A kamaboko-like food having a good texture can be produced by preparing a paste by mixing 1 to 20% by weight of a powdery vegetable protein, 1 to 8% by weight of gellan gum, 0.005 to 0.15% by weight of a divalent metal salt as a divalent metal, and 60 to 95% by weight of water, and molding and heating the paste. The powdery vegetable protein is preferably soybean protein. Additionally, the paste may also be molded uniformly to a thickness of 5 mm or less. By using this paste, a crab-flavored kamaboko-like food, which has many process restriction, can be produced.

This disclosure provides a technology for developing individual proteins for use in industrial processes that include food production. The technology mines sequence data from protein databases by a process that is done partly in silico. Instead of sampling and testing a vast library of compounds, machine learning and implementation narrows the field of functional candidates by predictive modeling based on known protein structure. Candidate proteins that are selected by this analysis are then produced and screened in a high-throughput manner by recombinant expression and testing to determine whether they have a target function. Multiple cycles of the machine learning, database mining, expression, and testing are done to yield potential ingredients suitable for use in the production of foods, cosmetics, agricultural feed, pharmaceutical excipients, and other industrial products.

This disclosure provides a technology for developing individual proteins for use in industrial processes that include food production. The technology mines sequence data from protein databases by a process that is done partly in silico. Instead of sampling and testing a vast library of compounds, machine learning and implementation narrows the field of functional candidates by predictive modeling based on known protein structure. Candidate proteins that are selected by this analysis are then produced and screened in a high-throughput manner by recombinant expression and testing to determine whether they have a target function. Multiple cycles of the machine learning, database mining, expression, and testing are done to yield potential ingredients suitable for use in the production of foods, cosmetics, agricultural feed, pharmaceutical excipients, and other industrial products.

The invention relates to methods for the production of an oleogel, comprising providing a dispersion of protein in oil and slowly adding and mixing water to the dispersion to produce a solid oleogel. The invention further relates to 5 an oleogel comprising protein, water and oil, preferably an oleogel that is produced by a method of the invention, and to products comprising an oleogel of the invention.

The invention relates to methods for the production of an oleogel, comprising providing a dispersion of protein in oil and slowly adding and mixing water to the dispersion to produce a solid oleogel. The invention further relates to 5 an oleogel comprising protein, water and oil, preferably an oleogel that is produced by a method of the invention, and to products comprising an oleogel of the invention.

The invention provides a method for obtaining a tuber protein isolate, comprising a) peeling at least one tuber, thereby obtaining at least one peeled tuber and a tuber peel composition; b) processing said at least one peeled tuber to obtain an aqueous liquid comprising tuber protein; and c) subjecting said aqueous liquid to a protein isolation step to obtain said tuber protein isolate. It has been found that peeling potatoes prior to protein isolation has several benefits: the isolated crude protein, or a hydrolysate thereof, is more clean, and has a composition enriched in tyrosine, proline, arginine glutamine, glutamate, asparagine and aspartate. The protein composition obtained from the tuber peels on the other hand is enriched in the essential amino acids threonine, leucine, isoleucine, methionine and phenylalanine.

This disclosure provides a technology for developing alternative protein sources for use in industrial food production. The technology mines natural sources by a process that is done partly in silico. Instead of sampling and testing a vast library of compounds, machine learning and implementation narrows the field of functional candidates by predictive modeling based on known protein structure. Candidate proteins that are selected by this analysis are then produced and screened in a high-throughput manner by recombinant expression and testing to determine whether they have a target function. Multiple cycles of the machine learning, database mining, expression and testing are done to yield potential ingredients suitable for assessment as part of a commercial food product.

This disclosure provides a technology for developing alternative protein sources for use in industrial food production. The technology mines natural sources by a process that is done partly in silico. Instead of sampling and testing a vast library of compounds, machine learning and implementation narrows the field of functional candidates by predictive modeling based on known protein structure. Candidate proteins that are selected by this analysis are then produced and screened in a high-throughput manner by recombinant expression and testing to determine whether they have a target function. Multiple cycles of the machine learning, database mining, expression and testing are done to yield potential ingredients suitable for assessment as part of a commercial food product.

Systems and methods describe continuously and progressively hydrating material, such as food material for meat analogue products. First, material is provided to be conveyed through a material passage between an exterior tube and a rotating inner shaft, with the rotating inner shaft including one or more agitation and/or progression features. The progression features could be, e.g., a series of imbricated protruding filled paddles arranged in a helical pattern, while the agitation features could take the form of, e.g., unfilled hoops, hooks, or paddles. Concurrent to conveying and hydrating the material through the material passage, a number of lumps, clumps, and/or unhydrated pieces of the material are broken up via one or more agitation features configured to produce uniform hydration and consistent dispersal of the material. Also concurrently or subsequently, water is continuously and/or progressively provided to the material to produce hydrated material particles.

This disclosure provides a technology for developing alternative protein sources for use in industrial food production. The technology mines natural sources by a process that is done partly in silico. Instead of sampling and testing a vast library of compounds, machine learning and implementation narrows the field of functional candidates by predictive modeling based on known protein structure. Candidate proteins that are selected by this analysis are then produced and screened in a high-throughput manner by recombinant expression and testing to determine whether they have a target function. Multiple cycles of the machine learning, database mining, expression and testing are done to yield potential ingredients suitable for assessment as part of a commercial food product.