Method and test kit for estimating germination characteristics of plant seed

Abstract

The present invention relates to a method for estimating the germination characteristics of a plant seed, comprising the following method steps: a) providing a plurality of individual plant seed parts, each plant seed part comprising at least one plant seed, b) producing or providing a test composition comprising: i) water, ii) a two-stage redox indicator, and iii) a fermentation microorganism, c) contacting the plant seed parts with a respective test volume of a test composition in order to obtain a plurality of individual test systems and incubating the test systems, d) measuring the test composition of the incubated test systems using an optical measurement method, in order to determine the optical absorption properties of the respective test composition for the electromagnetic radiation of at least the first wavelength [lambda] 1 and the second wavelength [lambda] 2, in order to obtain a plurality of absorption data records assigned to the respective test system, wherein the absorption data record comprises information about the absorption characteristics of the respective test composition at the first wavelength [lambda] 1 and the second wavelength [lambda] 2, where [lambda] 1 and [lambda] 2 differ by 10 nm or more, e) evaluating the absorption data record assigned to the test system using a machine learning-based estimation module, so as to estimate germination characteristics of the plant seed in the respective plant seed portion.

Description

Technical Field

[0001] The present invention relates to a method for estimating the germination characteristics of plant seeds, a test composition for such a method, a kit for producing the corresponding test composition, and a computer program product for performing the steps of the corresponding method. An electronic data processing device for such a method is also disclosed. Background Art

[0002] One of the most pressing challenges facing humanity is how to reliably provide sufficient food for the growing world population. In this context, past and future innovations in agricultural engineering and the agricultural field have special significance. For many of the most important agricultural products, obtaining high-performance seeds that can reliably and highly produce usable plants is crucial for achieving maximum yields on the available land surface.

[0003] In this context, in the agricultural field, seeds are important commodity entities that are subject to strict requirements. In particular, these relate to the ability of seeds to exhibit sufficient germination ability and seed vigor after sowing. In this context, the assessment of seed quality and ensuring that this quality remains consistent play such a prominent role that international associations dedicated to ensuring seed quality and its testing and certification, such as the Association of Official Seed Analysts (AOSA) and the International Seed Testing Association (ISTA), have existed as early as 1908 and 1924.

[0004] The tests defined by the ISTA for determining the germination characteristics of seeds and the defined quality standards constitute an undisputed industry standard in most countries of the European Union and are continuously updated by the ISTA. Thus, in the context of the present invention, first and foremost, the ISTA methods and standards are used as a basis, where in many cases, these methods and standards only differ slightly from the AOSA standards.

[0005] In short, the classical method for determining germination characteristics is based on extracting a representative seed portion from the seed batch to be evaluated, where the obtained test seeds are placed under conditions where the seeds can germinate, where the presence of germination of individual seeds in the test group and the degree of seed vigor are determined, and the specific germination characteristics of the seed batch are obtained through random evaluation.

[0006] Although it is undisputed to date that the tests stipulated by the ISTA are of the utmost relevance to the industry, determining germination characteristics by the actual germination of seeds is considered disadvantageous in many cases considering the time required and the workload involved. In this context, there has always been an interest in providing a method that can reliably and quickly estimate the germination characteristics of plant seeds while seeking a satisfactory correlation with the germination characteristics stipulated by the ISTA.

[0007] For example, a germination ability test of plant seeds known from the prior art is disclosed in DE 1020200567 A1. This germination ability test is based on a method in which a number of plant seeds are brought into contact with a predetermined volume of a test solution in separate containers, the test solution containing resazurin (i.e., a two-stage redox indicator) and yeast (i.e., a fermenting microorganism). The plant seeds are brought into contact with the test solution and cultivated. Generally, it is assumed that the germination ability of plant seeds decreases with an increase in seed coat damage. During cultivation, this damage to the seed coat causes glucose and other organic substances to escape from the interior of the seeds into the surrounding test solution. These substances can be reacted at least in part by the yeast, which causes the two-stage redox indicator resazurin to be converted into fluorescein or dihydrofluorescein. This conversion causes a color change. DE10202020567A1 makes use of this by measuring the absorption of the test solution at 570 nm in order to correlate the normalized absorption value with the actual germination characteristics obtained in the germination test described subsequently. By linear regression, a function can be estimated that allows the germination characteristics to be estimated based on the absorption values of similar plant seeds. According to the inventor's assessment, the method of DE 1020200567 A1 basically provides a very promising germination ability test, in particular the method can be implemented much faster than methods based on actual sowing. This method, which allows at least a rough estimate of the germination characteristics of various seeds, also has advantages in terms of time and cost efficiency, is non-destructive, and in most cases only has a minor impact on the environment.

[0008] Although the methods known from the prior art have the above advantages, many experts - including the inventors of the present invention - believe that the method needs to be improved in many respects. In particular, the prediction quality that can be achieved using the methods of the prior art is generally considered to be insufficient, especially because in order to obtain a robust correlation between the absorption at 570 nm and the germination ability, usually a very large data set is required, and despite sometimes complex normalization processes, a major scatter is still observed around the obtained regression. The present inventors believe that this defect in terms of the core function (i.e., the estimation of the germination characteristics) explains why this otherwise advantageous method for analyzing the germination ability of plant seeds, despite the above advantages, has not been fully established in the industry. Summary of the Invention

[0009] The main object of the present invention is to eliminate or at least mitigate the above disadvantages of the prior art.

[0010] In particular, it is an object of the present invention to disclose a method for estimating the germination characteristics of plant seeds by which the germination characteristics of plant seeds can be estimated particularly reliably and precisely.

[0011] In this context, the object of the present invention is to allow the provided method to be carried out as a non-invasive method that does not damage the plant seeds being tested and is preferably as unproblematic as possible in terms of environmental and health aspects.

[0012] Furthermore, the object of the present invention is to allow the provided method for estimating germination characteristics and evaluation to be much faster than the germination ability tests known from the prior art, so that the provided method shows higher time and cost efficiency.

[0013] Another object of the present invention is to allow the provided method to be efficiently implemented in at least a partially decentralized configuration, so that the estimation of the germination characteristics of plant seeds can also be carried out outside research institutions with the lowest possible equipment requirements.

[0014] Another object of the present invention is to allow the provided method to be applicable to various possible plant seeds and, regardless of the plant seeds used, to allow reliable estimation of germination characteristics.

[0015] In this context, the object of the present invention is to allow the provided method to be able to reliably estimate specific germination characteristics such as germination ability and seed vigor.

[0016] Another object of the present invention is to allow the provided method to be easily automated and suitable for high throughput. Additionally, the provided method is preferably configured in such a way that it allows particularly simple operation and relatively few requirements for the education and training of the personnel using the method.

[0017] Another object of the present invention is to provide a particularly advantageous test solution for use in the provided method and a kit for producing a test composition specific to plant seeds.

[0018] Furthermore, another object of the present invention is to provide a computer program product by which the basic method steps of the provided method can be executed on an electronic data processing device.

[0019] The second object of the present invention is to provide an electronic data processing device for use in the provided method.

[0020] The inventors of the present invention have now found, surprisingly, that if starting from the germination ability test based on the use of resazurin and yeast and the photometric evaluation of the color change of resazurin at 570 nm known from the prior art, as defined in the claims, instead detecting the absorption characteristics at two sufficiently different wavelengths and using machine learning techniques to evaluate the absorption data set obtained in this way, the above objects can indeed be achieved.

[0021] Surprisingly, by this modification of a method known from the prior art, namely the detection of at least two wavelengths sufficiently separated from each other and the use of machine learning methods, a favorable method has been obtained that is capable of evaluating the germination characteristics of plant seeds with a significantly increased prediction accuracy, wherein the resulting method is fast, inexpensive, safe, and seed-saving, and in particular can be used in a decentralized configuration with relatively low equipment requirements and a relatively low amount of training for the personnel used. A significant and consistent improvement has been achieved in predicting the most important germination characteristics compared to the germination tests known from the prior art.

[0022] Accordingly, the above object is achieved by the subject matter of the present invention as described in the claims. Preferred embodiments of the present invention are described in the dependent claims and in the following explanations.

[0023] In the embodiments specified as preferred below, the features of the other embodiments specified as preferred are combined in the particularly preferred embodiments. Accordingly, a combination of two or more of the embodiments specified as particularly preferred below is most particularly preferred. Also preferred are embodiments in which the features of one embodiment described as preferred to any extent are combined with one or more other features of the other embodiments described as preferred to any extent. The features of the preferred test compositions, kits, electronic data processing devices, and computer program products can be derived from the features of the preferred methods.

[0024] Below, for a certain element, such as for a two-stage redox indicator or a fermentation microorganism, if the specific amount or ratio of the element and the preferred embodiments of the element are disclosed, then in particular the specific amount or ratio of the element in the preferred configuration is also disclosed. Additionally, it is disclosed that at least a part of the element can be preferably configured in the corresponding specific total amount or total ratio of the element, and in particular, it is also disclosed that the element in the preferred configuration can also be present in the specific total amount or total ratio in a specific amount or ratio.

[0025] The present invention relates to a method for estimating the germination characteristics of plant seeds, comprising the following method steps:

[0026] a) providing a plurality of individual plant seed parts, each plant seed part comprising at least one plant seed,

[0027] b) producing or providing a test composition, the test composition comprising:

[0028] i) water,

[0029] ii) a two-stage redox indicator, and

[0030] iii) a fermentation microorganism,

[0031] c) Contact the plant seed part with a test composition of a corresponding test volume to obtain a plurality of individual test systems, and cultivate the test systems.

[0032] d) Measure the test composition of the cultivated test systems using an optical measurement method to determine the optical absorption characteristics of the corresponding test composition for electromagnetic radiation at least at a first wavelength λ1 and a second wavelength λ2, so as to obtain a plurality of absorption data records assigned to the corresponding test systems.

[0033] Wherein the absorption data set includes data on the absorption characteristics of the corresponding test composition at the first wavelength λ1 and the second wavelength λ2.

[0034] Wherein λ1 and λ2 differ by 10 nm or more.

[0035] e) Use an electronic data processing device to evaluate the absorption data set assigned to the test systems to estimate the germination characteristics of the plant seeds in the corresponding plant seed part, wherein the electronic data processing device includes a storage unit, and an estimation module based on machine learning is stored on the storage unit.

[0036] Wherein the electronic data processing device is configured to input the absorption data set obtained for the plant seed part as an input into the estimation module, and use the estimation module to estimate the germination characteristics of the plant seeds in the plant seed part.

[0037] Wherein the estimation module is trained to estimate the germination characteristics of the plant seeds in the plant seed part according to the absorption data set, and the training is carried out by supervised learning using a set of training data, and the set of training data includes a plurality of training absorption data sets of training plant seed parts of plant seeds with known germination characteristics.

[0038] The method according to the invention is used to estimate the germination characteristics of plant seeds. In terms of language, this contrasts with the expression "germination ability test" used in DE1020200567 A1, but it seems more appropriate to the inventor. This is because strictly speaking, in the method according to the invention, the germination characteristics of plant seeds are not tested or analyzed, as is the case with the ISTA test. Instead, the method according to the invention is based on predicting the germination characteristics of ungerminated plant seeds by correlation with the absorption values measured in similar plant seeds that germinate subsequently, so that the actual test of the germination characteristics can be avoided. As understood by those skilled in the art, the term "germination characteristics" is expressed similarly to various evaluation criteria developed by ISTA, and in addition to the germination ability, further parameters will be assigned to germination, i.e., for example, seed vigor, and these characteristics can also be advantageously determined by the method according to the invention. Thus, the method according to the invention is exemplary, where the estimated germination characteristics in each case indicate an estimate of the probability that the plant seeds of the plant seed portion will germinate physiologically, preferably indicating the probability that the plant seeds of the plant seed portion will germinate physiologically within a predetermined observation period after sowing, where the presence of physiological germination is particularly preferably checked every 24 hours. As understood by those skilled in the art, "physiological germination" can be understood as the emergence of a root with a length of at least 2 mm from the shell of the plant seed, where this is sometimes referred to as the "white root tip". However, in practice, the exact criteria are based on the current ISTA standards in each case. Additionally or alternatively, the method according to the invention is also exemplary, where the estimated germination characteristics in each case include an estimate of the probability that the plant seeds of the plant seed portion will form a normal seedling as defined by the ISTA standards. A normal seedling exists if the seed can be expected to form a healthy and complete seedling that can develop into a complete plant, as defined by the current ISTA standards.

[0039] Those skilled in the art understand that the method according to the invention can be used efficiently to determine the germination characteristics of multiple plant seeds. In practice, similar to the method developed by ISTA, the most relevant application is to determine the germination characteristics of a large batch of seeds by analyzing a representative partial number of plant seeds.

[0040] In the method according to the invention, in method step a), first a plurality of plant seeds are provided, which can be, for example, a corresponding representative partial number of seeds in a larger seed batch, and these plant seeds are divided into separate plant seed portions. To avoid any influence of contamination on the surface of the plant seeds, it is generally recommended to at least roughly clean the plant seeds before the method, for example, wash them off with distilled water.

[0041] As described below, according to the inventors' assessment, in most cases the method of the present invention is used to identify the overall evaluation parameters of whole plant seeds, i.e., the bulk parameters (bulk properties) of a seed lot or the prediction of average germination characteristics. In this context, two or more plant seeds can usually be used together as a plant seed portion, so that, for example, 100 plant seed portions can be used, each plant seed portion having 2 plant seeds. However, in the inventors' assessment, taking into account the achievable resolution of the germination characteristics, it is preferred to use a single plant seed as the plant seed portion in each case, because this prevents the differences in the germination characteristics of the plant seeds from disappearing due to averaging, as would occur, for example, if a plant seed with particularly favorable germination ability was measured together with a plant seed with lower germination ability in a common plant seed portion. For substantially all embodiments, a method according to the invention in which the plant seed portion comprises the same number of plant seeds in order to allow for an efficient experimental plan is preferred, according to the understanding of a person skilled in the art. For substantially all embodiments, a method according to the invention in which the plant seed portion consists of one or two, preferably one, plant seed, and / or in which each test system contains only one plant seed is particularly preferred.

[0042] In method step b), a test composition is produced or provided, where the latter can be accomplished, for example, by purchasing a finished test composition from a supplier. However, in the inventors' assessment, in most cases the test composition is affected by certain aging phenomena, so that, taking into account the achievable estimated quality, it is particularly preferred to produce the test composition in the method according to the invention.

[0043] For example, this can be achieved by mixing a two-stage redox indicator and a fermentation microorganism with water only immediately before further method steps in order to produce a test solution. This type of process management is highly advantageous because it prevents color changes of a part of the redox indicator that occur as signs of aging in the formulated test composition. Thus, a method according to the invention in which the test composition is produced in method step b) is preferred, where it is preferred to provide the test composition within 2 hours or less, particularly preferably within 1 hour or less, and most particularly preferably within 0.5 hour or less, especially immediately before combining the seed portions.

[0044] The method according to the invention is advantageously applicable to a wide variety of plant seeds. In fact, based on the available test data, the inventors have no reason to believe that the method according to the invention is not suitable for certain types of plant seeds. Through extensive testing, the inventors have determined the applicability of the method to a wide variety of plant seeds of the most widely different types. In this regard, thus, a method according to the invention is preferred, wherein the plant seeds are plant seeds selected from the group consisting of plant seeds of gymnosperms (Gymnospermae) and angiosperms (Angiospermae), preferably plant seeds selected from the group consisting of plant seeds of gymnosperms (Gymnospermae), monocotyledonous gymnosperms (Monocotyledoneae) and dicotyledonous gymnosperms (Dicotyledoneae), particularly preferably plant seeds selected from the group consisting of the following plant seeds:

[0045] - Sweet grass plants (Gramineae), such as Alopecurus, Avena, Hordeum, Lolium, Poa, Secale, Triticum, Triticale, Zea;

[0046] - Amaryllis plants (Amaryllidaceae), such as Allium;

[0047] - Asparagus plants (Asparagaceae), such as Asparagus;

[0048] - Umbelliferae, such as Apium, Daucus, Petroselinum;

[0049] - Compositae, such as Helianthus;

[0050] - Cruciferae, such as Brassica, Lepidium, Raphanus, Sinapis;

[0051] - Caprifoliaceae, such as Valeriana;

[0052] - Chenopodiaceae, such as Beta;

[0053] - Cucurbitaceae, such as Cucumis;

[0054] - Leguminosae, such as Glycine, Lens, Pisum, Trifolium, Vicia, Lupinus;

[0055] - Labiatae, such as Ajuga, Lavandula;

[0056] - Ranunculaceae, such as Delphinium;

[0057] - Rosaceae, such as Fragaria, Filipendula;

[0058] - Solanaceae, such as Lycopersicon, Solanum.

[0059] In the most particularly preferred method according to the invention, the plant seeds are the seeds of the plants shown in Tables 1 and 2.

[0060] In the method according to the present invention, the various components of the test composition have different meanings. According to the present invention, the solvent must contain water, since the presence of water is crucial for the subsequent required functions. In general, at least in addition to water, other solvents may also be present, so that the test composition as a whole constitutes an aqueous solvent. However, in practice, the inventors have found that, considering factors such as environmental compatibility, the seeding nature of the method, and the total cost, it will be advantageous if the solvent used in the test composition consists of 95% or more, preferably 98% or more, particularly preferably 99% or more, and most particularly preferably substantially entirely of water, based on the mass of the solvent, which advantageously can also prevent any other solvent component from having a negative impact on the activity of the fermentation microorganisms and / or the color change behavior of the redox indicator. Thus, a method according to the present invention in which the water is distilled water or softened water is preferred.

[0061] In the prior art, the test composition is sometimes referred to as a test solution. However, as understood by those skilled in the art, the test composition is not in all cases a solution in the narrow sense, i.e., a substantially homogeneous solution. Those skilled in the art understand that the term test composition is more appropriate, since in addition to test solutions, it also includes those liquid systems in which, for example, undissolved particles may optionally be present as a result of incompletely dissolved fermentation microorganisms, so that these systems should rather be referred to as test suspensions.

[0062] The text "composition" used includes two-stage redox indicators by which the change in the redox potential of the test composition can be indicated. According to the prior art, the redox indicator itself and its function are as well known to those skilled in the art as two-stage redox indicators. These two-stage redox indicators exhibit two transitions, in which, due to reduction or oxidation reactions, the absorption characteristics change, and thus in most cases, the visually perceptible color also changes. Thus, the two-stage redox indicators exhibit three different redox states, and these two redox transitions are located between these three states.

[0063] In the assessment of the inventors, the concept on which the method according to the invention is based can generally be implemented using any two-stage redox indicator. However, due to the applications known from the prior art, the favorable risk profile and the absorption wavelength ranges of the various redox states that are favorable for the assessment of the device, resazurin is particularly destined for use in the method according to the invention and is preferred as a two-stage redox indicator for all embodiments. Thus, according to the prior art, resazurin is well-known to the person skilled in the art and, like other redox indicators, can be obtained from different manufacturers. The method according to the invention in which the two-stage redox indicator is selected from the group consisting of dyes having a phenoxazine-3-one basic skeleton, such as resazurin, and in which the preferred two-stage redox indicator is resazurin is preferred.

[0064] As another component, the test composition contains fermenting microorganisms. The term "fermenting microorganisms" is clear to the person skilled in the art and designates microorganisms that are capable of transforming organic substances, in particular glucose, in a microbial or enzymatic manner by means of so-called fermentation, such as bacteria and fungi. In the assessment of the inventors, unicellular fungi known to the person skilled in the art, such as unicellular fungi in the form of yeast, such as Saccharomyces cerevisiae, are particularly suitable as fermenting microorganisms. Thus, the method according to the invention in which the fermenting microorganisms are selected from the group consisting of unicellular fungi, preferably selected from the group consisting of yeast, and particularly preferably selected from the group consisting of Saccharomyces cerevisiae and Saccharomyces bayanus is preferred. With regard to the production of the solution, the method according to the invention in which the test composition is produced in method step b) is preferred, in which the fermenting microorganisms are used as cold-treated microorganisms, preferably as freeze-dried microorganisms.

[0065] In the method according to the invention, the test composition, at least in its broadest defined form, has the same purpose as known from the prior art, while the preferred test compositions disclosed below also achieve additional functions. Without wishing to be bound by this theory, the inventors hypothesize that the processes occurring in the method according to the invention can be described as follows. In the dry state, the selective permeable membrane does not function in the plant seed. After contact with water, the plant seed begins to absorb water. Depending on the physiological structure, complete reconstruction or insufficient reconstruction of the selective permeable membrane occurs, resulting in varying degrees of leaching of storage substances such as carbohydrates, proteins or fats. In a well-structured case, the seed controls the leaching so that only low concentrations of these organic substances are present in the test composition. In contrast, in dead plant seeds, there is hardly any reconstruction of the selective permeable membrane. This leads to strong leaching of the storage substances, and thus these storage substances are found in increased concentrations in the test composition. Thus, the state of the plant seed is related to the amount of organic substances available to the fermenting microorganisms in the test composition during cultivation. Then, depending on the viability of the plant seed, the activity of the microorganisms in the test composition will vary. Redox indicators are now used to make the different activities of the micro-fermenting microorganisms detectable and even visible. This can be illustrated using resazurin as an example. Resazurin is a two-stage redox indicator that is dark blue-violet when fully oxidized. After the first reduction, the color changes in the direction of pink (irreversible), while after the second reduction, a colorless form is obtained (reversible). In other words, this is the method according to the invention, wherein during the fermentation of the fermentable compound, the fermenting microorganisms can change the redox potential of the test composition, and / or wherein the test composition is configured such that the fermentation of the fermentable compound by the fermenting microorganisms in the test composition first triggers the first color change in the two-stage redox indicator and then the second color change, or first triggers the first reduction and then the second reduction.

[0066] The inventors have realized that, in order to obtain particularly advantageous estimation results, it is advisable, in the method according to the invention, to adapt the composition of the test composition to the plant seeds to be tested, particularly with regard to the concentration of the redox indicator and the fermentation microorganism. In this regard, the inventors propose that, in the kit according to the invention, as described below, production instructions specific to the plant seeds can in particular be provided for producing the corresponding test composition from a prefabricated solid (such as a powder) by mixing with water or an aqueous solvent, so as to allow a specific mixing of the components. In this regard, the inventors have successfully determined the generally preferred ranges, and according to the evaluation of the inventors, particularly advantageous test compositions can be obtained through these ranges. Even in the absence of production specific to the plant seeds, the test compositions can be used for a variety of plant seeds. Specifically, the method according to the invention in which, based on the mass of the test composition, the mass fraction of water in the test composition is 70% or more, preferably 80% or more, particularly preferably 90% or more, more particularly preferably 95% or more, and most particularly preferably 99% or more is preferred. Additionally or alternatively, the method according to the invention in which, based on the mass of the test composition, the mass fraction of the two-stage redox indicator in the test composition is in the range of 0.00001 to 5%, preferably in the range of 0.001 to 0.5%, and particularly preferably in the range of 0.01 to 0.05% is preferred. Furthermore, additionally or alternatively, the method according to the invention in which, based on the mass of the test composition, the mass fraction of the fermentation microorganism in the test composition is in the range of 0.01 to 10%, preferably in the range of 0.05 to 5%, and particularly preferably in the range of 0.1 to 0.5% is preferred.

[0067] In method step c), the plant seed portion is brought into contact with a test volume of the test composition. As understood by those skilled in the art, this means bringing each plant seed into contact with a separate portion of the test composition. This contact can advantageously be carried out in a suitable container, and it is particularly advisable to use a so-called multi-well plate, since a large number of plant seeds are usually studied simultaneously in the method according to the invention. Thus, in many cases, the method according to the invention is effective in which the plant seed portions are each brought into contact with a test volume of the test composition in one of the plurality of test wells from a first test plate, so as to maintain separate test systems in the respective test wells of the first test plate.

[0068] As described above, in order to prevent aging effects in the test composition that may adversely affect the results of the method according to the invention, it is advisable to prepare the test composition immediately before bringing the seeds into contact. In this regard, in order to achieve particularly advantageous results, the inventors propose that the fermenting microorganisms should be used as cold-treated microorganisms before being mixed with the solvent, as also disclosed above. Furthermore, in order not to increase the activity of the fermenting microorganisms prematurely, the inventors propose to use relatively cold water when producing the test composition. In a favorable refinement, this can be done to such an extent that the temperature of the test composition is kept as low as possible even before the moment of contact with the plant seeds in order to prevent an undesired premature increase in the activity of the fermenting microorganisms. Compared with the method of preparing the solution at room temperature, this makes it possible to advantageously prevent any redox processes in the prepared solution from causing an initial color change of the redox indicator that could distort the results of subsequent optical measurements. Thus, the preparation method according to the invention in which the test composition is produced in method step b) is preferred, in which the water temperature during production is 8 °C or lower, preferably 7 °C or lower, preferably 6 °C or lower, particularly preferably 5 °C or lower. In this regard, the method according to the invention in which the temperature of the test composition is 8 °C or lower, preferably 7 °C or lower, preferably 6 °C or lower, particularly preferably 5 °C or lower when bringing the plant seed parts into contact with each other is particularly preferred.

[0069] By bringing the individual plant seed parts into contact with the corresponding parts of the test composition, a plurality of subunits are obtained which, for the sake of clear identification, are referred to as test systems in the context of the present invention. Thus, these test systems comprise the test composition and the plant seed parts. The person skilled in the art understands that the volume of the test composition added separately should preferably depend on the size of the plant seeds and / or the number of plant seeds in the plant seed parts. For a single small plant seed, the person skilled in the art will advantageously choose a smaller volume than for a larger plant seed part for a relatively large seed. In this regard, the person skilled in the art also understands that the efficiency of transfer of the organic substances identified by the combination of the fermenting microorganisms and the two-stage redox indicator from the plant seeds to the test composition generally depends essentially on the contact surface between the plant seeds and the test composition. Thus, although it is generally possible to use a small test volume of the test composition that only contacts part of the plant seeds, according to the inventors' assessment, this is not the preferred process control method. At the same time, a large amount of the test composition also means a large amount of redox indicator, and even in the case of dead seeds, most of the redox indicator may not be able to change color. In the inventors' assessment, for the test volume of the test solution in the test system, it is particularly preferred to be in the range of 1*V s ~500*V s preferably in the range of 2*V s ~250*V swithin the range of, particularly preferably at 3*V s to 125*V s wherein V s is the total volume of the plant seeds in the plant seed portion. Additionally or alternatively, a method according to the invention in which the plant seed portion is brought into contact with a test solution in a corresponding test volume in the range of 0.1 to 500000 μL, preferably in the range of 1 to 50000 μL, particularly preferably in the range of 10 to 5000 μL, is preferred.

[0070] To increase the efficiency of the method according to the invention, the inventors of the present invention propose taking measures to improve the wettability of the plant seeds, as this increases the interface and allows organic substances to be more effectively transferred from the plant seeds to the test composition. In particular, by corresponding measures, the time required to carry out the method according to the invention can be significantly reduced.

[0071] As a first less preferred option for adjusting the wettability to a favorable level, the inventors propose that the test system can be centrifuged to achieve favorable wettability. Thus, these examples relate to a method according to the invention in which, by a mechanical treatment step, preferably by immersing the plant seeds or centrifuging the test system, particularly preferably by centrifuging the test system, the contact of the plant seed portion with a test composition in a corresponding test volume is supported before cultivation.

[0072] However, the inventors believe that, compared to methods known from the prior art, the use of specific surfactants (so-called detergents) can achieve a profitable wettability of the plant seed portion and is particularly advantageous and preferred for substantially all embodiments of the method according to the invention. This allows for a significant increase in the efficiency of the method, particularly with respect to the time required, without the need for, for example, centrifugation. In implementing these preferred embodiments, the inventors consider it important that surfactants that have no negative impact or at least only a minimal negative impact on the biological activity of the fermenting microorganisms are used, rather than any desired surfactant. Accordingly, a method according to the invention in which, based on the mass of the test solution, the test composition additionally contains one or more surfactant compounds that are biocompatible with the fermenting microorganisms in a total mass fraction in the range of 0.001 to 25%, preferably in the range of 0.01 to 5%, particularly preferably in the range of 0.1 to 0.5%, is particularly preferred.

[0073] The term "biocompatible" is clear to those skilled in the art and means that in the presence of a surfactant compound, the activity of the microorganism is reduced by less than 5%, preferably less than 1%, particularly preferably less than 0.1%, and most particularly preferably is essentially not reduced at all. Many suppliers of commercially available detergents state in their documentation the degree of tolerance of the surfactant compounds provided to microorganisms, i.e., the degree of biocompatibility, or can provide corresponding data upon request. Additionally or alternatively, those skilled in the art can also determine whether the surfactant compounds available are sufficiently biocompatible by means of a relatively simple test on the fermentation microorganisms they use. Specifically, in the assessment of the inventors, nonionic surfactants are most particularly suitable for use in the method according to the invention, and in particular, polysiloxane-based polymers have produced excellent results in the experiments conducted by the inventors. Particularly favorable results have been achieved with nonionic surfactants based on trisiloxane. The corresponding anionic surfactants are available from various manufacturers under the Break-Thru brand, for example, Break-Thru SD260 from Evonic Operations GmbH. Thus, the method according to the invention in which the surfactant compound is selected from the group consisting of nonionic surfactants, preferably from the group consisting of polysiloxane copolymers, and particularly preferably from the group consisting of polyether-polysiloxane copolymers, is preferred.

[0074] In the method according to the invention, the test system obtained is then incubated. The incubation step is clear to those skilled in the art and refers to the development of the test system over a predetermined period of time (usually at an elevated temperature). In the method according to the invention, the incubation time in particular enables organic substances to escape from the plant seeds and these substances to be converted by the fermentation microorganisms. In this regard, the inventors have successfully determined particularly favorable incubation conditions by means of which good results can be obtained for the majority of plant seeds of industrially relevant crop plants, and in these examples, the problem of the trade-off between time and energy efficiency on the one hand and the most reliable possible estimation of the germination characteristics on the other hand has been solved in a particularly favorable manner. Thus, the method according to the invention in which the incubation time is 0.5 to 24 hours, preferably 1 to 12 hours, and particularly preferably 1.5 to 8 hours, is preferred. Additionally or alternatively, the method according to the invention in which the incubation is carried out under light-protected conditions is also preferred. Additionally or alternatively, the method according to the invention in which the incubation is carried out at a temperature of 6 to 40°C, preferably 8 to 37°C, and particularly preferably 10 to 35°C, is also preferred. Against this background, the method according to the invention in which the incubation is carried out for a predetermined time depending on the type of plant is generally preferred, where the predetermined time is preferably specified by the plant-seed-specific instructions in the kit according to the invention.

[0075] Those skilled in the art understand that at the end of the cultivation in the cultivated test system, a test composition is obtained, in which part of the two-stage redox indicator changes color once or twice, and these test compositions of the test system are measured in a further process according to the method of the present invention.

[0076] This measurement of the test composition of the cultivated test system is carried out in method step d), in which an optical measurement method is used. Suitable optical measurement methods are known to those skilled in the art based on their expertise, and suitable optical measurement devices, especially high-performance digital embodiments, are available from different suppliers. For almost all embodiments, the method according to the present invention in which the optical measurement method is an absorption measurement method, preferably a transmission measurement method, is relevant considering the measurement data to be detected. Since the data processing step according to the method of the present invention can be performed on the optical measurement device, the method according to the present invention in which the optical measurement unit includes an electronic data processing device is preferred.

[0077] Those skilled in the art understand that as a result of one or more color changes in the two-stage redox indicator, the wavelength of the absorption maximum of the test composition shifts, at least for changes in the visible light range, which can be perceived by those skilled in the art as a change in color. Those skilled in the art also understand that in the method according to the present invention, parameters associated with the absorption characteristics of the corresponding test compositions from the cultivated test system are referred to.

[0078] In this regard, it is clear to those skilled in the art that the simplest and most direct absorption characteristics can consist of directly measured absorption values. However, at the same time, instead of directly determined absorption characteristics, it is advisable to determine parameters associated with the absorption characteristics and / or values derived therefrom and / or determine them in subsequent evaluations. In particular, subsequent evaluations can be carried out on standardized or normalized absorption values, which are corrected, for example, by computational operations according to the absorption characteristics determined on a blank system, a reference system and / or a control system, as also described below. According to the understanding of those skilled in the art, an absorption data set is accordingly obtained for each test system, which data set includes data on the absorption characteristics of the corresponding test compositions, and in which, in addition to the directly determined absorption values defined above, these absorption characteristics can also include values derived therefrom.

[0079] Contrary to methods known from the prior art, the absorption characteristics of the individually cultivated test compositions are not measured only at 570 nm. Instead, the optical absorption characteristics of the respective test compositions are determined at a first wavelength λ1 and at a second wavelength λ2, where according to the invention these wavelengths must differ by at least 10 nm. Thus, in other words, this is a method according to the invention in which the measurement is carried out using an optical measuring device, preferably an optical photometer, where the optical measuring device is configured to determine the optical absorption characteristics of the test composition for electromagnetic radiation at least at the first wavelength λ1 and the second wavelength λ2, where λ1 and λ2 differ by 10 nm or more.

[0080] According to the above explanation, the method according to the invention is exemplary, where the absorption characteristics at least include

[0081] i) the absorption values of the respective test composition at the first wavelength λ1 and the second wavelength λ2, or

[0082] ii) values derived from these absorption values,

[0083] where the values derived from the absorption values are preferably obtained by one or more calculations, in particular, for example, depending on the absorption values determined on a control system, a reference system or a blank system, obtained by value normalization or value correction.

[0084] For example, the absorption characteristics of the respective test composition for electromagnetic radiation at least at the first wavelength λ1 and the second wavelength λ2 can be determined selectively only at the corresponding wavelengths, for example, by using a substantially monochromatic radiation source or by using a suitable filter. However, alternatively, a spectrum over a wide wavelength range can be acquired, and the first wavelength and the second wavelength can be read based on specific wavelengths.

[0085] In addition to the specific form of the evaluation of the data set, the advantages of the method according to the invention also result from the fact that the absorption data set determined for different test systems includes data on the absorption behavior at two different wavelengths at a certain minimum distance from each other. The total absorption spectrum of the cultivated test composition can be regarded in a simplified manner as the superposition, i.e., the overlap, of the mass fraction-weighted absorption spectra of the various states of the components contained therein, in particular the redox indicator. Based on the minimum distance between the first wavelength and the second wavelength, the absorption characteristics of the solution are determined at two wavelengths separated by a certain distance, wherein the contributions of the various different oxidation states of the two-stage redox indicator to the combined absorption spectrum are different and have a deviation ratio. The inventors believe that the data content of the obtained absorption data set is crucial for the downstream evaluation by machine learning. In their own experiments, the inventors determined 10 nm as a suitable lower limit that still enables useful results to be obtained with acceptable resource requirements before the contributions of the changes in the redox states are no longer sufficiently different. However, the inventors suggest that the greater the wavelength difference, the better the estimation results can generally be obtained in most cases, because in such cases, the relative contributions of the redox states of the redox indicator are usually more significantly deviated from the typical absorption spectrum. Thus, a method according to the invention in which the first wavelength λ1 and the second wavelength λ2 differ by 15 nm or more, preferably 20 nm or more, particularly preferably 25 nm or more, is preferred.

[0086] Although it is generally possible to measure the absorption behavior at three or more different wavelengths and additionally, for example, to determine the absorption characteristics at a third wavelength λ3 and a fourth wavelength λ4 in order to include the corresponding data in the absorption data set and take this into account in the subsequent computer-aided estimation, this is not preferred according to the inventors' evaluation. Those skilled in the art understand that although each additional measured wavelength can potentially improve the estimation accuracy, this will result in an increase in the equipment cost in the measurement as well as an increase in the required storage capacity and computing power. In the opinion of the inventors, the excellent estimation quality that can already be achieved with two wavelengths makes it unnecessary to measure more wavelengths, and thus the additional effort is considered unjustified, especially when the absorption characteristics are determined at wavelengths close to the corresponding absorption maxima of the relevant redox states of the redox indicator (for example, 570 nm and 600 nm for resazurin).

[0087] In practice, the effective selection of the first wavelength and the second wavelength is basically determined by the selection of the two-stage redox indicator. In this regard, the inventors propose that if the above-mentioned minimum distance is maintained, the first wavelength should be in the range of (m1 - 20) nm to (m1 + 20) nm, preferably in the range of (m1 - 10) nm to (m1 + 10) nm, where m1 is the wavelength of the absorption maximum of the redox state of the two-stage redox indicator, and / or the second wavelength should be in the range of (m2 - 20) nm to (m2 + 20) nm, preferably in the range of (m2 - 10) nm to (m2 + 10) nm, where m2 is the wavelength of the absorption maximum of the redox state of the two-stage redox indicator.

[0088] Regarding the particularly preferably used two-stage redox indicator resazurin, the inventors recommend the wavelength ranges in which they were able to obtain excellent results in their experiments and generate absorption data sets that can be evaluated in a particularly advantageous manner in subsequent evaluations. Specifically, the method according to the invention in which the first wavelength λ1 is in the range of 585 to 630 nm, preferably in the range of 590 to 620 nm, and particularly preferably in the range of 595 to 610 nm is preferred. Additionally or alternatively, and preferably additionally, the method according to the invention in which the second wavelength λ2 is in the range of 540 to 585 nm, preferably in the range of 550 to 580 nm, and particularly preferably in the range of 560 to 575 nm is preferred.

[0089] In the inventors' evaluation, at least theoretically, transmission measurements can be directly performed on the test system, i.e., without prior removal of the plant seed part. However, in practice, since this may lead to an otherwise avoidable deterioration of the obtained absorption measurement data, the inventors propose to take out a corresponding sub-volume of the test composition from the cultivated test system and perform the measurement by an optical measurement method in order to achieve as efficient a process control as possible. In this regard, the method according to the invention in which the test composition to be measured of the cultivated test system is separated before measurement of the plant seed part is preferred, where preferably a sub-volume is removed from the test system, particularly preferably a sub-volume in the range of 50 to 150 μL of each test composition, for measurement, where preferably the individual sub-volumes are transferred to the test of a second transparent test plate.

[0090] Before the subsequent evaluation of the absorption data set obtained for estimating the germination characteristics is explained in further detail below, it is advisable to address the various aspects particularly related to method steps a) to d). It will be readily understood by those skilled in the art that it is easier to correlate the absorption characteristics of the test system to be detected subsequently with the germination characteristics of the plant seeds in order to estimate the germination characteristics, and the fewer the other confounding factors and deviations, the lower the required computing power.

[0091] For a person skilled in the art, this means that the absorption data set of the test system should naturally be generated under as uniform conditions as possible. This method corresponds to the natural method of a person skilled in the art when designing a measurement series. This means, for example, using preferably similar plant seeds, forming plant seed parts of equal size, using test compositions of substantially equal volume, the test compositions used in this method having as uniform a composition as possible, bringing the plant seed parts into contact with the test volume as uniformly as possible, and determining the absorption characteristics, for example, under as identical conditions as possible, in particular using the same optical measurement device and measurement method.

[0092] Theoretically, multiple deviations, especially if they are relatively small, can even be well compensated in a downstream evaluation by resorting to machine learning, at least provided that the available computing power is increased accordingly and a high-performance estimation module is used. However, in the understanding of a person skilled in the art, for all embodiments, it is advisable to use the method according to the invention in which multiple equal-sized plant seed parts of similar plant seeds with substantially the same volume are brought into contact with substantially the same test composition and cultivated under substantially the same conditions, where the test composition of the test system cultivated in this way is measured by the same optical measurement method under the same measurement conditions to determine the optical absorption characteristics of the corresponding test composition. In this regard, the method according to the invention in which each test system contains the same number of plant seeds and substantially the same amount of test composition is particularly preferred. The method according to the invention in which the test composition of the cultivated test system is measured under the same conditions is also particularly preferred.

[0093] According to the inventor's assessment, another factor that is particularly advantageous for efficient process control is the use of a suitable comparison system that allows for an efficient evaluation of the plant seed parts measured together.

[0094] Initially, in the inventor's assessment, it is particularly advisable to provide so-called blank systems that do not contain plant seed parts or test compositions but only the solvent used in the test composition. These blank systems can be particularly used to correct the optical absorption characteristics determined in method step d) with respect to any influence of the solvent or sample carrier. Thus, the method according to the invention in which, in method step c), in addition to multiple test systems, one or more blank systems are produced, cultivated together with the test systems, and measured in method step d) is preferred, where the blank system contains only the solvent of the test composition, in particular water. The method according to the invention in which the optical absorption characteristics of the corresponding test composition determined in method step d) are preferably corrected by subtracting the blank value according to the optical absorption characteristics of one or more blank systems is particularly preferred.

[0095] In addition, the inventors propose that the control system can also be processed simultaneously. In the control system, instead of the plant seed part, a known amount of fermentable compound is prepared in a controlled manner to verify that the test composition has sufficient functionality, that is, to verify that, in particular, the fermenting microorganisms are active and can cause a color change in the redox indicator. Thus, the method according to the invention, in which in method step c), in addition to a plurality of test systems, one or more control systems are also produced, cultivated together with the test systems, and measured in method step d), is preferred, wherein the control system contains a fermentable compound, preferably a carbohydrate, in the test composition.

[0096] The inventors consider the use of these control systems to be particularly highly preferred because, by using the absorption characteristics determined for the control systems, the absorption characteristics determined for the corresponding test compositions for the test systems can be standardized. In practice, in fact, the observed biological processes are usually limited in their reproducibility, where various environmental factors affect the activity of the microorganisms and the intensity of leaching. In practice, by using control systems, both biological fluctuations, such as the actual concentration of the microorganisms used or their motility, i.e., their adaptability, and abiotic fluctuations, such as temperature fluctuations or different water qualities between measurements on different dates and / or at different locations and / or under different weather conditions, can be compensated, so that comparable data can be obtained under changing conditions. Therefore, the standardization preferred for substantially all embodiments makes it possible to at least partially, and in many cases almost completely, hide these factors, which advantageously makes it particularly easy to carry out the method according to the invention outside highly controlled laboratory conditions, so that, for example, the implementation of a decentralized method of absorption measurement by agronomists is much easier. In this case, for example, the standardization can be carried out using the wavelength measurement values corresponding to the fully oxidized form of the two-stage redox indicator. In any case, the method according to the invention, in which the optical absorption characteristics of the corresponding test compositions determined in method step d) are standardized based on the optical absorption characteristics of one or more control systems, is particularly preferred.

[0097] In addition to the blank system and the control system, additionally or alternatively, the inventors propose that a reference system containing only the test composition can also be used. These reference systems correspond in their color behavior or absorption to the initial state in which fermentation has not yet caused any color change in the redox indicator. In this way, the reference system is particularly suitable for setting appropriate cultivation conditions. When there is a sufficient color contrast between the control system in which fermentation occurs and the reference system, it is usually possible to determine by the naked eye alone that the cultivation has been carried out for a sufficient length of time. Thus, a method according to the invention in which, in method step c), in addition to a plurality of test systems, one or more reference systems are also produced, cultivated together with the test systems, and measured in method step d) is preferred, wherein the reference system contains only the test composition. In addition, a method according to the invention in which the test systems are cultivated until a predetermined color contrast between the control system and the reference system is observed is preferred.

[0098] In method step e), an estimation of the germination characteristics of the plant seeds is carried out, which will be disclosed in detail below.

[0099] In practice, isolated data on the estimated germination characteristics of plant seeds or parts of plant seeds are usually insignificant. The added value of the method according to the invention lies particularly in the fact that a larger number of plant seeds representing a larger seed lot can be analyzed separately for their germination characteristics in order to derive therefrom a prediction of the average germination characteristics of the seed lot, which prediction of the average germination characteristics can be used as a characteristic variable for the seed lot. Initially, a method according to the invention in which the number of parts of plant seeds and the number of test systems are 20 or more, preferably 40 or more, particularly preferably 60 or more, and most particularly preferably 80 or more is therefore preferred. Thus, a method according to the invention additionally comprising the following method steps is preferred:

[0100] f) Calculating a prediction of the average germination characteristics by averaging the germination characteristics estimated for the parts of plant seeds.

[0101] In most cases, it is advisable to output the data obtained in method step e) in a suitable manner, for example via a display, and it is also conceivable to output via an electronic interface for further data processing. Thus, a method according to the invention additionally comprising the following method steps is preferred:

[0102] g) Outputting the estimated germination characteristics or the calculated prediction of the average germination characteristics, wherein the output is preferably carried out via the data interface of an electronic data processing unit or an electronic display device. For the simple processing of the output data, the output is preferably carried out in an easy-to-understand graphical representation, for example in the form of a so-called box plot, such that a direct and intuitive comparison of the available data can be carried out in a particularly simple manner.

[0103] In the method according to the invention, with the help of a computer based on machine learning concepts (sometimes also referred to as "artificial intelligence" or "AI-based"), an absorption data set assigned to a test system is evaluated to estimate germination characteristics. In this method, the absorption data set determined for each test system as described above is input as an input to an estimation module based on machine learning in an electronic data processing device, and the estimation module performs an actual estimation and outputs corresponding estimation data regarding the germination characteristics, and the absorption data set includes data regarding the absorption characteristics of the corresponding test composition at a first wavelength λ1 and a second wavelength λ2.

[0104] For example, the electronic data processing device can be, for example, a separate computer of an end user connected to an optical measurement device. However, in the context of an integrated solution, it can also be an electronic data processing device that is a component of the optical measurement device. However, in a particularly preferred embodiment, the electronic data processing device is a central electronic data processing device, for example, a server in a cloud-based evaluation, which can be connected to a plurality of optical measurement devices in a network, for example.

[0105] In the context of the present invention, the module based on machine learning is designated as an estimation module in order to correctly name and identify the module by its function, wherein the ability of the module to estimate germination characteristics based on the absorption data set assigned to the test system comes from the training of the estimation module. For example, the estimation module can be part of an integrated software and exists on a storage unit of the electronic data processing device, where the term storage unit refers to a storage device that can be accessed by the electronic data processing device and does not necessarily need to be physically connected to the electronic data processing device, but can also be accessed, for example, through a wireless communication network.

[0106] The concept of machine learning itself and suitable machine learning algorithms are generally well known to those skilled in the art based on their expertise. A machine learning-based module or computer program product that can be adapted to the requirements of the present invention through the training data sets specified here can be purchased from numerous manufacturers or developers. In an exemplary method according to the invention, the estimation module is based on a machine learning algorithm selected from the group consisting of supervised learning algorithms, preferably selected from the group consisting of supervised learning algorithms for solving regression problems, and particularly preferably selected from the group consisting of artificial neural networks, and / or wherein the estimation module is obtained by applying a machine learning algorithm to the group of training data, and the algorithm is selected from the group consisting of supervised learning algorithms, preferably selected from the group consisting of supervised learning methods for solving regression problems, and particularly preferably selected from the group consisting of artificial neural networks.

[0107] The inventors specifically refer to the following publications, which are considered particularly helpful in this regard:

[0108] - R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL: https: / / www.R - project.org / .

[0109] - Mlr3: Lang M, Binder M, Richter J, Schratz P, Pfisterer F, Coors S, Au Q, Casalicchio G, Kotthoff L, Bischl B (2019). “mlr3: A modern object - oriented machine learning framework in R.” Journal of Open Source Software. doi:10.21105 / joss.01903 (URL: https: / / doi.org / 10.21105 / joss.01903), URL: https: / / joss.theoj.org / papers / 10.21105 / joss.01903.

[0110] - Mlr3viz: Michel Lang, Patrick Schratz, Raphael Sonabend, Marc Becker and Jakob Richter (2021). mlr3viz: Visualizations for'mlr3'. R package version 0.5.7. https: / / CRAN.R - project.org / package=mlr3viz.

[0111] - Tidyverse: Wickham et al., (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686, https: / / doi.org / 10.21105 / joss.01686.

[0112] To obtain the estimation module (i.e., training), a person skilled in the art uses a training set of training data, which includes corresponding absorption data sets of plant seed parts of plant seeds with known germination characteristics. These absorption data sets are referred to as training absorption data sets of the training plant seed parts in the context of the present invention. Thus, the identification module can be trained by so-called supervised learning to achieve the required functions.

[0113] Fortunately, for a person skilled in the art, obtaining a suitable training set is not a problem in practice because conducting germination tests that are usually oriented to the ISTA specification is an everyday activity for a person skilled in the art, and the data required for "supervised learning" is also binary in many cases (i.e., germinated / non-germinated). Although the prediction quality can be improved by appropriate measures if needed, as described in the practical examples below, a person skilled in the art only needs to generate the corresponding absorption data sets in the germination tests that he would conduct anyway, which can also be carried out through method steps a) to d).

[0114] In the following, an example is given to illustrate how to generate a set of exemplary training data. Additionally, in Table 3 below, a set of exemplary measured training data is disclosed. Taking rapeseed as an example, these data enable a person skilled in the art to understand the suitable format of the training data and can also be used as an initial basis for training an exemplary estimation module for testing purposes.

[0115] The inventors propose that, for example, a set of training data suitable for a plant seed can be generated as follows:

[0116] A. First, carry out the method according to the present invention in the context of a rough screening to evaluate the conditions (changing volume, concentration of redox indicator, cultivation period, cultivation temperature, etc.) under which effective differentiation of the absorption characteristics of the test system will be found.

[0117] B. When sufficient differentiation is achieved, perform method steps a) to d) on a large number of plant seeds, where the cultivation is carried out under the conditions determined according to point A. In this case, as described above, a blank system, a control system, and a reference system are used to obtain background-corrected and normalized data on the absorption characteristics of the corresponding test compositions at the first wavelength λ1 and the second wavelength λ2.

[0118] C. After that, sow the previously examined plant seeds according to the ISTA standard in a manner that still allows the recorded data sets to be assigned to the sown plant seeds. Among them (depending on the objective of subsequent training), for example, it can be determined which plant seeds germinate physiologically, the duration before physiological germination, and / or which plant seeds form "normal seedlings".

[0119] D. Although the method is non-destructive, the cultivation in the test composition exerts stress on the plant seeds. In particular, the hypoxia during complete coverage with the test composition affects the viability of the plant seeds to be tested. It is not possible to avoid this effect of cultivation on the germination characteristics when compiling the training data. However, the inventors have recognized that in order to improve the estimation quality, this effect can be compensated by subsequently calibrating the data obtained at point C based on the non-cultivated (i.e., "normal") sowing according to ISTA. For this purpose, the same number of plant seeds are sown without prior cultivation according to the same ISTA standard, the same germination characteristics are recorded, and then the calibration function is determined.

[0120] For the training data set, the inventors generally recommend a size of at least 10 batches of different batches and qualities, with at least 400 seeds per batch, from which at least a simpler estimation module can be obtained. In many cases, even a significantly smaller training data set can be used to obtain a simpler estimation module.

[0121] Using the obtained training data set, in order to obtain an estimation model in the usual way by machine learning, for example, a regression model can be trained, which is trained using the normalized absorption data set of at least two wavelengths, the corresponding normalization factors, and the recorded germination characteristics (such as physiological germination yes / no). The quality of the model can be directly verified, for example, using a certain proportion (such as about 20%) of the previously separated training data, in order to determine, for example, the mean square error and / or the maximum error of the obtained estimation module in this matter and to decide whether further training is optionally required.

[0122] Based on the above explanation, particularly preferred embodiments of the method according to the invention can be achieved individually or in combination of two or more features.

[0123] Preferred is the method according to the invention in which the set of training data is 3000 or more, preferably 4000 or more, particularly preferably 5000 or more training absorption data sets of a training plant seed portion of plant seeds with known germination characteristics.

[0124] Preferred is generally the method according to the invention in which the set of training data is obtained by performing method steps a) to d) of the method according to the invention for a plurality of training plant seed portions, preferably in substantially the same manner, in order to obtain training absorption data sets, wherein the germination characteristics of the training plant seeds are determined in a subsequent germination characteristics test.

[0125] Also preferred is a method according to the invention in which the germination characteristics recorded in subsequent tests on the germination characteristics of plant seeds for a plant seed portion are corrected by a correction factor which takes into account the germination characteristics reduced due to cultivation in the test composition, wherein the correction factor is obtained by correlation with the same batch of plant seeds not used in the method in the germination characteristics test.

[0126] It will be readily understood by those skilled in the art that the training data set and training on which the estimation module is based should preferably be as widely relevant as possible to the process control of the method according to the invention. Hyperbolically speaking, those skilled in the art do not expect an estimation module trained with a training data set generated for plant seeds of type A1, cultivation conditions B1 and wavelength C1 to be meaningfully used in a method according to the invention for processing plant seeds of type A2 under cultivation conditions B2 and determining wavelength C2 if A1, B1 and C1 are completely different from A2, B2 and C2. Thus, preferred is a method according to the invention in which the set of training data comprises a plurality of training absorption data sets obtained by method steps a) to d) of the method according to the invention for estimating germination characteristics, wherein substantially the same process parameters and / or equipment are used.

[0127] In the understanding of those skilled in the art, in particular, the method according to the invention is generally also preferred, wherein the training absorption data sets of the set of training data:

[0128] i) are obtained for plant seeds corresponding to the type of plant seeds provided in the method, and / or

[0129] ii) are obtained for a training plant seed portion comprising the same number of plant seeds as the plant seed portion provided in the method, and / or

[0130] iii) are obtained for a training test system in which a test composition having the same test volume as the test system obtained in the method is used, and / or

[0131] iv) are obtained for a training test system cultivated under the same conditions as the test system cultivated in the method, and / or

[0132] v) are obtained using the same optical measurement method, preferably using the same optical measuring device as the absorption data set obtained in the method, and / or

[0133] vi) comprise the same data on absorption characteristics as the absorption data set obtained in the method, preferably having the same data structure; wherein it is particularly preferred to set 2 or more, preferably 3 or more, particularly preferably 4 or more, most particularly preferably 5 or more conditions, in particular all of these conditions.

[0134] The present invention also relates to a particularly preferred test composition for use in the method according to the present invention, said test composition comprising:

[0135] i) water,

[0136] ii) a two-stage redox indicator,

[0137] iii) a fermenting microorganism, and

[0138] iv) a surfactant compound biocompatible with the fermenting microorganism, wherein the surfactant is selected from the group consisting of non-ionic surfactants. According to the inventors' assessment, such a test composition is particularly advantageous because it ensures good wettability of the plant seeds with the test composition without having to resort to mechanical means to increase wettability.

[0139] The present invention also relates to a kit for producing a test composition according to the present invention, said kit comprising:

[0140] A1) a starting mixture, said starting mixture comprising:

[0141] ii.b) a two-stage redox indicator,

[0142] iii.b) a fermenting microorganism, and

[0143] iv.b) a surfactant compound biocompatible with the fermenting microorganism, wherein the surfactant is selected from the group consisting of non-ionic surfactants, or

[0144] A2) in a separate container, sub-components for producing the starting mixture, and

[0145] B) production instructions specific to the plant seeds, including production specifications for producing a test composition specific to the plant seeds by mixing the starting mixture or the sub-components of the starting mixture with an aqueous solvent, in particular water.

[0146] In the inventors' assessment, it is preferred to add other components useful to the end user when performing the method according to the present invention to the kit.

[0147] Specifically, for example, a kit according to the present invention comprising a test plate having a plurality of test notches for receiving a test system from a plant seed portion and a test composition is preferred, wherein the test plate preferably comprises 45 or more, preferably 90 or more test notches.

[0148] A kit according to the present invention comprising a metering and filling aid for filling a test plate having a plurality of test notches is also preferred.

[0149] Also preferred is a kit according to the invention comprising an optical measuring device, wherein the optical measuring device is configured to determine the optical absorption properties of a test composition for electromagnetic radiation at least at a first wavelength λ1 and a second wavelength λ2, wherein λ1 and λ2 differ by 10 nm or more.

[0150] The invention also relates to a computer program product comprising commands which, when executed by an electronic data processing device, cause the device to carry out step e) of the method according to the invention, preferably method steps d) and e), wherein the computer program product comprises an estimation module based on machine learning and is preferably stored on a portable storage unit, preferably on a USB-readable data carrier.

[0151] Furthermore, an electronic data processing device for use in a method according to the invention for estimating the germination properties of plant seeds in a plant seed portion is disclosed, comprising a storage unit and an estimation module based on machine learning stored on the storage unit, wherein the electronic data processing device is configured to input an absorption data set obtained for the plant seed portion in the method according to the invention as an input into the estimation module and to use the estimation module to estimate the germination properties of the plant seeds in the plant seed portion, wherein the estimation module is trained to estimate the germination properties of the plant seeds in the plant seed portion based on the absorption data set, and wherein the training is carried out by supervised learning using a set of training data, the set of training data comprising a plurality of training absorption data sets of training plant seed portions of plant seeds with known germination properties. Description of the Drawings

[0152] Hereinafter, the invention and its preferred embodiments will be explained in more detail with reference to the drawings. These drawings show the following:

[0153] Figure 1 is a schematic flow chart of the method according to the invention; and

[0154] Figure 2 is the absorption spectra of two redox states of the two-stage redox indicator resazurin. Detailed Description

[0155] Figure 1 is a schematic view of the method steps of the method according to the invention for estimating the germination properties of plant seeds in a preferred embodiment. The estimated germination properties are the germination ability and seed vigor according to the ISTA standard.

[0156] In method step a)100 of the exemplary method according to the invention, 80 individual plant seed portions are provided, each plant seed portion consisting exactly of one rapeseed plant seed (i.e., rapeseed). The plant seed portions are arranged individually in test recesses in a so-called multi-well plate.

[0157] In method step b) 102, a test composition is produced which, based on the mass of the test composition, comprises 99.9705% distilled water, 0.0005% resazurin, 0.004% Saccharomyces cerevisiae, and 0.025% of a surfactant compound biocompatible with microorganisms, which compound is available from Evonik Operations GmbH under the brand name BreakThru SD260.

[0158] The fermenting microorganism is prepared as a powder together with resazurin and the surfactant compound in a freeze-dried state, and the powder is mixed with cold water (about 6 °C) in accordance with instructions specific to plant seeds to prepare the test composition. The test composition is produced immediately before method step c) 104 such that there is only a short time interval between the production of the test composition and its contact with the plant seed portion. This advantageously makes it possible to largely avoid the occurrence of warming and aging effects of the test composition during this period. In method step c) 104 of the method according to the invention, rapid and complete wetting of the plant seeds with the test composition is achieved by means of the surfactant compound used.

[0159] In method step c) 104, each plant seed portion is brought into contact with a test volume of 150 μL of the test composition in order to obtain 80 individual test systems. The microtiter plate with the test systems is then incubated for 4 hours at a temperature of about 21 °C in the dark. After incubation, 100 μL of the test composition is removed from each test system and transferred to a new microtiter plate for subsequent measurement.

[0160] In addition to the 80 test systems, in method step c) 104, a total of 4 blank systems (containing only 150 μL of distilled water) and a total of 4 control systems (which contain 0.3125 mmol of a first carbohydrate (i.e. sucrose) in the test composition), 4 control systems (which contain 0.078125 mmol of a second carbohydrate (i.e. sucrose) in the test composition), and 4 reference systems (which contain only 150 μL of the test composition) are produced and incubated together with the test systems on the same microtiter plate.

[0161] In method step d) 106, using a photometer of model absorbance96 from Byonoy GmbH, a transmission measurement method is adopted to finally measure the test compositions of the cultivated test system as well as the blank system, control system and reference system for electromagnetic radiation at the first wavelength λ1 = 600 nm and the second wavelength λ2 = 570 nm, so as to determine the optical absorption characteristics (OD values) of the corresponding test compositions. In this way, a number of absorption data sets corresponding to the number of test systems are obtained. Thus, the absorption data sets contain data on the absorption characteristics of the corresponding test compositions at 600 nm and 570 nm, where the values derived from the corresponding absorption values are further processed, and the values are corrected considering the blank measurement or normalized according to the control measurement.

[0162] In method step e) 108, a computer-aided analysis is performed on the absorption data sets used to estimate the germination characteristics of plant seeds in the corresponding plant seed parts. For this purpose, software with an estimation module is used, and this software is based on a machine learning algorithm, as disclosed above. For this purpose, the algorithm is trained by supervised learning using a set of training data from the training absorption data sets of the training plant seed parts of rapeseed with known germination characteristics.

[0163] In method step f) 110, then the average germination characteristic prediction is calculated in the form of a box plot by taking the average of the germination characteristics estimated for the plant seed parts. In method step g) 112, the estimated average germination characteristic prediction is output via an electronic computer.

[0164] Table 3 summarizes a set of exemplary training data, which is obtained by the above method steps a) 100 to d) 108 for 4 groups, with 80 plant seeds in each group. Thus, Table 3 contains the normalized and corrected absorptions of the plant seeds (numbered) at 570 nm (A1) and 600 nm (A2). In addition, for the purpose of supervised learning, the corresponding relevant results of the sowing test are input, that is, whether the plant seeds germinate physiologically (P) and form normal seedlings (N) (1 = yes / 0 = no).

[0165] For example, for different plant seeds or different seed batches, the inventors compared the estimation accuracy (ACC600 / 570) according to the method of the present invention and the estimation accuracy obtained when only a single wavelength was evaluated (ACC570 and ACC600). For this purpose, 80% of all available data was used for training. The remaining 20% was provided to the estimation module trained in this way and compared with the actual germination results (P and N respectively, also referred to as "ground truth"). Predictions with a value >= 0.5 were evaluated as 1, and predictions with a value < 0.5 were evaluated as 0; the percentage of correctly predicted cases is the accuracy. For example, for wheat, the effect of fungicide treatment was also estimated. For this purpose, untreated samples and samples treated with a fungicide (brand name: Vibrance Trio; Syngenta) and mixtures ("partially fungicide-treated", ratio: 1:1) were taken from the same batch and the method was performed once for each. The corresponding estimation module was trained separately for the treated or untreated seeds on the one hand and for all treated / untreated seeds on the other hand.

[0166] The results of the germination ability and seed vigor are shown in Tables 1 and 2.

[0167] Table 1 - Accuracy values when estimating germination ability

[0168]

[0169]

[0170] Table 2 - Accuracy values when estimating seed vigor

[0171]

[0172] The estimation according to the method of the present invention always shows an improved estimation quality and provides the best estimation in all cases. It should be remembered that the comparison values ACC600 and ACC570 were also evaluated using a machine learning-based estimation module, and thus the comparison values shown here already show a significantly improved estimation quality compared to the relatively simple regression methods known from the prior art.

[0173] In Figure 2In order to graphically illustrate the influence of the color change of a redox indicator, the absorption spectra of the two-stage redox indicator resazurin are graphically illustrated in different redox states A and B, where the Y-axis represents absorption and the X-axis represents the wavelength in nanometers. Absorption spectrum A was measured for resazurin and shows a maximum absorption at 600 nm. Absorption spectrum B was measured for fluorescein, i.e., for reduced resazurin, and shows a maximum absorption at 570 nm. Thus, the reduction of resazurin to fluorescein results in a shift of the absorption maximum by approximately Δ = 30 nm. In the method according to the invention, the determination of the absorption characteristics of the two absorption maxima is particularly efficient because, as can be seen from Figure 2 it can be seen that a particularly distinct difference between the absorptions of the two components can be seen at these wavelengths.

[0174] Table 3 - Exemplary training data set

[0175]

[0176]

[0177]

[0178]

[0179]

[0180] List of reference numerals

[0181] 100 Method step a)

[0182] 102 Method step b)

[0183] 104 Method step c)

[0184] 106 Method step d)

[0185] 108 Method step e)

[0186] 110 Method step f)

[0187] 112 Method step g)

Claims

1. A method for estimating the germination characteristics of plant seeds, comprising the following method steps: a) Providing a plurality of individual plant seed parts, each plant seed part containing at least one plant seed, b) Producing or providing a test composition, the test composition comprising: i) Water, ii) A two-stage redox indicator, and iii) Fermenting microorganisms, c) Contacting the plant seed parts with a corresponding test volume of the test composition to obtain a plurality of individual test systems, and culturing the test systems, d) Measuring the test composition of the cultured test systems using an optical measurement method to determine the optical absorption characteristics of the corresponding test composition for electromagnetic radiation at at least a first wavelength λ1 and a second wavelength λ2, so as to obtain a plurality of absorption data records assigned to the corresponding test systems, wherein the absorption data set includes data on the absorption characteristics of the corresponding test composition at the first wavelength λ1 and the second wavelength λ2, wherein λ1 and λ2 differ by 10 nm or more, e) Using an electronic data processing device to evaluate the absorption data set assigned to the test systems to estimate the germination characteristics of the plant seeds in the corresponding plant seed parts, wherein the electronic data processing device includes a storage unit, and an estimation module based on machine learning is stored on the storage unit, wherein the electronic data processing device is configured to input the absorption data set obtained for the plant seed parts as an input into the estimation module and use the estimation module to estimate the germination characteristics of the plant seeds in the plant seed parts, wherein the estimation module is trained to estimate the germination characteristics of the plant seeds in the plant seed parts based on the absorption data set, and the training is performed by supervised learning using a set of training data, the set of training data including a plurality of training absorption data sets of training plant seed parts of plant seeds with known germination characteristics.

2. The method according to claim 1, wherein the two-stage redox indicator is resazurin.

3. The method according to any one of claims 1 or 2, wherein the fermenting microorganisms are selected from the group consisting of single-celled fungi.

4. The method according to any one of claims 1 to 3, wherein the test composition further comprises one or more surfactant compounds biocompatible with the fermenting microorganisms.

5. The method according to any one of claims 1 to 4, wherein the surfactant compounds are selected from the group consisting of surfactants acceptable to the fermenting microorganisms.

6. The method according to any one of claims 1 to 5, wherein the first wavelength λ1 and the second wavelength λ2 differ by 15 nm or more.

7. The method according to any one of claims 1 to 6, wherein the first wavelength λ1 is in the range of 585 to 630 nm, and / or wherein the second wavelength λ2 is in the range of 540 to 585 nm.

8. A test composition for use in the method according to any one of claims 1 to 7, the test composition comprising: i) Water, ii) A two-stage redox indicator, iii) Fermenting microorganisms, and iv) a surfactant compound that is biocompatible with the fermentation microorganism, wherein the surfactant is selected from the group consisting of non-ionic surfactants.

9. A kit for producing the test composition according to claim 8, the kit comprising: A1) a starting mixture, the starting mixture comprising: ii.b) a two-stage redox indicator, iii.b) a fermentation microorganism, and iv.b) a surfactant compound that is biocompatible with the fermentation microorganism, wherein the surfactant is selected from the group consisting of non-ionic surfactants, or A2) in a separate container, sub-components for producing the starting mixture, and B) production instructions specific to plant seeds, including production specifications for producing a test composition specific to plant seeds by mixing the starting mixture with an aqueous solvent.

10. A computer program product comprising instructions that, when the program is executed on an electronic data processing device, cause the device to perform method steps e), preferably method steps d) and e), of the method according to any one of claims 1 to 8, wherein the computer program product comprises a machine learning-based estimation module.

Images