Method for estimating germination characteristics of plant seeds and test kit

Abstract

a) providing a plurality of separate groups of plant seeds, each group comprising at least one plant seed; b) preparing or preparing a test composition comprising: i) water; ii) a two-step redox indicator; and iii) a fermenting microorganism; c) contacting the groups of plant seeds with test liquid amounts of each of the test compositions to obtain a plurality of separate test systems; and culturing the test systems; d) measuring the test compositions in the cultured test systems using an optical measurement method to determine optical absorption characteristics of each of the test compositions for electromagnetic radiation at at least a first wavelength λ1 and a second wavelength λ2, thereby obtaining a plurality of absorption datasets assigned to each of the test systems, wherein the absorption datasets include data on absorption characteristics of each of the test compositions at the first wavelength λ1 and the second wavelength λ2, where λ1 and λ2 differ by 10 nm or more; and e) using a machine learning-based estimation module to evaluate the absorption datasets assigned to the test systems to estimate the germination characteristics of the plant seeds in each of the groups of plant seeds.

Description

[Technical Field]

[0001] The present invention relates to a method for predicting the germination characteristics of plant seeds, a test composition for use in such a method, a kit for preparing the corresponding test composition, and a computer program product for carrying out the steps of the corresponding method. Furthermore, a data processing device for use in such a method is disclosed.

[0002] One of the most pressing challenges facing humanity is ensuring there is enough food for a growing world population. Against this backdrop, past and future innovations in agricultural engineering and agriculture are particularly important. For many major agricultural products, obtaining high-performance seeds that reliably produce high yields of usable plants is crucial to maximizing yields on available land area.

[0003] In this context, seeds are important commodities in the agricultural sector that have strict requirements. These requirements are particularly related to the ability of seeds to exhibit sufficient germination and seed vigor after sowing. In this context, assessing seed quality and keeping that quality consistent plays a very important role, and international associations that focus on seed quality assurance, testing, and certification, such as the Association of Official Seed Analysis Systems (AOSA) and the International Seed Testing Association (ISTA), have existed since as early as 1908 and 1924.

[0004] The tests and quality standards established by ISTA for determining seed germination characteristics are the undisputed industry standard in most countries of the European Union and are continuously updated by ISTA. Therefore, in the context of this invention, ISTA methods and standards are primarily based, and in many cases, these differ only slightly from AOSA standards. [Background technology]

[0005] In summary, the conventional method for determining germination characteristics involves extracting a representative portion of seeds from the seed batch being evaluated, placing the extracted test seeds in an environment in which the seeds can germinate, determining whether or not each seed in the test group germinates and the degree of seed vigor, and deriving the germination characteristics specific to the seed batch through probabilistic evaluation.

[0006] Although the tests specified by ISTA have undoubtedly had the greatest impact on industry to date, determining germination characteristics by actual germination of seeds is often considered to be disadvantageous in terms of the time and effort required. Against this background, there is still growing interest in providing a method that can reliably and quickly estimate the germination characteristics of plant seeds while still correlating well with the germination characteristics specified by ISTA.

[0007] A germination test for plant seeds known from the prior art is disclosed, for example, in German Patent Application Publication No. 102020200567. This germination test is based on a method in which multiple plant seeds are placed in separate containers and contacted with a predetermined amount of a test solution containing the two-step redox indicator resazurin and the fermenting microorganism yeast. The plant seeds are contacted with the test solution and incubated. It is generally believed that the germination ability of plant seeds decreases with increasing damage to the seed coat. During incubation, damage to the seed coat causes organic substances such as glucose to leak from the seed's interior into the surrounding test solution. These substances are at least partially reacted by the yeast, converting the two-step redox indicator resazurin to resorufin or dihydroresorufin. This conversion results in a color change. In German Patent Application Publication No. 102020200567, the absorbance of the test solution at 570 nm is measured and the standardized absorbance value is correlated with the actual germination characteristics obtained in the germination test described below. Using linear regression, a function can be derived for the same plant seed, estimating the germination characteristics from the absorbance. According to the inventor's assessment, the method of German Patent Application Publication No. 102020200567 essentially provides a very promising method for testing germination ability, which can be carried out much more quickly than methods based on actual sowing. This method allows for at least a rough estimation of the germination characteristics of a wide range of plant seeds, and furthermore offers the advantages of time and cost efficiency, is non-destructive, and in most cases has little impact on the environment. Summary of the Invention [Problem to be solved by the invention]

[0008] Despite the aforementioned advantages of the methods known from the prior art, many experts, including the inventors of the present invention, believe that these methods need improvement in many respects. In particular, the prediction quality obtained by the prior art methods is generally considered to be insufficient. This is because, despite the fact that very large data sets are usually required to obtain a strong correlation between absorbance at 570 nm and germination ability and sometimes complex standardization processes are required, the obtained regressions tend to vary widely. The inventors believe that these deficiencies in this core function, i.e., the estimation of germination characteristics, are the main reasons why this method for analyzing the germination ability of plant seeds has not been widely adopted in the industry, despite the above-mentioned advantages.

[0009] A primary object of the present invention is to eliminate or at least mitigate the aforementioned drawbacks of the prior art.

[0010] In particular, the present invention aims to disclose a method for estimating the germination characteristics of plant seeds, which allows for particularly reliable and accurate estimation of the germination characteristics of plant seeds.

[0011] In this context, the present invention aims to ensure that the method provided can be carried out in a non-invasive manner that does not destroy the plant seeds to be examined and is preferably as problem-free as possible in terms of environmental and health aspects.

[0012] Furthermore, the present invention aims to make the method for estimating and evaluating germination characteristics significantly faster to carry out than the germination ability tests known from the prior art, so that the method provided is more time- and cost-effective.

[0013] A further object of the present invention is to enable the provided method to be efficiently implemented in an at least partially decentralized configuration, so that estimation of the germination characteristics of plant seeds can be carried out outside of a research institution with as minimal equipment requirements as possible.

[0014] A further object of the present invention is that the method provided is applicable to a wide range of plant seeds, regardless of the plant seeds used, and allows reliable estimation of germination characteristics.

[0015] In this context, the present invention aims to provide a method that allows reliable estimation of certain germination characteristics, such as germination ability and seed vigor.

[0016] It is a further object of the present invention that the provided methods be easily automatable and suitable for high throughput. Furthermore, the provided methods are preferably configured to be particularly simple to operate and with relatively low requirements regarding education and training of personnel involved in the methods.

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

[0018] It is a further object of the present invention to provide a computer program product enabling the steps required for the provided methods to be carried out on an electronic data processing device.

[0019] A second object of the present invention is to provide an electronic data processing apparatus suitable for use in the methods provided. [Effects of the Invention]

[0020] The inventors of the present invention have surprisingly found that, starting from the germination ability test known from the prior art, which is based on the use of resazurin and yeast and the photometric evaluation of the color change of resazurin at 570 nm, the aforementioned objective can indeed be achieved if instead absorption characteristics at two sufficiently different wavelengths are detected and the absorption data set thus obtained is evaluated using machine learning techniques, as defined in the claims.

[0021] Surprisingly, by modifying the methods known from the prior art in this way, i.e. by detecting at least two wavelengths that are sufficiently differentiated from one another and by using machine learning methods, an advantageous method is obtained that can assess the germination characteristics of plant seeds with significantly improved prediction accuracy, the method thus obtained is fast, cheap, safe, seed-saving, and in particular can be used in a decentralized configuration with relatively low equipment requirements and low training requirements for the personnel using it. Compared to germination tests known from the prior art, significant and consistent improvements can be achieved in the prediction of the most important germination characteristics. [Means for solving the problem]

[0022] Therefore, the above-mentioned object is achieved by the subject matter of the invention as set forth in the claims. Preferred embodiments according to the invention are set forth in the dependent claims and in the following description. DETAILED DESCRIPTION OF THE INVENTION

[0023] In the following, preferred embodiments are combined with features of other preferred embodiments in particularly preferred embodiments. Thus, combinations of two or more of the particularly preferred embodiments below are most particularly preferred. Also preferred are embodiments in which features of any preferred embodiment are combined with one or more features that are preferred in any degree in other embodiments. Features of preferred test compositions, kits, electronic data processing devices, and computer program products can be derived from features of preferred methods.

[0024] Hereinafter, when both a specific amount or ratio and a preferred embodiment of an element, such as a two-stage redox indicator or a fermentation microorganism, are disclosed, the specific amount or ratio of the element that is preferably configured is also disclosed. Furthermore, it is disclosed that at least some of the element may be preferably configured in a corresponding specific total amount or ratio of the element, and in particular that the element that is preferably configured may also be present in a specific amount or ratio within the specific total amount or ratio.

[0025] The present invention provides a method for predicting germination characteristics of plant seeds, comprising the steps of: a) providing a plurality of separate groups of plant seeds, each group comprising at least one plant seed; b) preparing or providing a test composition comprising: i) water, ii) two-stage redox indicators, and iii) fermenting microorganisms; c) contacting the plant seeds with test volumes of each of the test compositions to obtain a plurality of separate test systems, and culturing the test systems; d) measuring the test compositions in the incubated test systems using optical metrology to determine optical absorption characteristics of each of the test compositions for electromagnetic radiation at at least a first wavelength λ1 and a second wavelength λ2, thereby obtaining a plurality of absorption data sets assigned to each of the test systems; the absorption dataset includes data about absorption characteristics of each of the test compositions at the first wavelength λ and the second wavelength λ; λ1 and λ2 differ by 10 nm or more, and e) using an electronic data processing device to evaluate the absorption data sets assigned to the test systems to estimate the germination characteristics of the plant seeds in each of the plant seed groups; the electronic data processing device includes a storage unit, and the machine learning estimation module is stored in the storage unit; the electronic data processing device is configured to input the absorption dataset obtained for the group of plant seeds into the estimation module as input values ​​and to estimate germination characteristics of plant seeds in the group of plant seeds using the estimation module; the estimation module is trained to estimate germination characteristics of plant seeds in the group of plant seeds using the absorption dataset; The training is performed by supervised learning using a training dataset that includes multiple training absorption datasets for a training group of plant seeds with known germination characteristics.

[0026] The method of the present invention is used to estimate the germination characteristics of plant seeds. The term "germination ability test" used in German Patent Application Publication No. 102020200567 differs from this term, but the inventors believe it is more appropriate. This is because, strictly speaking, the germination characteristics of plant seeds in the method of the present invention are not tested or analyzed as in the ISTA test. Rather, the method of the present invention is based on predicting the germination characteristics of ungerminated plant seeds by correlation with absorption values ​​measured in similar seeds that subsequently germinated, thereby avoiding the testing of actual germination characteristics. The term "germination characteristics," as understood by those skilled in the art, refers to more than the various evaluation criteria used by ISTA and indicates that, in addition to germination ability, additional parameters, such as seed vigor, are assigned to germination, and these characteristics can also be advantageously determined by the method of the present invention. Thus, in one example of the method of the present invention, the estimated germination characteristics indicate in each case the probability that a plant seed from the plant seed group will physiologically germinate, preferably the probability that a plant seed from the plant seed group will physiologically germinate within a predetermined observation period after sowing, and particularly preferably, the presence or absence of physiological germination is checked every 24 hours. As understood by those skilled in the art, "physiological germination" can be understood as the emergence of a root from the plant seed with a length of at least 2 mm, which is also sometimes called a "white root tip." However, in practice, the exact criteria are based in each case on the current ISTA standard. Additionally / alternatively, in one example of the method of the present invention, the estimated germination characteristics in each case include an estimate of the probability that a plant seed from the plant seed group will form a normal seedling as defined by the ISTA standard. A normal seedling as defined by the current ISTA standard is recognized when the seed forms a healthy, complete seedling that can be expected to develop into a complete plant.

[0027] Those skilled in the art will appreciate that the method of the present invention can be effectively used to determine the germination characteristics of multiple plant seeds. Indeed, similar to the method developed by ISTA, the most relevant application is to determine the germination characteristics of large seed batches by analysis of a representative portion of the plant seeds.

[0028] In the method of the present invention, in step a), first provide a plurality of plant seeds.This may be, for example, a representative portion corresponding to a large-scale seed batch, and divide these plant seeds into separate plant seed groups.In order to prevent any influence of soiling on the surface of the plant seeds, it is usually desirable to at least roughly clean the plant seeds before the method of the present invention, for example, by rinsing with distilled water.

[0029] As will be described below, the inventors' assessment is that the method of the present invention is most often used to determine overall evaluation parameters for an entire plant seed, i.e., bulk parameters (bulk characteristics) or to predict the average germination characteristics of a seed batch. Against this background, it is generally possible to use two or more plant seeds together as a plant seed group. That is, for example, 100 plant seed groups containing two plant seeds per plant seed group can be used. However, the inventors' assessment is that, in terms of the achievable resolution, it is preferable to use separate plant seeds for each plant seed group. This is to prevent differences in the germination characteristics of plant seeds from being lost through averaging, for example, when a plant seed with particularly favorable germination ability is measured together with a plant seed with lower germination ability in the same plant seed group. In substantially all embodiments, those skilled in the art will understand that the method of the present invention preferably uses the same number of plant seeds in the plant seed groups to efficiently realize experimental design. In substantially all embodiments, the method of the present invention is particularly preferably used with one or two plant seed groups, preferably one plant seed, and / or with each test system containing only one plant seed.

[0030] In step b), the test composition is prepared or supplied. The latter can be achieved, for example, by purchasing a finished test composition from a supplier. However, according to the inventor's assessment, in most cases the test composition is subject to certain degradation phenomena over time, so that preparing the test composition in the method of the present invention is particularly preferred in terms of the estimated quality that can be achieved.

[0031] For example, this can be achieved by not mixing the two-stage redox indicator and the fermentation microorganism with water until immediately prior to carrying out the step of preparing the test solution. Such process control is highly advantageous in that it prevents inappropriate color changes on the part of the redox indicator that may occur in the prepared test composition as an indication of deterioration over time. Thus, the method of the present invention is such that the test composition is prepared in step b), and the test composition is preferably prepared within 2 hours, particularly preferably within 1 hour, and most particularly preferably within 0.5 hours, particularly immediately prior to combining the seed groups.

[0032] The method of the present invention is advantageous in that it can be applied to a wide range of plant seeds. In fact, based on available test data, the inventors have not reached the conclusion that the method of the present invention is not suitable for specific types of plant seeds. Through extensive testing, the inventors have established the applicability of the method of the present invention to a wide range of plant seeds of the most widely different types. From this perspective, the method of the present invention is preferably applied to plant seeds selected from the group consisting of gymnosperms and angiosperms, preferably gymnosperms, monocotyledons, and dicotyledons, and particularly preferably plant seeds of the following plants:

[0033] sweet-tasting grasses, such as, for example, Alopecurus, Avena sativa, Barley, Rye, Wheat, Triticale and Maize; - Amaryllidaceae plants, for example Allium; - Asparagaceae plants, such as Asparagus genus; - Apiaceae, such as honeysuckle, carrot, and water parsley; -Asteraceae, for example, Helianthus; - Brassicaceae family, such as Brassica, Brassica napus, Radish, and White mustard; - Caprifoliaceae, for example, the genus Caprifoliaceae; - Chenopodiaceae, for example, the genus Chard; - Cucurbitaceae, for example, Cucumis; - Leguminosae, such as Glycine max, Lentil, Pisum sativum, Celeriac, Vicia faba, Lupinus, etc. - Lamiaceae, for example, the genus Lavandula, the genus Lavandula; - Ranunculaceae, for example, Delphinium; - Rosaceae, for example, the genera Fragaria and Spiraea; -Solanaceae, e.g., Tomato, Solanum, etc.

[0034] In the method of the present invention, it is most preferred that the plant seeds are seeds of the plants shown in Tables 1 and 2.

[0035] In the method of the present invention, various components of the test composition have different importance. According to the present invention, the solvent must contain water, since its presence is essential for achieving the desired functionality later. At least generally, other solvents can be present in addition to water, and thus the test composition as a whole is composed of an aqueous solvent. In practice, however, the inventors have found that it is advantageous in terms of environmental compatibility, the properties of the seeds in the method, and overall cost 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, water, based on the weight of the solvent. This is also advantageous in that it prevents additional solvent components from adversely affecting the activity of the fermentation microorganism and / or the color change behavior of the redox indicator. For this reason, in the present invention, the water is preferably distilled or demineralized water.

[0036] In the prior art, a test composition is sometimes referred to as a test solution. However, as understood by those skilled in the art, a test composition is not always a solution in the strict sense, i.e., a substantially homogeneous solution. As understood by those skilled in the art, the term test composition is more appropriate, since it includes, in addition to a test solution, liquid systems in which undissolved particles are optionally present, for example, due to fermentation microorganisms that are not completely dissolved, and these systems should rather be called test suspensions.

[0037] The test composition used contains a two-stage redox indicator, which can indicate a change in the redox potential of the test composition. Redox indicators themselves and their function are well known to those skilled in the art from the prior art as two-stage redox indicators. These two-stage redox indicators undergo two transitions, most often a visually noticeable color change, due to a change in absorption characteristics caused by a reduction or oxidation reaction. Two-stage redox indicators exhibit three different redox states with two redox transition points between them.

[0038] In the inventors' assessment, the concept underlying the method of the present invention can generally be implemented using any two-stage redox indicator. However, due to its applicability already known from the prior art, its favorable risk profile, and its range of absorption wavelengths in various redox states suitable for device evaluation, the method of the present invention particularly contemplates the use of resazurin, and in all embodiments, resazurin is the preferred two-stage redox indicator. Resazurin is well known to those skilled in the art from the prior art and, like other redox indicators, is commercially available from various manufacturers. In the method of the present invention, the two-stage redox indicator is preferably selected from the group consisting of phenoxazin-3-one-based dyes, e.g., resazurin, and it is preferred that the two-stage redox indicator is resazurin.

[0039] The test composition further comprises a fermenting microorganism. The term "fermenting microorganism" is clear to those skilled in the art and refers to a microorganism, such as a bacterium or a fungus, capable of microbial or enzymatic conversion of organic matter, especially glucose, by so-called fermentation. In the inventor's opinion, unicellular fungi, known to those skilled in the art in the form of yeast, such as brewer's yeast, are particularly suitable as fermenting microorganisms. Therefore, in the method of the present invention, the fermenting microorganism is preferably selected from the group consisting of unicellular fungi, preferably selected from the group consisting of yeasts, and particularly preferably selected from the group consisting of Saccharomyces cerevisiae and Saccharomyces bayanus. Regarding the preparation of the solution, the method of the present invention preferably uses a test composition prepared in step b) and a fermenting microorganism as a low-temperature-treated microorganism, preferably a freeze-dried microorganism.

[0040] In the method of the present invention, the test composition, at least in its broadest form as a test composition, has the same purpose as that known in the prior art, and the preferred test composition disclosed below has additional functions. Based on the inventors' assumptions, the process performed in the method of the present invention can be explained as follows, but is not limited to this theory. The selectively permeable membrane does not function in plant seeds in a dry state. After contact with water, the plant seeds begin to absorb water. Depending on the physiological composition, complete or incomplete reconstitution of the selectively permeable membrane occurs, resulting in varying degrees of leaching of storage substances, such as carbohydrates, proteins, or fats. In a good composition, the seeds control leaching, so these organic substances are present in only low concentrations in the test composition. In contrast, in dead plant seeds, reconstitution of the selectively permeable membrane does not substantially occur. As a result, large amounts of storage substances are leached and appear in high concentrations in the test composition. Therefore, the condition of the plant seeds correlates with the amount of organic material available to fermenting microorganisms during cultivation in the test composition. Therefore, the activity of microorganisms in the test composition varies depending on the vitality of the plant seeds. Redox indicators have been used to make various activities of micro-fermenting microorganisms detectable or even visible. This can be illustrated by taking resazurin as an example. Resazurin is a two-stage redox indicator that exhibits a dark blue-purple color when fully oxidized. After a first reduction, the color changes toward pink (irreversible), and after a second reduction, it becomes colorless (reversible). In other words, this is the method of the present invention, where during the fermentation of a fermentable compound, the fermenting microorganism can change the redox potential of the test composition, and / or the test composition is configured such that fermentation of the fermentable compound by the fermenting microorganism in the test composition first causes a first color change and then a second color change, or a first reduction and then a second reduction, in the two-stage redox indicator.

[0041] The inventors have found that, in order to obtain particularly advantageous estimation results, it is desirable to adapt the composition of the test composition, particularly the concentrations of the redox indicator and the fermentation microorganism, to the plant seed to be tested in the method of the present invention. In this regard, the inventors recommend that, in the kit of the present invention, preparation instructions specific to the plant seed are provided to enable a specific mixing of components when preparing the corresponding test composition by mixing a prepared solid, such as a powder, with water or an aqueous solvent, as disclosed below. In this regard, the inventors have succeeded in finding a generally preferred range that allows for a particularly advantageous test composition that can be used for a wide range of plant seeds without requiring any preparation specific to the plant seed. Specifically, in the method of the present invention, it is preferred that 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, based on the mass of the test composition. Further / alternatively, in the method of the present invention, the mass fraction of the two-step redox indicator in the test composition is preferably in the range of 0.00001 to 5%, preferably 0.001 to 0.5%, and particularly preferably 0.01 to 0.05%, based on the mass of the test composition. Further / alternatively, in the method of the present invention, the mass fraction of the fermentation microorganism in the test composition is preferably in the range of 0.01 to 10%, preferably 0.05 to 5%, and particularly preferably 0.1 to 0.5%, based on the mass of the test composition.

[0042] In step c), the group of plant seeds is contacted with a test volume of the test composition. As understood by those skilled in the art, this means that each plant seed is contacted with a separate test composition. Such contacting can be advantageously carried out in a suitable container. Since the method of the present invention typically involves simultaneously testing a large number of plant seeds, the use of a so-called multi-well plate is particularly desirable. Therefore, in many cases, the method of the present invention is effective in that the group of plant seeds is each contacted with a test volume of the test composition in one test well among multiple test wells of a first test plate, and separate test systems are held in multiple test wells of the first test plate.

[0043] As mentioned above, to prevent deterioration of the test composition over time, which may adversely affect the results of the method of the present invention, it is desirable to prepare the test composition as close as possible to the time of contact with the seeds. In this regard, to obtain particularly advantageous results, the inventors recommend using fermentation microorganisms that have been subjected to low-temperature treatment before mixing with the solvent, as disclosed above. Furthermore, to prevent premature increases in the activity of the fermentation microorganisms, the inventors recommend using relatively cold water in preparing the test composition. In a more advantageous embodiment, the temperature of the test composition may also be kept as low as possible until the moment of contact with the plant seeds to prevent undesirable premature increases in the fermentation microorganisms. This advantageously prevents any redox reactions in the prepared solution from causing an initial color change in the redox indicator, which could cause inaccurate results in subsequent optical measurements, compared to methods in which the solution is prepared at room temperature. Therefore, in the method of the present invention, the test composition is prepared in step b), and the water used during preparation is preferably at a temperature of 8°C or lower, preferably 7°C or lower, more preferably 6°C or lower, and particularly preferably 5°C or lower. In this regard, it is particularly preferred that the method of the present invention is such that when the test composition is brought into contact with the plant seeds, the temperature is below 8°C, preferably below 7°C, preferably below 6°C, particularly preferably below 5°C.

[0044] By contacting a large number of plant seed groups with respective portions of the respective test compositions, multiple test units are obtained. For clarity, these are referred to in the present invention as test systems. Accordingly, these test systems include test compositions and plant seed groups. As understood by those skilled in the art, the amount of test composition to be added is preferably determined according to the size and / or number of plant seeds in the plant seed group. Those skilled in the art will advantageously select a smaller amount for each small plant seed than for a relatively large plant seed group consisting of larger seeds. Furthermore, as understood by those skilled in the art, the efficiency of the passage of organic matter from the plant seed into the test composition, which can be detected by combining a fermentation microorganism and a two-stage redox indicator, usually depends substantially on the contact surface between the plant seed and the test composition. While it is generally possible to use a small amount of test composition that contacts only a portion of the plant seed, the inventors have determined that this is not a preferred process control method. At the same time, a large amount of test composition means that a large amount of redox indicator is present, but even in dead seeds, most of the redox indicator may not cause a color change. According to the evaluation by the present inventors, the amount of test solution in the test system is 1 * V s ~500 * V s in the range of, preferably 2 * V s ~250 * V s In the range of 3 * V s ~125 * V s It is particularly preferred that V s is the total volume of the plant seeds in the group of plant seeds. Further / alternatively, in the method of the present invention, it is preferable that the group of plant seeds is contacted with a test solution having a test solution volume in the range of 0.1 to 500,000 pL, preferably 1 to 50,000 pL, and particularly preferably 10 to 5,000 pL.

[0045] To improve the efficiency of the method of the present invention, the inventors recommend taking measures to improve the wettability of the plant seeds, since this will increase the interface and allow for a more efficient passage of organic matter from the plant seeds into the test composition. In particular, such measures can significantly reduce the time required for the method of the present invention.

[0046] First, as a somewhat preferred option for adjusting the wettability to an advantageous degree, the inventors recommend that the test system can be centrifuged to obtain an advantageous wettability. These embodiments can be implemented in the method of the present invention in that the contacting of the plant seeds with the test solution amount of the respective test composition before cultivation is assisted by a mechanical treatment step, preferably by immersion of the plant seeds or by centrifugation of the test system, particularly preferably by centrifugation of the test system.

[0047] However, in contrast to methods known from the prior art, the inventors believe that it is particularly advantageous and preferable to use specific surfactants, so-called detergents, in substantially all embodiments of the method of the present invention to obtain favorable wetting of the plant seeds. This can, for example, eliminate the need for centrifugation, significantly increasing the efficiency of the method of the present invention, especially in terms of the required processing time. In implementing these preferred embodiments, the inventors believe that surfactants that have no negative effect, or at least minimally negative effects, on the biological activity of fermentation microorganisms should be used rather than any desired surfactant. Therefore, it is particularly preferable for the method of the present invention that the test composition further comprises one or more surfactant compounds that are biocompatible with fermentation microorganisms, preferably in a total mass fraction ranging from 0.001 to 25%, preferably from 0.01 to 5%, and particularly preferably from 0.1 to 0.5%, based on the mass of the test solution.

[0048] The term "biocompatibility" is clear to those skilled in the art and indicates that the activity of microorganisms in the presence of a surfactant compound is reduced by less than 5%, preferably less than 1%, particularly preferably less than 0.1%, and particularly preferably not at all. Many commercial detergent suppliers indicate in their documentation the degree to which their surfactant compounds are tolerated by microorganisms, i.e., biocompatible, or can provide corresponding data upon request. Additionally / alternatively, those skilled in the art can determine whether an available surfactant compound is sufficiently biocompatible by relatively simple testing with the fermentation microorganisms used. Specifically, according to the inventors' evaluation, nonionic surfactants are most particularly suitable for use in the method of the present invention, and in particular, polysiloxane-based polymers have produced excellent results in the experiments conducted by the inventors. Particularly advantageous results were obtained using trisiloxane-based nonionic surfactants. Corresponding anionic surfactants are commercially available from various manufacturers under the trade name Break-Thru, for example, Break-Thru SD260 from Evonic Operations GmbH. Therefore, in the method of the present invention, the surfactant compound is preferably selected from the group consisting of nonionic surfactants, preferably selected from the group consisting of polysiloxane copolymers, and particularly preferably selected from the group consisting of polyether-polyalkylsiloxane copolymers.

[0049] In the method of the present invention, the resulting test system is then cultured. The culture procedure will be clear to those skilled in the art and refers to the transformation of the test system for a predetermined period of time, usually under heated conditions. In the method of the present invention, the culture time allows, in particular, organic substances to escape from the plant seeds and allow these substances to be transformed by fermentation microorganisms. In this regard, the inventors have succeeded in finding particularly advantageous culture conditions that can obtain good results for plant seeds of most industrially important crop plants. This advantageously resolves the trade-off between time efficiency and energy efficiency, on the one hand, and the feasibility of the most reliable estimation of germination characteristics, on the other hand. Therefore, in the method of the present invention, it is preferable that the culture be carried out for a period of 0.5 to 24 hours, preferably 1 to 12 hours, and particularly preferably 1.5 to 8 hours. Furthermore / alternatively, in the method of the present invention, it is preferable that the culture be carried out under light-shielded conditions. Furthermore / alternatively, in the method of the present invention, it is preferable that the culture be carried out at a temperature of 6 to 40°C, preferably 8 to 37°C, and particularly preferably 10 to 35°C. In this regard, it is generally preferred that the method of the present invention involves culturing for a predetermined period depending on the plant species, and the predetermined period is preferably specified by plant seed-specific instructions in the kit of the present invention.

[0050] As will be understood by those skilled in the art, once the incubation is complete, a test composition is obtained in which a portion of the redox indicator changes color once or twice in the incubated test system, and this test composition in the test system is measured in the further steps of the method of the present invention.

[0051] The aforementioned measurement of the test composition in the cultured test system is carried out in step d) by optical measurement. Suitable optical measurement methods are known to those skilled in the art based on their expertise, and suitable optical measurement devices, in particular high-performance digital devices, are commercially available from many suppliers. In substantially all embodiments, in view of the measurement data detected, the optical measurement method in the method of the present invention is suitably an absorption measurement method, preferably a transmittance measurement method. Since the data processing steps of the method of the present invention can be performed in the optical measurement device, it is preferred that the optical measurement unit in the method of the present invention comprises a data processing device.

[0052] Those skilled in the art will understand that one or more color changes in a two-step redox indicator will result in a shift in the wavelength of maximum absorption of the test composition, a change in at least the visible light range that will be discernible by those skilled in the art as a color change. Those skilled in the art will also understand that the methods of the present invention refer to parameters that correlate with the absorption characteristics of each test composition in an incubated test system.

[0053] In this regard, it is clear to those skilled in the art that the simplest and most straightforward absorption characteristic can be composed of a directly measured absorption value. However, at the same time, it may be desirable to determine parameters correlated with the directly measured absorption characteristic and / or values ​​derived therefrom instead of the directly measured absorption characteristic, and / or to determine these in a subsequent evaluation. In particular, the subsequent evaluation can be performed on standardized or normalized absorption values, which are corrected, for example, by calculation with reference to the absorption characteristics determined for the blind system, the reference system, and / or the control system, as described below. As understood by those skilled in the art, in this way, an absorption data set containing data on the absorption characteristics of each test composition is obtained for each test system, and these absorption characteristics may include, in addition to the directly measured absorption values ​​defined above, values ​​derived therefrom.

[0054] In contrast to methods known from the prior art, the absorption characteristics of each incubated test composition are not measured only at 570 nm. Rather, the optical absorption characteristics of each test composition are determined at a first wavelength λ1 and a second wavelength λ2, which, according to the present invention, must differ by at least 10 nm. That is, in the method of the present invention, measurements are carried out using an optical measurement device, preferably a photometer, configured to determine the optical absorption characteristics of the test composition for electromagnetic radiation at at least a first wavelength λ1 and a second wavelength λ2, where λ1 and λ2 differ by at least 10 nm.

[0055] In light of the above, an example of a method of the present invention includes an arrangement in which the absorption characteristics include at least the following: i) the absorbance value of each test composition at a first wavelength λ and a second wavelength λ; or ii) Values ​​derived from these absorption values Here, the value derived from the absorption value is preferably obtained by one or more calculations, in particular by standardization of the value or correction of the value according to an absorption value determined for example for a control system, a reference system or a blind system.

[0056] The absorption characteristics of each test composition for electromagnetic radiation at at least the first wavelength λ1 and the second wavelength λ2 may be selectively determined only at the corresponding wavelengths, for example, by using a substantially monochromatic radiation source or suitable filters. Alternatively, a spectrum over a wide wavelength range may be obtained and the first and second wavelengths may be detected based on the specific wavelengths.

[0057] In addition to the specific evaluation format of such datasets, the method of the present invention is advantageous in that the absorption datasets determined for various test systems contain data on the absorption behavior at two different wavelengths that have a predetermined minimum difference from each other. The overall absorption spectrum of the incubated test composition can be viewed, in a simplified manner, as a superposition, or overlap, of the mass fraction-weighted absorption spectra of the components contained therein, particularly the various states of the redox indicator. Based on the minimum difference between the first and second wavelengths, the absorption characteristics of the solution are determined at two wavelengths that are separated from each other. Here, the various oxidation states of the two-stage redox indicator contribute to the combined absorption spectrum to different degrees and proportions. The inventors believe that the data content of the absorption datasets thus obtained is necessary for downstream evaluation by machine learning. In their own experiments, the inventors have identified 10 nm as a suitable lower limit that allows for useful results without requiring excessive resources and while still showing sufficient differences in the contributions of the redox states. However, the inventors recommend that larger wavelength differences generally result in better estimation results. This is because, compared with a normal absorption spectrum, the relative contribution of the redox state of the redox indicator in such a case shows a more significant difference than normal. For this reason, in the method of the present invention, it is preferable that the difference between the first wavelength λ1 and the second wavelength λ2 is 15 nm or more, preferably 20 nm or more, and particularly preferably 25 nm or more.

[0058] Furthermore, although it is generally possible to measure the absorption behavior at three or more different wavelengths and further determine absorption characteristics at, for example, a third wavelength λ3 and a fourth wavelength λ4, and include the corresponding data in the absorption data set for subsequent computer-aided estimation, this is not preferred in the inventors' opinion. As will be appreciated by those skilled in the art, measuring additional wavelengths may improve the accuracy of the estimation, but this increases the cost of the measurement equipment and the required storage and computing power. In the inventors' opinion, since estimations at two wavelengths can achieve very good quality, measuring additional wavelengths is not necessary, and the additional effort is not justified, especially when determining absorption characteristics at wavelengths close to the respective absorption maxima of the redox state of the relevant redox indicator, e.g., 570 nm and 600 nm for resazurin.

[0059] In practice, the effective selection of the first and second wavelengths is substantially determined by the selection of the two-step redox indicator, whereby the inventors recommend that the first wavelength 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 redox state absorption maximum of the two-step redox indicator, and / or the second wavelength 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 redox state absorption maximum of the two-step redox indicator, provided that the aforementioned minimum difference is maintained.

[0060] For resazurin, which is considered particularly preferred as a two-step redox indicator, the inventors have achieved excellent results in their own experiments and recommend wavelength ranges that provide absorption data sets that can be evaluated in a particularly favorable manner in subsequent evaluations. Specifically, the present invention recommends that the first wavelength λ1 be in the range of 585 to 630 nm, preferably 590 to 620 nm, and particularly preferably 595 to 610 nm. And / or, the method of the present invention further recommends that the second wavelength λ2 be in the range of 540 to 585 nm, preferably 550 to 580 nm, and particularly preferably 560 to 575 nm.

[0061] According to the inventors' evaluation, it is at least theoretically possible to subject the test system to a transmittance measurement directly, i.e., without prior removal of the plant seeds. However, in practice, this can lead to essentially avoidable degradation of the obtained absorption measurement data. Therefore, in order to achieve the most effective process control, the inventors recommend taking a subvolume of each test composition from the cultured test system and measuring it by optical measurement. In this regard, in the method of the present invention, the test compositions to be measured are separated from the cultured test system before measuring the plant seeds, preferably one subvolume per test composition, particularly preferably in the range of 50 to 150 μL, and then the various subvolumes are preferably transferred to a second transparent test plate for testing.

[0062] Before describing in more detail below the subsequent evaluation of the obtained absorption data set for the purpose of estimating germination characteristics, it is desirable to address certain aspects, particularly with respect to steps a) to d). As the skilled artisan will readily appreciate, correlation between the detected absorption characteristics of the test system and the germination characteristics of the plant seeds to estimate germination characteristics is more easily possible and requires less computational power when there are fewer other confounding factors and anomalies.

[0063] This means to those skilled in the art that the absorption data sets of the test system should be generated under conditions as uniform as possible. Such correspondence is a natural correspondence for those skilled in the art when designing a continuous measurement. This means, for example, that preferred similar plant seeds are used, uniformly sized groups of plant seeds are formed, substantially equal amounts of test composition are used, the test composition used in the method of the invention is as uniform as possible, the group of plant seeds is contacted with the amount of test liquid in as uniform a manner as possible, and the absorption properties are determined, for example, under as identical conditions as possible, in particular using the same optical measuring device and measuring method.

[0064] Theoretically, multiple deviations, especially when relatively small, can be fairly well corrected in downstream evaluations by using machine learning, at least provided that available computing power is increased and a correspondingly high-performance estimation module is used. However, as will be understood by those skilled in the art, in all embodiments of the method of the present invention, it is desirable to contact a plurality of equally sized groups of similar plant seeds with substantially the same test composition in substantially equal amounts, culture them under substantially the same conditions, and measure the test compositions of the thus cultured test systems using the same optical measurement method under the same measurement conditions to determine the optical absorption characteristics of each test composition. In this regard, it is particularly preferred that the method of the present invention be such that each test system contains the same number of plant seeds and substantially the same amount of test composition. Furthermore, it is particularly preferred that the test compositions of the cultured test systems are measured under the same conditions.

[0065] In the inventors' assessment, the effective evaluation of groups of plant seeds measured together, by using a suitable comparison system, is also particularly advantageous for effective process control.

[0066] First, according to the inventors' evaluation, it is particularly desirable to prepare a so-called blind system that does not contain either plant seeds or test compositions, but only the solvent used in the test compositions. Such a blind system can be used, in particular, to correct the light absorption characteristics determined in step d) for any influence of the solvent or sample carrier. Therefore, in step c), the method of the present invention preferably prepares one or more blind systems in addition to the multiple test systems, incubates them with the test systems, and measures them in step d), where the blind system preferably contains only the solvent of the test composition, in particular water. It is particularly preferred that the method of the present invention corrects the light absorption characteristics of each test composition determined in step d) according to the light absorption characteristics of the one or more blind systems, preferably by subtracting a blind value.

[0067] Furthermore, according to the present inventors' recommendation, a control system prepared in a controlled manner using a predetermined amount of fermentable compound instead of the plant seeds may be simultaneously processed to verify that the test composition is fully functional, i.e., that the fermenting microorganism is particularly active and capable of inducing a color change in the redox indicator. Therefore, in step c) of the method of the present invention, in addition to the multiple test systems, one or more control systems are prepared, cultured together with the test systems, and measured in step d), and it is preferred that the control system contains a fermentable compound, preferably a carbohydrate, in the test composition.

[0068] The inventors believe that the use of these control systems is most highly preferred, particularly because it allows the absorption characteristics determined for each test composition of the test system to be standardized using the absorption characteristics determined for the control systems. In practice, the biological processes being investigated typically have limited reproducibility, and various environmental factors can affect the activity and intensity of microbial exudation. In practice, the use of a control system allows correction for both biotic variability, such as the actual concentration and motility, i.e., fitness, of the microorganisms used, and abiotic variability, such as temperature and water quality variations between measurements on different days and / or at different locations and / or under different weather conditions, resulting in comparable data under different conditions. Therefore, in virtually all embodiments, this standardization is preferred, as it at least partially, and in many cases almost completely, neutralizes the influence of such factors. This is advantageous, particularly because it facilitates the implementation of the method of the present invention outside of highly controlled laboratory conditions, thereby significantly facilitating the implementation of decentralized methods involving absorption measurements, for example, by farmers. Standardization can be performed, for example, using wavelength measurements corresponding to the fully oxidized form of a two-step redox indicator. In any case, it is particularly preferred that the method of the invention comprises normalizing the light absorption properties of each test composition determined in step d) to the light absorption properties of one or more control systems.

[0069] Furthermore, the inventors recommend using a reference system containing only the test composition in addition to or instead of the blind system and the control system. This reference system responds in color behavior or absorption to the initial state when fermentation has not yet caused any color change in the redox indicator. As such, the reference system is particularly well suited to setting suitable culture conditions. If there is a sufficient color difference between the control system in which fermentation occurs and the reference system, it is usually possible to determine with the naked eye alone that the culture has been carried out for a sufficiently long time. Therefore, in step c) of the method of the present invention, in addition to multiple test systems, one or more reference systems are preferably prepared, cultured together with the test systems, and measured in step d), and the reference system preferably contains only the test composition. Furthermore, in the method of the present invention, it is preferable to culture the test systems until a predetermined color difference between the control system and the reference system is observed.

[0070] In step e), the germination characteristics of the plant seeds are estimated, as will be disclosed in detail below.

[0071] In practice, a single piece of data on the estimated germination characteristics of a plant seed or a group of plant seeds is usually of little relevance. The added value of the method of the present invention is that it analyzes the germination characteristics of a large number of plant seeds representative of a larger seed batch, respectively, and derives an average germination characteristic prediction for the seed batch from the analysis, which can be used as a characteristic variable for the seed batch. Therefore, the method of the present invention is preferably used when the number of plant seed groups and test systems is 20 or more, preferably 40 or more, particularly preferably 60 or more, and most particularly preferably 80 or more. It is also preferable that the method of the present invention further comprises the following steps:

[0072] f) calculating an average predicted germination characteristic by averaging the estimated germination characteristics for said group of plant seeds; In most cases it will be desirable to output the data obtained in step e) in a suitable manner, for example via a display, possibly via an electronic interface for further data processing. Preferably, the method of the present invention also further comprises the steps of:

[0073] g) an output step, preferably carried out via a data interface of an electronic data processing device or an electronic display device, of the estimated germination characteristics or the calculated average germination characteristics prediction, in the form of an easily understandable graphical representation, e.g. a so-called box-and-whisker plot, in order to facilitate the handling of the output data, so as to allow a direct and intuitive comparison of the available data in a particularly simple manner.

[0074] In the method of the present invention, the absorption data sets assigned to the test systems for estimating the germination characteristics are evaluated by computer aid based on the concept of machine learning (sometimes called "artificial intelligence" or "AI type"), in which the absorption data sets determined as described above for the various test systems, including data on the absorption characteristics of the respective test compositions at the first wavelength λ1 and the second wavelength λ2, are input as input values ​​by an electronic data processing device to a machine learning type estimation module, which performs the actual estimation and outputs estimated data on the corresponding germination characteristics.

[0075] The electronic data processing device can be, for example, a separate computer of the end user connected to the optical measurement device. However, for an integrated method, the data processing device can also be a component of the optical measurement device. However, in a particularly preferred embodiment, the data processing device is a central electronic data processing device, for example a server in a cloud-based evaluation that can be connected in a network to multiple optical measurement devices.

[0076] In the present invention, the machine learning module is referred to as a prediction module based on its function for proper naming and identification, and the ability to predict germination characteristics from the absorption data set assigned to the test system is due to the training of the prediction module. For example, the prediction module may be part of comprehensive software or may reside on a memory unit of an electronic data processing device. Here, the term memory unit refers to a storage device accessible by the electronic data processing device, and does not necessarily have to be physically connected to the electronic data processing device, but may be accessible, for example, via a wireless communication network.

[0077] The concept of machine learning itself and suitable machine learning algorithms is generally well known to those skilled in the art based on their expertise. Machine learning type modules or computer program products that can be adapted to the requirements of the present invention using the training data set specified herein are commercially available from a large number of manufacturers or developers. In one example of the method of the present 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, particularly preferably selected from the group consisting of artificial neural networks. And / or the estimation module is obtained by applying a machine learning algorithm to a training data set, the algorithm being selected from the group consisting of supervised learning algorithms, preferably selected from the group consisting of supervised learning algorithms for solving regression problems, particularly preferably selected from the group consisting of artificial neural networks.

[0078] The inventors particularly cite the following documents as being useful in this regard: ― 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 / . ― 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. ― Tidyverse: Wickham et al., (2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686, https: / / doi.org / 10.21105 / joss.01686.

[0079] To obtain (i.e., train) the estimation module, a person skilled in the art utilizes a training set of training data comprising an absorption dataset corresponding to a plant seed group of plant seeds with known germination characteristics, which is referred to in the present invention as a training absorption dataset of a training plant seed group, and the identification module can be trained by so-called supervised learning to obtain the desired functionality.

[0080] Fortunately, obtaining a suitable training set is not a practical problem for those skilled in the art, since they usually perform germination tests according to the ISTA specifications in their daily practice, and the data required for "supervised learning" are often binary (i.e., germinate / not germinate). While the quality of predictions can be improved by suitable means if necessary, those skilled in the art only need to generate a corresponding absorption data set for any germination test, which can be achieved by steps a) to d) below, as described as an example.

[0081] The following provides an example of how an example of a training dataset can be generated. Furthermore, Tables 3-1 to 3-4 below disclose an example of an actually measured training dataset. This allows a person skilled in the art to know a suitable format for training data using the rapeseed example, and this can also be used for testing as a first standard for training in an example of an estimation module.

[0082] According to the inventors' recommendations, a suitable training data set for a type of plant seed may be generated, for example, as follows.

[0083] A. First, in the method of the present invention, a rough screening is performed to evaluate the conditions (amount of change, concentration of redox indicator, incubation period, incubation temperature, etc.) under which significant differences in the absorption characteristics of the test system appear.

[0084] B. Once sufficient change has been achieved, a quantity of plant seeds is subjected to steps a) through d), wherein incubation is carried out under the conditions determined by item A. Also, as described above, blind, control, and reference systems are used to obtain background-corrected and standardized data regarding the absorption characteristics of each test composition at the first wavelength λ1 and the second wavelength λ2.

[0085] C. The tested plant seeds are then sown according to the ISTA standard, such that the recorded data set is assigned to the plant seeds. Here, depending on the purpose of the subsequent training, it may be possible to determine, for example, which plant seeds physiologically germinate, the time until physiological germination, and / or which plant seeds form "normal seedlings."

[0086] D. Despite its non-destructive nature, incubation in the test composition places a burden on the plant seeds. In particular, the complete coverage with the test composition results in a lack of oxygen, which can affect the vigor of the plant seeds being tested. This effect of incubation on germination characteristics cannot be avoided when collecting training data. However, the inventors have found that, in order to improve the quality of the estimation, it is possible to compensate for this effect by subsequently calibrating the data obtained in section C on seeds sown according to the same ISTA standard without prior incubation, recording the same germination characteristics, and then determining a calibration function.

[0087] For training datasets, we typically recommend at least 10 different batch sizes and 400 seed qualities, which at least allows for a simpler estimation module, and is often possible with significantly smaller training datasets.

[0088] The resulting training dataset can be used to train a regression model using, for example, the standardized absorption dataset at at least two wavelengths, the corresponding standardized coefficients, and the recorded germination characteristics (e.g., whether or not the seeds have germinated physiologically) to obtain an estimation model in the usual way by machine learning. The quality of the model can be directly verified, for example, using a portion of the training data (e.g., about 20%) that has been previously separated. This allows, for example, determining the mean square error and / or maximum error of the resulting estimation module to determine whether further training is optionally required.

[0089] From the above description, particularly preferred embodiments of the method according to the invention can be derived, which can be implemented individually or in combination of two or more features.

[0090] In the method of the present invention, the training dataset is training absorption data of a training plant seed group of plant seeds having known germination characteristics, and is preferably 3,000 or more, preferably 4,000 or more, and particularly preferably 5,000 or more training absorption datasets.

[0091] It is generally preferred that the method of the present invention involves performing steps a) to d) of the method of the present invention, preferably in a substantially identical manner, to obtain a training absorption dataset for a plurality of training plant seed groups, whereby the training dataset is obtained and the germination characteristics of the plant seeds of the training plant seed groups are determined in a subsequent germination characteristic test.

[0092] In addition, in the method of the present invention, the germination characteristics of the plant seeds of the training plant seed group recorded in the subsequent germination characteristic test are corrected by a correction coefficient that takes into account the decrease in germination characteristics due to cultivation in the test composition, and it is preferable that the correction coefficient is obtained by correlation with plant seeds from the same lot as the lot used in the germination characteristic test but that are not used in the method of the present invention.

[0093] Those skilled in the art will readily understand that the training data set and training on which the estimation module is based preferably correlate as broadly as possible with the process control of the method of the present invention. To put it bluntly, those skilled in the art would not expect that an estimation module trained on a training data set generated for plant seeds of type A1, culture conditions B1, and wavelength C1 in the method of the present invention can be meaningfully used to treat plant seeds of type A2 under culture conditions B2 and determine at wavelength C2, where A1, B1, and C1 are completely different from A2, B2, and C2. Therefore, the method of the present invention preferably uses a training data set that includes multiple training absorption data sets obtained by steps a) to d) of the method of the present invention for estimating germination characteristics, and substantially the same process parameters and / or equipment are used.

[0094] In particular, those skilled in the art will appreciate that the methods of the present invention also generally prefer that the training absorption data set of the training data set is: i) obtained for a plant seed corresponding to the type of plant seed provided in the method of the invention, and / or ii) obtained for a training plant seed group containing the same number of plant seeds as the plant seed group provided in the method of the present invention; and / or iii) obtained for a training test system using the same test composition as the test system obtained in the method of the present invention in the test liquid volume; and / or iv) obtained for a training test system cultured under the same conditions as the test system cultured in the method of the invention; and / or v) obtained using the same optical measurement method, preferably using the same measurement equipment, as the absorption data set obtained in the method of the present invention; and / or vi) The data set contains data on the same absorption characteristics as the absorption data set obtained in the method of the present invention, preferably in the same data structure, particularly preferably two or more, preferably three or more, particularly preferably four or more, most particularly preferably five or more, and in particular when all of these conditions are met.

[0095] Furthermore, it is particularly preferred in the present invention that the test composition used in the method of the present invention comprises: i) water, ii) two-stage redox indicators; iii) fermenting microorganisms, and iv) a surfactant compound that is biocompatible with said fermentation microorganism, said surfactant compound being selected from the group consisting of non-ionic surfactants. According to the inventors' assessment, this test composition is particularly advantageous because it achieves excellent wetting of the plant seeds without relying on mechanical methods to improve wetting.

[0096] The present invention also relates to a kit for preparing a test composition according to the invention, comprising: A1) a starting mixture comprising: ii.b) Two-stage redox indicators; iii.b) fermenting microorganisms, and iv.b) a surfactant compound that is biocompatible with the fermenting microorganism, selected from the group consisting of nonionic surfactants, or A2) the accessory ingredients for preparing the starting mixture, contained in separate containers; and B) Preparation instructions specific to a plant seed, comprising preparation specifications for preparing a test composition specific to a plant seed by mixing the starting mixture or the subcomponents of the starting mixture with an aqueous solvent, in particular water.

[0097] In the inventors' assessment, it may be desirable to include additional elements in the kit that may be useful to the end user in carrying out the methods of the present invention.

[0098] Specifically, for example, the kit of the present invention includes a test plate having a plurality of test wells for accommodating a test system using a group of plant seeds and a test composition, and the test plate preferably includes 45 or more test wells, preferably 90 or more test wells.

[0099] The kits of the present invention also preferably include a pouring and filling aid for filling a test plate having a plurality of test wells.

[0100] The kit of the present invention further includes an optical measurement device configured to determine the optical absorption characteristics of the test composition for electromagnetic radiation at at least a first wavelength λ1 and a second wavelength λ2, preferably λ1 and λ2 differing by 10 nm or more.

[0101] Furthermore, the present invention relates to a computer program product, which comprises commands for causing an electronic data processing device to perform step e), preferably step d) and step e), of the present invention upon execution of the program by said device, and which further comprises a machine learning type estimation module, and which is preferably stored on a portable storage unit, preferably a USB readable data carrier.

[0102] The present invention further discloses an electronic data processing device used in the method of the present invention for estimating the germination characteristics of plant seeds in a group of plant seeds, comprising a memory unit and a machine learning estimation module stored in the memory unit, wherein the electronic data processing device is configured to input an absorption dataset for the group of plant seeds obtained in the method of the present invention as an input value to the estimation module and estimate the germination characteristics of plant seeds in the group of plant seeds using the estimation module, and the estimation module is trained to estimate the germination characteristics of plant seeds in the group of plant seeds from the absorption dataset, and the training is performed by supervised learning using a training dataset including a plurality of training absorption datasets for a training group of plant seeds of plant seeds having known germination characteristics.

[0103] The invention and preferred embodiments thereof will now be described in more detail with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0104] [Figure 1] FIG. 1 is a schematic flow diagram of the method of the present invention. [Figure 2] Figure 2 Absorption spectra of the two redox states of the two-step redox indicator resazurin.

[0105] Figure 1 is a schematic diagram of a preferred embodiment of the steps of the method of the invention for the estimation of germination characteristics of plant seeds. The germination characteristics estimated are germination ability and seed vigor according to the ISTA criteria. [Example]

[0106] In step a) 100 of one example of the method of the present invention, 80 separate groups of plant seeds, each consisting of only one rapeseed or oilseed rape plant seed, are provided and the groups of plant seeds are individually placed in test wells of a so-called multi-well plate.

[0107] In step b) 102, a test composition is prepared containing, by weight of the test composition, 99.9705% distilled water, 0.0005% resazurin, and 0.004% Saccharomyces cerevisiae, as well as 0.025% of a surfactant compound that is biocompatible with microorganisms and commercially available from Evonik Operations GmbH under the trade name BreakThru SD260.

[0108] The fermentation microorganism is prepared as a freeze-dried powder together with resazurin and a surfactant compound, which is then mixed with cold water (about 6°C) according to instructions specific to the plant seeds to prepare a test composition. The test composition is prepared just before step c) 104 so that there is only a short time lag between preparation of the test composition and contact with the plant seeds. This advantageously significantly prevents the test composition from warming and deteriorating over time. The surfactant compound used in step c) 104 of the method of the present invention allows the plant seeds to be quickly and completely wetted by the test composition.

[0109] In step c) 104, each group of plant seeds is contacted with 150 μL of test composition to obtain 80 separate test systems. The multiwell plate containing the test systems is then incubated for 4 hours at a temperature of about 21° C. under the exclusion of light. After incubation, 100 μL of the test composition is removed from each test system and transferred to a new multiwell plate for subsequent measurement.

[0110] In step c) 104, in addition to the 80 test systems, a total of four blind systems containing only 150 μL of distilled water, and a total of four control systems containing 0.3125 mmol of the first carbohydrate (i.e., sucrose) in the test composition, four control systems containing 0.078125 mmol of the second carbohydrate (i.e., sucrose) in the test composition, and four reference systems containing only 150 μL of the test composition are prepared and incubated together with the test systems on the same multi-well plate.

[0111] In step d) 106, the incubated test systems and the test compositions of the blind, control, and reference systems are finally measured using a photometer model Absorbance 96 manufactured by Byonoy GmbH in the transmittance measurement method at electromagnetic radiation with a first wavelength λ1 = 600 nm and a second wavelength λ2 = 570 nm to determine the absorbance characteristics (OD value) of each test composition. 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 absorbance characteristics of each test composition at 600 nm and 570 nm, and the values ​​obtained from the corresponding absorption values ​​are further processed, and the values ​​are corrected taking into account the standardization of the brand or control measurements.

[0112] In step e) 108, the absorption datasets are subjected to a computer-assisted analysis to estimate the germination characteristics of the plant seeds in each plant seed group. For this purpose, software is used having an estimation module based on a machine learning algorithm, which has been trained for this purpose by supervised learning using a training dataset obtained from a training absorption dataset of a training plant seed group of rapeseed with known germination characteristics, as disclosed above.

[0113] In step f) 110, the estimated germination characteristics for the plant seed population are averaged to calculate a mean germination characteristic prediction in the form of a box plot. In step g) 112, this estimated mean germination characteristic prediction is output via an electronic computer.

[0114] Tables 3-1 to 3-4 summarize an example of a training dataset generated through steps a) 100 to d) 108 for four sets of 80 plant seeds each. Therefore, Tables 3-1 to 3-4 contain the standardized and corrected absorbance at 570 nm (A1) and 600 nm (A2) for each plant seed (No.). Furthermore, for supervised learning, the corresponding results of each seeding test, i.e., whether the plant seeds germinated physiologically (P) and formed normal seedlings (N) (1 = yes / 0 = no), were entered.

[0115] As an example, for various plant seeds or various seed batches, the inventors compared the estimation accuracy (ACC600 / 570) of the method of the present invention with the estimation accuracy (ACC570 and ACC600) obtained by evaluation at only a single wavelength. For this purpose, 80% of all available data was used. The remaining 20% ​​was input into the estimation module trained in this way and compared with the actual germination results (P and N, respectively, also referred to as "true values"). (Predictions ≥ 0.5 were scored as 1, predictions < 0.5 were scored as 0, and the percentage of correct predictions represented the accuracy.) For example, for bread wheat, the effect of fungicide treatment was also estimated. For this purpose, untreated samples, samples treated with a fungicide (Syngenta, trade name: Vibrance Trio), and a mixture (1:1 ratio, referred to as "partially fungicide-treated") were taken from the same lot and each was subjected once to the method of the present invention. Each estimation module was trained on treated or untreated seeds separately on the one hand and on all treated / untreated seeds on the other hand.

[0116] The results of germination ability and seed vigor are shown in Tables 1 and 2.

[0117] [Table 1]

[0118] [Table 2]

[0119] The estimation by the method of the present invention showed a consistently improved estimation quality and achieved the best estimation in all cases. It should be noted that the comparison values ​​ACC600 and ACC570 were evaluated in comparison with methods known from the prior art using machine learning-based estimation modules, so the comparison values ​​shown here already show a significantly improved estimation quality compared to the relatively simple regression methods known from the prior art.

[0120] In Figure 2, the absorption spectra of the two-stage redox indicator resazurin in various redox states A and B are plotted to demonstrate the color change effect of the redox indicator, where the Y axis represents absorbance and the X axis represents wavelength in nanometers. Absorption spectrum A was measured for resazurin, which exhibits an absorption maximum at 600 nm. Absorption spectrum B was measured for resorufin, i.e., reduced resazurin, which exhibits an absorption maximum at 570 nm. Thus, the reduction of resazurin to resorufin caused a shift in the absorption maximum of Δ = approximately 30 nm. Determining the absorption characteristics of the two absorption maxima is particularly useful in the method of the present invention, because, as can be seen from Figure 2, at these wavelengths, a particularly significant difference is observed between the absorption of the two components.

[0121] [Table 3-1] [Table 3-2] [Table 3-3] [Table 3-4] [Explanation of symbols]

[0122] 100 process a) 102 Process b) 104 Process c) 106 Project d) 108 Project e) 110 Engineering f) 112 Engineering g)

Claims

1. A method for estimating the germination characteristics of plant seeds, a) A step of supplying multiple separate groups of plant seeds, each containing at least one plant seed. b) A step of preparing or supplying a test composition comprising the following: i) Water, ii) Two-stage redox indicators, and iii) Fermentation microorganisms, c) A step of bringing the group of plant seeds into contact with the amount of test solution of each of the test compositions to obtain a plurality of separate test systems, and culturing the test systems. d) The test composition in the cultured test system is measured using an optical measurement method, and at least the first wavelength λ １ and the second wavelength λ ２ A step of obtaining a plurality of absorption data sets assigned to each of the inspection systems by determining the optical absorption characteristics of each of the inspection compositions with respect to electromagnetic radiation in the inspection system, The absorption dataset is the first wavelength λ １ and the second wavelength λ ２ This includes data on the absorption characteristics of each of the aforementioned test compositions. λ １ and λ ２ This means there is a difference of 10 nm or more, and e) A step of estimating the germination characteristics of plant seeds in each of the plant seed groups by evaluating the absorption dataset assigned to the inspection system using an electronic data processing device, The aforementioned electronic data processing device includes a storage unit, The machine learning type estimation module is stored in the memory unit, The electronic data processing device is configured to input the absorption dataset obtained for the plant seed group as input values ​​to the estimation module, and to use the estimation module to estimate the germination characteristics of the plant seeds in the plant seed group. The estimation module is trained to estimate the germination characteristics of plant seeds in the plant seed group using the absorption dataset. The aforementioned training is performed by supervised learning using a training dataset that includes multiple training absorption datasets for a group of training plant seeds having known germination characteristics.

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

3. The method according to claim 1 or 2, wherein the fermenting microorganism is selected from the group consisting of single-celled fungi.

4. The method according to claim 1 or 2, wherein the test composition further comprises one or more surfactant compounds that are biocompatible with the fermenting microorganisms.

5. The surfactant compound is selected from the group consisting of nonionic surfactants. The method according to claim 4, wherein the nonionic surfactant is selected from the group consisting of the polysiloxane copolymer.

6. the first wavelength λ １ and the second wavelength λ ２ The method according to claim 1 or 2, wherein the difference is 15 nm or more.

7. the first wavelength λ １ is within the range of 585 to 630 nm, and / or the second wavelength λ ２ is within the range of 540 to 585 nm, the method according to claim 1 or 2.

8. A computer program product, Includes commands, The command allows the electronic data processing device to execute a program, thereby causing the device to perform step e), preferably step d) and step e) described in claim 1 or 2. Furthermore, a computer program product including the aforementioned machine learning-type estimation module.

Images