The locations from which Sclerocarya birrea fruits were collected for this study were:

Pre-treatment of the samples and determination of the physicochemical and nutritional parameters were carried out in the laboratory of food technology department of IRSAT/CNRST laboratory in Burkina Faso.

Sclerocarya birrea ripe fruits were collected in the North (ND), Hauts-Bassins’s (HB) and Mouhoun loop’s (ML) regions of Burkina Faso between May and July, 2018.

In total, eleven (11) batches of ripe fruit samples, of two (02) kilograms each, were collected in the three (03) regions. They were collected directly from the feet of the trees, put in plastic baskets with lids, labeled and then transported to the laboratory where they were thoroughly washed in tap water and left to dry in a plastic basket. Using a mortar and strainer, the nuts were collected and dry in the shade before the kernels were extracted by manual crushing. The kernels obtained were crushed using a porcelain mortar, packed in plastic jars, coded and kept in the refrigerator at 4˚C before analysis. The oil samples were obtained in the laboratory by extraction with organic solvent (n-hexane) and by direct sampling during the hydraulic press extraction sessions at the Union-Yanta center. Figures 2-5 show the fruit, nuts, kernels and oil of Sclerocarya birrea during sample pre-treatment, respectively.

Moisture, total ash, total fat, protein, total sugars and energy of Sclerocarya birrea kernels were determined using the official methods of analysis described by AOAC International, 16th edition [8] .

Total amino acid profiles were determined by reverse phase HPLC using the Waters Pico-Tag method [9] and amino acid values were calculated [10] . Minerals in Sclerocarya birrea kernel samples were determined by flame atomic absorption spectrometry using the international AOAC method [11] .

The Abbe refractometer was used to determine the refractive index of S. birrea oils according to the official methods of the American Oils Chemists’ Society [12] , which were used to determine the saponification and iodine indices.

The data were analyzed with XLSTAT2014.5.03 software. The calculation of the means and standard deviations of the various physicochemical and nutritional characteristics carried out in triplicate was done with the “mean” and “standard deviation” functions of the XLSTAT2014.5.03 software. The analysis of variance (ANOVA), was used to test the significant differences between the means of the parameters determined from the different sample. For P < 0.05, the difference was considered significant.

The proximate composition of S. birrea kernel is given in Table 1 . The results of analysis of dried nut kernels showed 4.02% ± 0.47% moisture with 4.47 ± 0.18 g ash; 29.50 ± 2.19 g protein; 55.78 ± 2.62 g fat; 10.25 ± 1.02 g carbohydrates and 661.02 ± 12.85 Kcal energy value per 100 g dry matter (DM). M L and HB samples showed the highest fat and moisture contents with high energy values, while the highest protein and carbohydrate contents were obtained with seeds from ND samples. total ash content of S. birrea kernels was relatively similar in the three regions ( Table 1 ).

DM = Dry Mater. P-value > α = 5%, so no significant difference between proximal parameter values in the same column.

Analysis of variance (ANOVA) at the significance level α = 5% showed that there was no significant difference between S. birrea kernels from the three regions in terms of proximate composition and energy value (P-value > α).

Moisture contents in the present study are lower than those obtained by [6] (10.47% ± 0.33%) in Nigeria, but similar to those obtained by [13] (4.17% ± 0.23%) in Ghana. These authors obtained lower carbohydrate contents (9.76% ± 0.06%) than in the present study. The different drying methods for fresh S. birrea nuts and agroclimatic conditions in different study areas could explain the difference in moisture content. The ash contents were similar to those obtained by [6] and [13] (4.67% ± 0.15% and 4.63% ± 0.08%, respectively), but lower than those obtained by [14] (5.43%) using kernels of the same species. The protein and fat contents obtained by [6] (30.97% ± 1.17% and 58.88% ± 0.59%, respectively) are higher than those obtained in the present study. Another study carried out in Niger by [15] showed higher average protein (35.23%) and carbohydrate (14.77%) contents with lower average fat content (46.03%) than in the present study. This difference could be explained by differences in agro-climatic conditions. It could also be explained by variability of genetic factors of plant. Our results for fat content are similar to those obtained (51.30% - 56.23%) for almonds of the same species from the Centre and Cascades regions [16] . They are higher than those obtained (47.5%) by [17] from almond oils of S. birrea subsp. Caffra from different countries in South Africa (Malawi, Mozambique, Zambia and Zimbabwe). This shows that S. birrea subsp. Caffra is less rich in fat than the S. birrea subsp. birrea subspecies of the present study.

Protein contents of Lannea kerstingii (26.39%), and Lannea microcarpa (21.14%) kernels [18] [19] are lower than S. birrea kernel of the present study. On the other hand, fat content is lower than that of Lannea kerstingii (57.85%) and Lannea microcarpa (64.90%) kernels [18] [19] . Another study on Balanites aegyptiaca seed showed lower protein (28.28%) and fat (55.78%) contents than the present study [20] . These results show that S. birrea kernel is a potential source of protein, lipids and calories. It is also a potential raw material for the oil industry, the production of energy bars and ingredients for seasoning sauces like peanut powders and pastes.

Table 2 shows amino acid content of S. birrea seeds. A total of seventeen (17) alpha-amino acids were detected. S. birrea seeds contain mainly glutamic acid (16.88 ± 1.55 g/100g), arginine (7.47 ± 0.86 g/100g) and proline (6.88 ± 0.72 g/100g). These values are lower than those found in Nigeria by [6] who reported that glutamic acid (28.09%), arginine (14.99%) and aspartic acid (13.35%) were the major amino acids in S. birrea seeds. They are higher than the values obtained by [14] , who also confirmed the predominance of glutamic acid, arginine and aspartic acid in S. birrea kernels compared to the other amino acids. The differences observed could be explained by the differences in agro-climatic conditions in study areas or by genetic variability of plants.

These results show that S. birrea kernel is a potential source of glutamate, which is often used to enhance the flavour of foods. In nutritional terms, this molecule is one of the most important neurotransmitters in the brain and aids memory and learning [21] [22] . It would be advisable for students to consume S. birrea almonds on a regular basis.

An estimate of essential amino acid content shows that the protein of S. birrea consists of 19.98% ± 0.24% essential amino acids. This percentage is lower than that obtained by [6] (35.01%). These differences could be due to the agro-climatic conditions of the study areas or to the genetic variability of the plants. Essential amino acid values in relation to the FAO recommendations show that 100 g of S. birrea almond protein is likely to meet the nutritional requirements for the aromatic amino acids, Phe + Tyr (value = 142.67%) as well as for histidine (value = 107.33%). The other essential amino acids are present in limited amounts in 100 g of S. birrea almond protein [10] .

Mineral composition of S. birrea seeds in the three regions of Burkina Faso is given in Table 3 . The values obtained show that S. birrea kernels contain an average of 259.64 mg calcium, 911.30 mg potassium, 443.96 mg magnesium and 35.38 mg sodium per 100 g dry matter as major minerals, with an Na/K ratio of 0.04. The highest values were found from ND fruit kernels, while the lowest values were obtained from M L and HB fruit kernels, which showed almost similar values of major mineral elements ( Table 3 ).

P-value < α = 5%, so the difference between the mineral values in the same column is significant.

The kernels of the collected S. birrea fruits also contained minor mineral elements such as iron (3.69 mg/100g DM) and zinc (7.43 mg/100g DM). The highest values were found in the kernels of ND fruits, while the lowest values were found in those of HB fruits ( Table 3 ).

Analysis of variance (ANOVA) at α = 5% significance level showed a significant difference between the mineral contents of S. birrea kernels from the three regions (P-value < α).

Calcium and Fe contents in S. birrea kernels from the present study are lower than those obtained from the same species in Nigeria by [6] (403.08 ± 1.97 mg/100g DM and 27.49 ± 0.11 mg/100g DM, respectively). These authors reported lower levels of K (366.00 ± 2.00 mg/100g), Mg (206.14 ± 3.14 mg/100g), Na (4.76 ± 0.47 mg/100g) and Zn (3.29 ± 0.12 mg/100g DM). This difference in mineral content could be explained by the different agro-climatic conditions in the study areas. The calcium, potassium, magnesium, sodium, iron and zinc contents of the present study are higher than those of Lannea kerstingii (78.33 mg/100g; 674.18 mg/100g; 317.15 mg/100g; 2.48 mg/100g; 4.46 mg/100g and 6.34 mg/100g, respectively) according to [18] . S. birrea seeds are therefore a potential source of major and minor mineral elements. This is a great advantage for the consumer, especially for the strengthening of bones and teeth in children by Ca, respiration and cellular energy metabolism with Fe, heart rate regulation and insulin production by Mg. The low Na/K ratio (< 1) shows that the consumption of Sclerocarya birrea kernels could be an asset in reducing blood pressure and contributing to the reduction of the risk of cardiovascular diseases [23] - [25] .

The results of physico-chemical parameters of S. birrea oil are given in Table 4 . S. birrea oil gave an average refractive index of 1.47; saponification value 192.25 mg KOH/g oil and iodine value 24.97 g I₂/100g oil. All the refractive indices are almost similar in the almond oils from the three locations. The highest iodine and saponification indices (28.02 g I₂/100g, 200.86 mg KOH/g oil) were found with the S. birrea kernel oils from HB and ND respectively, while the lowest indices (23.26 g I₂/100g, 183.89 mg KOH/g oil) were found with the oil sampled from Union-Yanta and that from kernels from BM ( Table 4 ).

Analysis of variance (ANOVA) at the significance level α = 5% showed that there was no significant difference between the oils from S. birrea seeds from the three regions in terms of refractive index and iodine value (P-value > α).

P-value > α = 5% (no significant differences between values in the same column) and P-value < α = 5%, (significant differences between values in the same column).

The refractive indices are similar to the specifications of the Codex Alimentarius [26] for palm oil (n20 = 1.46 to 1.47). They are also similar to those reported in Nigeria by [27] (n20 = 1.46), but higher than those found in Ghana by [13] for raw almond oil (n20 = 1.37) and roasted almond oil (n20 = 1.38). The difference with the refractive index values of these authors could be explained by the variation in agro-climatic conditions and whether the samples were pretreated or not.

The saponification index values are close to the Codex Alimentarius specifications [26] for palm oil (Is = 190 to 209 mg KOH/g oil). They are higher than the values published in Nigeria by [27] (178.60 mg KOH/g). This difference could be explained by the variation in agro-climatic conditions, but also by the genetic variability of the species. These saponification index values show that S. birrea oil has good saponification power and is mainly composed of fatty acids (FAs) with relatively low molecular weights [28] - [30] .

The iodine index values are all higher than those published in Ghana (19.56 ± 0.02 g I2/g oil) by [13] for raw S. birrea almond oils. However, they are lower than the Codex Alimentarius specifications [26] for palm oil (Ii = 50.0 to 55.0 g I2/g oil). These iodine index values obtained in the present study are higher than specifications for palm kernel oil (Ii = 14.1 to 21.0 d’I2/g oil) according to [26] . The iodine indices show that the fatty acid molecules of S. birrea oil in the present study have little unsaturation. As a result, this oil is stable at room temperature [29] .