of Agriculture. Food Safety
and Inspection Service, Office of Public Health Science.
Protein Determination by Combustion. (pdf). Available at
http://www.fsis.usda.gov/PDF/CLG_PRO_4_03 (Accessed
20 June 2010).
USDA (United States Department of Agriculture), 2011.
National Nutrient Database for Standard Reference,
Release 24. Available at: http://ndb.nal.usda.gov/ndb/foods/
list (Accessed 10 January 2011).
Waniska, R. D. 2000. Structure, phenolic compounds, and
antifungal proteins of sorghum caryopses, 72-106 in
Technical and institutional options for sorghum grain
mold management: proceedings of an international
consultation, 18–19 May 2000, Washington, DC.
2004:1–5.
Wedad, H., A. H. Abdelhaleem, I. M. Abdelmoneim, and E.
B. Elfadil. 2008. Effect of fermentation, malt-pretreatment
and cooking on antinutritional on protein digestibility of
sorghum cultivars. Pakistan Journal of Nutrition
7:335–341.
Whitney, E., and R. S. Rolfes. 2002. Understanding Nutrition.
Tenth edition. 9th ed. Peter Marshall, publisher, Belmont,
USA.
WHO (World Health Organization), 2004. Guiding principles
for feeding infants and children during emergencies.
(pdf). Available at http://www.who.int/nutrition/
publications/guiding_principles_feedchildren_emergencies
(Accessed 15 July 2011).
Wiley, J., 2001. Colorimetric Quantification of
Carbohydrates. (pdf). Available at (http://www.nshtvn.org/
ebook/molbio/Current%20Protocols/CPFAC/fae0101.pdf)
(Accessed 20 February 2012).
Zia-ur, R., and A. M. Salariya. 2005. The effects of
hydrothermal processing on antinutrients, protein and
starch digestibility of food legumes. Int. J. Food Sci.
Technol. 40:695–700.
538
© 2015 The Authors. Food Science & Nutrition published by Wiley Periodicals, Inc.