Soybean (Glycine max (L.) Merrill) is a highly nutritious legume with enormous potential to improve dietary quality for humans and livestock. However, the development of varieties with improved nutritional traits has been affected by the negative correlation that exists among the different traits and the high cost of the phenotypic assessment. The objectives of this study were: (1) to quantify the total protein, total oil and fatty acids of 52 soybean genotypes from different sources, (2) to identify correlations among total protein, total oil content and fatty acids. The total protein content was determined using the Modified Folin-Lowry Method. In contrast, the total oil and fatty acids methyl esters were determined using the chloroform/methanol gravimetric method and Gas Chromatography–Mass Spectrometry. The analysis of variance revealed that the studied traits varied significantly depending on genotypes and origin. Total protein content ranged from 30.07% to 50.40 %, while total oil content ranged from 14.94% to 23.48%. Total oil content varied significantly between origins with genotypes from the USA having the highest mean of 20.43%, while those from AVRDC had the lowest mean of 18.32 %. Palmitic acid (16:0) content ranged from 10.58% to 21.18%; 4.93% to 22.20% for stearic acid (18:0), 22.69% to 39.95% for oleic acid (18:1) and 30.60% to 51.72 % for linoleic acid (18:2). Genotypes from Uganda had the highest percentages of oleic acid, followed by genotypes from Japan. Six negative and three positive correlations were found to be significant in the current study further. The current study identified soybean genotypes with elevated protein and oil content above the average that can be used to improve the nutritional properties of soybean in Uganda and across the East African region hence boosting the soybean industry.