Канадский ученый призвал пересмотреть оценку качества соевого шрота: сырого белка недостаточно для определения питательной ценности

Soy Connext 2026 - The crude protein fallacy: Rethinking how soybean meal quality is measured Crude protein as a measure of digestible amino acid content just doesn’t make sense, but a Canadian researcher is working with USSEC to find a solution Editor’s note: Dr. Dominique Bureau presented at Soy Connext 2026, hosted by the U.S. Soybean Export Council, in Chicago August 5-7, 2026. More than 800 U.S. Soy buyers, sellers, growers and forward-thinkers from around the world came together to exchange ideas, open markets, and move the business forward. From the latest research and sustainability strategies to real-world trade opportunities and supply chain solutions, every session, conversation and connection was designed to create momentum for the global soybean industry. Soybean meal may meet a feed mill’s specifications for crude protein but still fail to deliver the nutritional value expected by a nutritionist. That disconnect is at the heart of the challenge Dominique Bureau, Ph.D., professor at the University of Guelph, sees in evaluating soybean and soybean meal quality. Physical damage before processing, excessive heat during drying or storage, and overprocessing at the crushing plant can alter proteins and amino acids in ways that conventional quality-control methods do not necessarily capture. At the same time, heat cannot simply be avoided. Thermal treatment is an essential part of producing nutritionally valuable soybean meal because it reduces antinutritional factors, particularly trypsin inhibitors. The challenge, Bureau explained during Soy Connext 2026, is finding the processing “sweet spot”: enough heat to improve nutritional value without crossing the threshold where heat begins damaging proteins and reducing amino acid availability. “Heat treatment of the bean is extremely important and is very beneficial,” Bureau said. “However, it’s a little bit of science and art – you have to find the sweet spot.” Understanding where quality is gained and lost requires looking beyond a soybean meal’s crude protein content and examining the entire chain, beginning with the physical condition of the soybean. Physical quality can signal nutritional quality For nutritionists, characteristics such as soybean color, size and the percentage of broken beans may initially seem separate from feeding value. A broken bean is still a bean, after all. But Bureau said physical characteristics can provide an important warning that something has happened to the soybean during production, harvest, drying, storage or transportation. Damage can take many forms, including moldy, sprouted, green, burnt or split beans. The visible damage itself may not always be a direct nutritional problem. Instead, it can be evidence that the beans experienced conditions capable of altering their chemical and nutritional characteristics. Bureau pointed to research conducted with partners in Vietnam in which one ton of Brazilian, Argentinian and U.S. soybeans were purchased and characterized bean by bean for size, color and damage. Brazilian soybeans had more physical damage, including smaller, darker and more broken beans. U.S. soybeans were larger and brighter with substantially less damage, while the Argentinian soybeans generally fell between the two. The researchers then asked the more important question: Did those physical differences translate into nutritional differences? The soybean samples were crushed to produce soybean meal, incorporated into tilapia diets and evaluated on nutrient digestibility and animal performance. The research indicated that they did translate into nutritional differences. Protein digestibility was 85% for the Brazilian soybeans compared with 91% for the U.S. soybeans, with Argentina again falling between the two. Differences were also observed regarding energy and other nutrients. “Physical damage to the bean represents a sign or a forewarning that something has happened to the bean,” Bureau said. Animal performance provided another layer of evidence. When graded levels of protein from the three soybean origins were fed to tilapia, differences emerged in feed efficiency, with U.S. soy supporting better results and Brazilian soy producing lower results. Over-formulation can provide enough excess nutrients to mask some differences among ingredients. But as ingredient costs encourage nutritionists to formulate closer to minimum requirements, differences in digestibility become more consequential. “If you're formulating on the edge – at the limit, then that's where you see the difference,” Bureau said. Soybean meal hides the visual clues Identifying damage becomes more difficult once soybeans are processed into soybean meal. With whole soybeans, buyers can see differences in color, size and physical damage. During processing, however, beans are dehulled, flaked, heat treated, oil is extracted and the remaining material is ground into meal. Thus, many of the visible indicators disappear. Quality assessment then moves to the laboratory, where different soybean meals may appear remarkably similar based on conventional analyses of crude protein, fat and amino acid concentrations. In Bureau’s example, the poorer-quality Brazilian beans produced meal with the highest crude protein concentration. This creates what Bureau described as the “crude protein fallacy.” Crude protein analysis does not directly measure protein. It measures nitrogen and then uses a conversion factor to estimate crude protein. The commonly used calculation multiplies measured nitrogen by 6.25, based on an assumption that amino acids contain, on average, 16% nitrogen. Bureau argued that the assumption does not adequately reflect the true composition of many feed ingredients. More importantly, nitrogen can remain present even when nutritional value has deteriorated. “We are basing our assessment on a 160-year-old technique that is basically a rule of thumb. It’s important to understand that crude protein is a shortcut, but it doesn't represent nutritive value,” Bureau said. “We're trading ingredients and adjusting the price based on crude protein, and it doesn't make any sense. We need to do better.” Nitrogen measured in a feed ingredient can also come from sources other than amino acids, including nucleic acids, creatine, uric acid, urea and ammonium salts. Bureau said analyses and mass-balance calculations conducted by his group suggest about 15% of nitrogen in many feed ingredients, with a range of roughly 5% to 25%, may come from compounds other than amino acids. That distinction matters because animals require available amino acids, not an analytical crude protein number. Amino acid content does not necessarily equal availability Even measuring amino acid concentrations does not completely solve the problem. A laboratory analysis can indicate that soybean meal contains adequate lysine, methionine and other amino acids, but it does not necessarily show whether the animal can digest and absorb them. Heat damage can change the chemical structure surrounding an amino acid or bind it to another molecule. Laboratory procedures may still detect the amino acid because analytical methods use harsh acid hydrolysis capable of breaking bonds that the animal’s digestive system cannot. As a result, the analytical amino acid profile can look acceptable while biological availability is reduced. “Your amino acid profile can look great,” Bureau said. “But to the animal, it's not great because it cannot extract and digest the amino acid then absorb it in a form that it can use.” This difference between concentration and availability is critical for precision formulation. Bureau said research sponsored by the U.S. Soybean Export Council has identified differences approaching 8 to 10 percentage points in digestibility. If diets are formulated close to an animal’s requirement and the assumed digestibility is substantially higher than the ingredient’s actual digestibility, the animal may effectively receive a deficient diet. The biological consequences can extend beyond growth. Bureau said reduced nutrient availability can influence feed conversion, meat yield, survival, disease resistance and waste output, ultimately affecting both economics and environmental performance. Quality starts before the crushing plant Some heat exposure is intentional and controlled. However, heat damage begins long before soybeans reach processing. One particularly important period is the interval between physiological maturity and reaching a moisture level suitable for storage and transportation. Bureau explained that soybeans can contain around 60% moisture at maturity and must dry to approximately 11% to 12% for safe handling and storage. Under favorable conditions, that process takes about two weeks. Climate, cropping systems and infrastructure can change how easily that drying process occurs. In temperate production regions such as the United States and Canada, farmers generally produce one crop annually and allow soybeans to dry before harvest. However, conditions often are different in Brazil, where farmers may be trying to harvest soybeans quickly enough to establish a second corn crop. Soybean harvest can also coincide with significant rainfall in some regions. Wet beans create several challenges. Soybeans contain nutrients that support microbial growth. When beans remain too wet, bacteria and fungi can proliferate, biological and chemical reactions generate heat, and localized self-heating can occur. Bureau described the development of “hotspots” or bin burn where temperatures can become extremely high. “All beans that are too wet for too long will degrade,” Bureau said. Moisture distribution also matters. A load may have an acceptable average moisture concentration while moisture migrates and creates localized pockets where conditions favor self-heating. Artificial drying introduces another potential source of heat damage. Bureau said drying should generally occur around 60°C to 70°C maximum, but less-controlled systems can reach 80°C or even 90°C for extended periods. The result can be heat damage before the soybean ever enters the crushing process. This helps explain why Bureau considers physical damage an important warning indicator. Broken beans may indicate the crop experienced excessive handling, moisture, drying challenges or other stressors. He described them as the “canary in a coal mine,” signaling that something happened to the beans. Why soybean processing requires heat While unintended heat exposure can damage soybeans, eliminating heat from processing is not an option. Heat serves several essential functions in soybean meal production. According to Bureau, heating soybeans: - Helps condition and soften beans for dehulling and flaking - Assists in removing solvent after oil extraction - Denatures heat-sensitive antinutritional factors Trypsin inhibitors are particularly important because they interfere with protein digestion. Raw or inadequately heated soybean products can therefore have poor protein digestibility and support poor animal performance. Bureau said trypsin inhibitor activity needs to be reduced to below approximately 10 international units per gram of product. Above that level, digestibility of protein and amino acids declines as trypsin inhibitor activity increases. Heating initially improves feeding value by reducing those antinutritional factors. But the relationship is not linear indefinitely. Animal trials in which soybeans are subjected to different heating durations or temperatures demonstrate what Bureau described as a bell-shaped response. Animal performance initially improves as antinutritional factors are destroyed, but it reaches an optimal value then declines as excessive heat begins damaging proteins and amino acids. “If you overdo it, then you have a risk of having other reactions that are deleterious,” Bureau said. “Lysine is particularly sensitive, although heat damage can affect numerous amino acids.” Existing quality tests have limitations Knowing that both insufficient and excessive heating are problematic creates an obvious quality-control question: How can a feed mill determine whether soybean meal falls within the optimal range? Bureau says current tools remain imperfect. Common measurements include potassium hydroxide, or KOH, protein solubility, protein dispersibility index and urease activity. These measurements are useful for tracking changes within a controlled processing system, but Bureau questioned their ability to consistently differentiate nutritional quality among soybean meals from numerous origins. He cited research showing a lack of correlation between KOH solubility and protein digestibility in soybean meals of different origins. The distinction is important because a test may effectively show changes when one batch of soybeans is processed under progressively different conditions but perform poorly as a screening tool across commercial samples from many sources. Bureau said there is growing agreement that reactive lysine is probably the best available test for identifying heat damage because lysine is among the amino acids most sensitive to excessive heat. However, it’s still an indirect solution. Reactive lysine focuses on one amino acid rather than the full amino acid profile, and Bureau estimated its predictive capability at roughly 60% to 70% at best. Near-infrared spectroscopy, or NIRS, could make reactive lysine assessment more practical if robust calibrations are developed. However, creating those calibrations requires hundreds of samples and considerable analytical and biological research. Moving toward digestible amino acid estimates Ultimately, Bureau believes the industry should move toward measuring what nutritionists actually need. If the objective is to formulate diets based on digestible nutrients, quality control systems should aim to predict digestible amino acid content rather than depend heavily on surrogate measurements such as protein solubility. Developing those tools will require substantial investment. Bureau estimated that creating an appropriate calibration for amino acid digestibility could require roughly 300 soybean samples for a single animal species, supported by in vitro work, in vivo research and animal trials. Until better tools become commercially practical, his recommendation is to use multiple quality-control measurements rather than depend on a single number. Reactive lysine, including NIRS calibrations where appropriate, currently represents one of the better available options. Animal trials can provide an additional layer of validation, particularly for feed manufacturers that consistently purchase significant volumes from particular suppliers. Most importantly, Bureau urged nutritionists to maintain some skepticism about what laboratory values actually reveal. Crude protein remains useful as a reproducible measurement, but it should not be confused with feeding value. Amino acid concentration provides more information, but concentration does not guarantee digestibility. Traditional processing indicators can add useful evidence, but no single existing test completely characterizes heat damage. For Bureau, soybean quality is ultimately a biological question. Physical characteristics can provide early warnings, chemical analysis can improve understanding and emerging technologies may offer better predictions. But the final measure is whether nutrients are actually available to the animal. “If you want to assess the quality and nutritive value of ingredients, the best answer is given by the animal,” Bureau said. That principle becomes increasingly important as nutritionists formulate diets closer to animal requirements. When over-formulation is reduced, small differences in amino acid digestibility can no longer be assumed away. The challenge for the soybean and feed industries is therefore not simply producing meal with enough crude protein. It is preserving the nutritional integrity of the soybean from field to feed mill, applying enough heat to eliminate antinutritional factors without damaging valuable amino acids, and developing analytical tools capable of measuring the feeding value that ultimately matters to the animal.