Method for determining target functional compound having target biodegradability

Abstract

The present invention relates to a method for determining a synthesis specification including a target biodegradability. A target biodegradability indicative of a biodegradation characteristic of the functional compound is provided. A digital representation of the potential target synthesis specification is provided. A habitat is provided, the habitat indicating a habitat descriptor value of the habitat descriptor. A model based on the habitat is provided, the model being adapted to determine the biodegradability of the functional compound in the habitat. The biodegradability of the potential target functional compound is determined based on the provided model and the digital representation. The determined biodegradability is then compared to the target biodegradability, and i) the potential target functional compound is determined as the target functional compound, or ii) a new potential target synthesis specification of the potential target functional compound is provided, and repeating the determination of biodegradability using the new potential target synthesis specification.

Description

Technical Field

[0001] The present invention relates to a method, an apparatus, and a computer program product for determining a target functional compound having a target biodegradability. Further, the present invention relates to a training method, a training apparatus, and a training computer program for training a data-driven biodegradation model, which can be used by the method, the apparatus, and the computer program product to determine a functional compound. Further, the present invention relates to a method and an apparatus for providing an interface for providing a target functional compound. Background Art

[0002] Generally, functional compounds refer to, for example, small molecules that are widely used in industrial and / or everyday-use products due to their wide range of application properties. The use of functional compounds particularly includes excipients, plasticizers, stabilizers, inhibitors, odorants, aromatic components, nutritional components, catalysts, radiation absorbers, lubricants, and surfactants, among others. However, on the other hand, this widespread application has led to a large amount of waste containing used functional compounds. When disposed of in a non-specified environment, non-degradable waste is a problem. In particular, the accumulation of chemicals in the environment, such as the accumulation of phosphates that cause algae blooms, is undesirable. Therefore, there is a need not only for functional compounds that will decompose, but also for knowledge about the biodegradability of functional compounds to be taken into account at an early stage of the product design process. In particular, it would be advantageous if it were possible to predict, during the design process, functional compounds that provide a specific biodegradability and are also suitable for the intended application. Therefore, it would be advantageous to provide the possibility of predicting functional compounds having a biodegradability suitable for the application in an accurate and computationally inexpensive manner. Summary of the Invention

[0003] An object of the present invention is to provide a method, an apparatus, and a computer program product that allow the determination of a target functional compound having a target biodegradability that allows accurate determination and is computationally inexpensive. Further, another object of the present invention is to provide a training method, a training apparatus, and a computer program product that allow the provision of a biodegradation model that can be used in the method, the apparatus, and the computer program and can be trained to provide good determination accuracy by using fewer computational resources.

[0004] In a first aspect of the present invention, a computer-implemented method for determining a target functional compound having a target biodegradability is provided, wherein the method comprises a) providing the target biodegradability, wherein the biodegradability indicates the biodegradation characteristics of the functional compound, b) providing a digital representation of a potential target functional compound, c) providing a biodegradation habitat, wherein the biodegradation habitat indicates habitat descriptor values of habitat descriptors that affect the biodegradation of the functional compound in the corresponding habitat, wherein the habitat descriptor indicates an environmental characteristic of the habitat, d) providing a biodegradation model based on the provided biodegradation habitat, wherein the biodegradation model is adapted to determine the biodegradability of the functional compound in the corresponding biodegradation habitat, wherein the biodegradation model is a data-driven model parameterized with respect to the biodegradation habitat to determine the biodegradability of the functional compound based on the digital representation of the functional compound, e) determining the biodegradability of the potential target functional compound based on the provided biodegradation model and the digital representation, and f) comparing the determined biodegradability of the potential target functional compound with the target biodegradability, and based on the comparison, i) determining the potential target functional compound as the target functional compound, or ii) providing a new potential target functional compound and repeating the determination of the biodegradability with the new potential target functional compound.

[0005] Since the biodegradation model is particularly suitable for determining the biodegradability of potential target functional compounds with respect to a specific biodegradation habitat, which is characterized by corresponding habitat descriptor values that affect the biodegradability of the functional compound in the respective habitat, the biodegradability of the functional compound, particularly the potential target functional compound, in the respective habitat can be determined very accurately. In addition, since the biodegradation model has been specifically trained for one or more specific biodegradation habitats, the training data required for training becomes less, and the biodegradation model becomes more flexible in determining the biodegradation of new functional compounds that are not part of the training data set. Thus, an accurate determination of the biodegradability of potential target functional compounds at low computational cost is provided. Since the determination of the target functional compound is based on an accurate and computationally inexpensive determination of biodegradability, the method also allows for an accurate and computationally inexpensive determination of the target synthesis specifications that indicate the target functional compounds having the corresponding target biodegradability. In addition, the current test methods for testing the biodegradability of functional compounds are extremely time-consuming and may take months or years to obtain results, while the above method allows for the provision of results essentially immediately, particularly for potentially suitable functional compounds. Thus, not only can the technical requirements for biodegradability determination be reduced, but also the time required for designing new biodegradable products can be significantly shortened. In addition, providing the simple possibility of having an accurate determination of biodegradability already considered during the product design process allows for the design of products such that waste can be avoided. In particular, it can be ensured that the functional compounds used in the product will biodegrade in the respectively expected environments, such as in marine habitats.

[0006] Developing new chemical products according to application requirements is a prominent issue in the modern chemical industry. Recently, additional requirements have also been proposed, relating to the environmental impact of chemical products throughout their life cycle. An important aspect of environmental impact is preventing the accumulation and bioaccumulation of chemicals in the environment. Chemical accumulation is an increasingly serious problem, which can be avoided if the functional compound materials are biodegradable. Bioaccumulation is the gradual accumulation of a compound in an organism and can be avoided if the functional compound is biodegradable. To evaluate biodegradation, a series of standardized tests are currently used. For biodegradability, there are tests with various specified conditions (e.g., ISO13432, ISO14852, ISO14855, ISO17556, and OECD 301). Standardized tests usually strike a balance between time efficiency tests (minimum 14 days, maximum 24 months) and actual conditions. In practice, higher temperatures than real conditions are often used to speed up the test time. Companies developing new functional compounds need to invest significant resources in self-assessing product sustainability and certification. The overall biodegradability assessment, including laboratory space and equipment, becomes expensive and time-consuming. Therefore, it is necessary to identify the biodegradability of new materials as early as possible in the development process. The proposed method for determining biodegradability as disclosed herein enables the method for developing new materials to be faster and more efficient. At an early stage, even before synthesizing the functional compound, the biodegradability can be determined. This allows determining whether the functional compound is suitable for market entry. This results in a faster time to market. This also allows reducing waste generation because it is not necessary to synthesize the functional compound to determine biodegradability. The proposed method provides a digital twin for measuring the biodegradability of functional compounds.

[0007] In addition, standard measurements and tests for biodegradability are usually time-consuming, for example, including waiting times of several months or even years. In particular, when developing new functional compounds for corresponding applications, these time-consuming tests can greatly limit the development process. In this context, the present invention allows providing results for new functional compounds that immediately greatly reduce the time after obtaining the results.

[0008] In addition, due to the incredibly high number of possible functional compounds that are usually even not fully explored and potentially applicable to a specific application, today, technical product engineers are given the technical task of finding functional compounds that are not only applicable to a specific application but also meet the corresponding target properties, especially the target biodegradability. It is necessary to synthesize and test a huge number of possible functional compounds, or browse through huge datasets and libraries storing potential functional compounds in order to find the corresponding functional compounds that may be suitable for the application. Even when using a sophisticated design of experimental methods, it is still necessary to synthesize and experimentally test an extremely high number of possible functional compounds. In this context, the above method allows to assist a user, such as a technical product engineer, to find potentially suitable functional compounds automatically and faster. In particular, by using the above method, the user only needs to synthesize and test potentially suitable functional compounds for which it has been determined that they are very likely to meet the corresponding target properties, especially the target biodegradability. Thus, unnecessary synthesis and testing of functional compounds can be avoided. Therefore, the method allows the user to perform the technical task of finding functional compounds suitable for technical applications faster and more efficiently.

[0009] The method refers to a computer-implemented method and can therefore be executed by a general-purpose or special-purpose computer suitable for executing the method, for example by executing a corresponding computer program. The method is suitable for determining a target functional compound having target biodegradability.

[0010] A biodegradable functional compound refers to a functional compound that can be degraded through biological processes. In particular, a biodegradable functional compound can refer to a functional compound that can be assimilated by bacteria and / or fungi to produce environmentally friendly products, i.e., functional compounds that decompose into non-polluting residues. Generally, biodegradability indicates the biodegradation characteristics of a functional compound. In particular, biodegradability refers to a measure of the degradation (i.e., decomposition) of a functional compound caused by biological processes (i.e., processes involving biological materials, especially microorganisms involved in the degradation process). Thus, biodegradability does not refer to a pure chemical degradation process that does not include microbial activity. Biodegradability is an inherent property of a functional compound. In this context, an inherent property of a functional compound refers to a property of a functional compound that is caused by the nature of the functional compound (i.e., its structure, composition, etc.) and thus reflects the essence of the functional compound, in a particular context. In particular, when present in a specific bioactive environment, biodegradability reflects the essence of the functional compound. Target biodegradability can refer to any quantification of the biodegradability of a functional compound. For example, target biodegradability can refer to only one value, such as the half-life of a functional compound in a corresponding habitat, or can refer to more than one value, such as a degradation function of a functional compound over time in a specific habitat. Preferably, the target biodegradability of a functional compound refers to any one of the mineralization characteristics, biotransformation characteristics, and / or decomposition half-life of the functional compound. Preferably, the target biodegradability is provided in the form of a percentage value of biodegradation after a predetermined time range.

[0011] Biodegradable compounds can be designed to degrade upon disposal through the action of living organisms. Biodegradability may be related to the environmental fate and / or behavior of a compound. Biodegradability may be related to the extent to which a compound can be decomposed by microorganisms such as bacteria, fungi, or algae. Biodegradability can depend on the compositional components of the chemical structure of the compound, molecular weight, physical factors (such as crosslink density, branching, crystallinity, or solubility), and exposure conditions (such as habitats, e.g., soil, compost, or aquatic systems). Regarding exposure conditions, microorganisms, microbial populations, nutrient concentration, temperature, pH, pO2, ionic conditions, or substrate characteristics (such as toxicity) affect biodegradability. Biodegradability can be measured based on the measured mass loss (mg / time), dissolved organic carbon (DOC, organic carbon concentration / time), oxygen consumption (e.g., by pressure measurement, e.g., Pa / time), or carbon dioxide production over time (e.g., by pressure measurement, e.g., Pa / time).

[0012] To quantify biodegradability in the sense of the measured properties of a compound, many measurement criteria have been developed. Different measurement methods have been defined to determine biodegradability under predefined laboratory conditions. For example, for Wastewater OECD Test No. 301: "Ready Bio-degradability" (July 17, 1992) describes six methods for determining biodegradability. Additionally, for example, ASTM D5988-18 "standard test method for determining aerobic biodegradation of plastic materials in soil" describes measuring the carbon dioxide produced by microorganisms as a function of exposure time to measure the degree of biodegradability relative to a reference material. Furthermore, for example, ISO 17556:2019 "plastics—determination of the ultimate aerobic biodegradability of plastic materials in soil by monitoring the oxygen demand in a respirometer or the amount of carbon dioxide evolved" generates the optimal biodegradation rate of plastic materials in test soil by controlling oxygen consumption or carbon dioxide production.In addition, for example, ISO 14855-1:2012 "Determination of the ultimate aerobic biodegradability of plastic materials under controlled composting conditions - Method by analysis of evolved carbon dioxide - Part 1: General method" and ASTM D5338-15 "Standard test method for determining aerobic biodegradation of plastic materials under controlled composting conditions, incorporating thermophilic temperatures" determine the ultimate aerobic biodegradability of organic compound-based plastics under controlled composting conditions (the way in which microorganisms completely consume chemical or organic substances in the presence of oxygen) by measuring the percentage of carbon converted to carbon dioxide and the degree of disintegration of the plastic at the end of the test. ASTM D6400-21 "Standard specification for labeling of plastics designed to be aerobically composted in municipal or industrial facilities" additionally includes elemental analysis, plant germination (phytotoxicity), and sieve filtration of the resulting granules. In ISO 17088:2021 "Plastics - Organic recycling - Specifications for compostable plastics", it includes an assessment of the negative impacts on the composting process and facilities, as well as on the quality of the resulting compost, including the presence of high levels of regulated metals and other harmful components.

[0013] For aerobic biodegradation, ISO 18830:2016 "Plastics - Determination of aerobic biodegradation of non-floating plastic materials in a seawater / sandy sediment interface - Method by measuring the oxygen demand in a closed respirometer" and ISO 19679:2020 "Plastics - Determination of aerobic biodegradation of non-floating plastic materials in a seawater / sediment interface - Method by analysis of evolved carbon dioxide" have been developed. The biodegradation assessment is measured by the oxygen demand or the evolution of CO2.Other standards include, for example, ISO 14853:2016 “Plastics — Determination of the ultimate anaerobic biodegradation of plastic materials in an aqueous system — Method by measurement of biogas production”, ISO 23977-1:2020 “Plastics — Determination of the aerobic biodegradation of plastic materials exposed to seawater — Part 1: Method by analysis of evolved carbon dioxide” and ISO 23977-2:2020 “Plastics — Determination of the aerobic biodegradation of plastic materials exposed to seawater — Part 2: Method by measuring the oxygen demand in closed respirometer”.

[0014] The quantitative biodegradation properties of a compound can depend on the measurement method and conditions used, the measurement environment and the measured values related to the degradation process, such as mass loss over time, DOC, oxygen consumption or carbon dioxide production. The measurement method and the measured characteristics can be provided as metadata for each measurement point related to biodegradability.

[0015] Generally, a functional compound can be any functional compound. A functional compound is typically a compound having application properties for a technical purpose (i.e., fulfilling a function in a chemical product). For example, a compound that provides UV protection in a sunscreen substance is a functional compound. A functional compound can contain an active ingredient, i.e., the ingredient that provides the function of the functional compound. Additionally, the active ingredient can provide the biological activity of the functional compound. For example, functional compounds can refer to antifungal agents, aroma chemicals, UV absorbers, food additives, vitamins, nutrients, dyes, surfactants. Functional compounds are typically characterized by their chemical structure. However, different chemical structures can also exist in one functional compound. For example, a functional compound can be composed of components that are molecules undergoing tautomerism, protonation or deprotonation, etc. A functional compound can be composed of more than one stereoisomer. Thus, a functional compound composed of one type of molecule can be associated with one or more chemical structures, such as a protonated structure and a non-charged structure and / or two stereoisomers. Therefore, generally, a functional compound can include all molecules related to one chemical formula through (de)protonation, isomerization (such as tautomerism and stereoisomerization). Additionally, a functional compound can refer to any functional compound that can be described by one or more chemical structures. In an embodiment, the chemical structure can be associated with one chemical formula.

[0016] Preferably, the functional compound consists of small molecules. Preferably, the functional compound has a molecular mass of less than 10,000 g / mol. More preferably, the compound has a molecular weight of less than 800 g / mol, even more preferably less than 400 g / mol. In addition, it is preferred that the functional compound is present in the environment in a form that allows the molecule of the functional compound to be fully described using a simple structural formula containing relevant information. A simple molecular structure refers to a molecule that can be clearly described by the covalent bonding between the atoms of the molecule. Examples of situations where this is not the case are, for example, systems with a dynamic equilibrium between several forms (such as monomers and oligomers) in the case of several inorganic acids, or ionic substances with very localized charges that have a strong interaction with the solvent (e.g., via hydrogen bonding). Preferably, the functional compound has at least one of the following properties: having an impact on living organisms, being suitable for affecting the structure of living organisms, or being suitable for affecting the function of living organisms. In one embodiment, the functional compound contains at least one of the following functional groups: ether group, hydroxyl group, peroxy group, hydroperoxide group, carboxyl group, carboxyl group derivative, carbonyl group, amine group, imine group, hydrazine group, urea group, urethane group, thiourethane group, nitrile group, azide group, azo group, cyanate group, isocyanate group, isocyanide group, pyridine group, alkane group, alkene group, alkyne group, phenyl group, ketone group, thione group, aldehyde group, thioaldehyde group, acetal group, ketal group, oxime group, hydrazone group, nitro group, nitroso group, thiol group, sulfide group, disulfide group, sulfonic acid derivative, sulfinic acid derivative, sulfate group, sulfate group derivative, sulfone group, sulfoxide group, mercapto group, sulfide group, phosphino group, phosphate group, phosphoric acid derivative, phosphonate, phosphine group, silyl group, silylamino group, organosilicon group, borate group, borane group, halide group, or a combination thereof. Preferably, the functional compound corresponds to one of the following compound classes: carboxyl derivative, ether group, amine group, hydroxyl group, carbonyl group, alkane group, alkene group, benzene derivative, pyridine derivative, halide group.

[0017] In a first step, the method includes providing a target biodegradability indicative of the biodegradation properties of a functional compound. In particular, providing may refer to receiving the target biodegradability from an input of a user using, for example, a corresponding input unit. Additionally, providing may also refer to accessing a storage unit that has stored the target biodegradability capacity. Further, the providing may also include receiving the target biodegradability capacity from other sources via, for example, a network connection and providing the received biodegradability capacity. Generally, the target biodegradability may refer to a target value, such as a target half-life of the functional compound in a specific habitat, or may refer to a range of values that the functional compound should meet in a specific habitat. Additionally, the target biodegradability capacity may also refer to any type of target function, such as, for example, a timely sequence of biodegradation. For example, the target biodegradability may indicate that the target functional compound should have a first range of biodegradability values during a first time range and then a second target biodegradability value range during a subsequent time range. Such a more complex target biodegradability may be advantageous in cases where it is desired that the functional compound not be biodegraded for a period of time (e.g., for the average usage time of the functional compound) in a specific habitat and then be rapidly biodegraded in the same or another habitat.

[0018] The method includes providing a digital representation of a potential target functional compound. In particular, providing may refer to receiving the digital representation from an input of a user using, for example, a corresponding input unit. Additionally, providing may also refer to accessing a storage unit that has stored the digital representation. The digital representation of the functional compound may be any representation that provides information allowing the definition of the functional compound and / or the derivation of corresponding characterization parameters (e.g., the physicochemical properties of the functional compound). Preferably, the digital representation includes an indication of characterization parameters relating to the chemical structure of the functional compound. Based on the digital representation, the physicochemical properties may optionally be derived and used.

[0019] In a preferred embodiment, the numerical representation is of the chemical structure and / or the synthetic specification of the functional compound. As described above, the functional compound may also comprise more than one chemical structure. In such a case, preferably, the numerical representation provides a chemical structure associated with and / or indicative of the ratio of one or more structural formulas and the number of more than one structural formula present in the functional compound. Preferably, at least two structural formulas provided for the functional compound by the numerical representation correspond to structural formulas related via chemical equilibrium. Generally, the synthetic specification includes instructions on how a particular associated functional compound can be produced. For example, the synthetic specification may refer to components, starting products, and / or production conditions which, if applied, will result in the synthesis of the functional compound during the production process. Thus, when the synthetic specification is executed, for example, using suitable laboratory or industrial equipment, the synthetic specification is always associated with the functional compound produced. In particular, the synthetic specification can also be regarded as a recipe for how to produce the associated functional compound. Additionally, if at least one synthetic specification of the functional compound is known, it is preferred to provide the synthetic specification as part of the numerical representation. However, in some cases, the target functional compound is first determined and then the synthetic specification of the target functional compound is determined. This approach can be particularly useful for potential target compounds for which the synthetic specification is not yet known.

[0020] Since the structural formula of the functional compound may be affected by the habitat, preferably, the chemical structure provided by the numerical representation depends on the habitat. For example, in an aqueous medium, a certain molecule may tend to exist in a protonated form, where the ratio of unprotonated to protonated is 1:3. In such a case, a numerical representation of such a molecule by a single chemical structure may be insufficient. Such a molecule can be represented by a numerical representation that includes a quantitative ratio indicating the equilibrium between different structures associated with one chemical formula, where more than one structure is in chemical equilibrium. Due to the incorporation of the statistical frequency of the molecules associated with each structure, the numerical representation can be referred to as a statistical representation. To illustrate the concept, as an example, the equilibrium of morpholine and one of its protonated structures, protonated morpholine, is illustrated by morpholine + H+(25%) protonated morpholine (75%). Both morpholine and protonated morpholine can be the result of introducing morpholine into water at a certain pH value, with a tendency towards protonated morpholine. For example, the ratio can be 1:3 of protonated morpholine to morpholine. Thus, the numerical representation of introducing morpholine into water can refer to the specification of the structures of morpholine and protonated morpholine and the corresponding amounts or ratios of the amounts of more than one structural formula present in the functional compound. Examples of the specification of the chemical structure can be the number and type of atoms and their corresponding connectivity. Another example includes the use of SMILE and / or SMARTS to represent the chemical structure of the functional compound.

[0021] In addition, the numerical representation can also be a chemical graph of a functional compound. A chemical graph is a representation of the structural formula of a compound according to graph theory. For example, a chemical graph can be a labeled graph where vertices correspond to the atoms of the compound and edges correspond to chemical bonds. Then, the numerical representation can refer to a graph where the atoms of the molecule are nodes and the atomic bonds of the molecule are the edges of the graph.

[0022] Preferably, the numerical representation includes characterization parameters, in particular the physico-chemical properties of the potential target functional compound. In particular, the physico-chemical properties of a functional compound can be quantified by physico-chemical parameters. Preferably, the numerical representation directly includes physico-chemical parameters, preferably referring to descriptors. In particular, the physico-chemical parameters indicate parameters that quantify the physico-chemical properties of the functional compound. In this context, the term "physico-chemical properties" refers to the physical and / or chemical properties of the functional compound. However, a numerical representation can also be provided such that it allows the derivation of physico-chemical properties, for example, in the form of descriptors, for example, by providing a representation of the potential target synthesis specifications for which the corresponding physico-chemical properties have been stored or can be determined, for example, by calculation with the corresponding descriptors. Preferably, the numerical representation refers to at least one of the synthesis specifications, structural formula, trade name, IUPAC name, chemical identifier, and CAS number of the functional compound.

[0023] The potential target synthesis specifications can also be regarded as starting synthesis specifications that indicate which functional compound or which region of the potential functional compound space should be determined first when looking for synthesis specifications leading to a functional compound with target biodegradability. Usually, the numerical representation of the potential target synthesis specifications can be provided by the user or provided automatically, for example, according to predefined rules or can also be provided arbitrarily. For example, the user can select promising potential target synthesis specifications as a starting point. However, any target synthesis specifications can also be used, or a set of rules can be used to provide potential target synthesis specifications without user intervention. Preferably, the potential target synthesis specifications are provided based on rules that take into account the constraints of the potential target synthesis specification space (i.e., the target functional compound space).

[0024] Preferably, the physico-chemical parameters refer to at least one of compositional descriptors, count descriptors, lists of structural fragments, fingerprints, graph invariants, 3D descriptors, and / or higher-dimensional descriptors, which indicate parameters that quantify the physico-chemical properties of the functional compound. In a preferred embodiment, the descriptor refers to a 3D descriptor, in particular a quantum chemical descriptor. In addition, the inventors have found that, in particular, the molar mass very accurately describes the biodegradation of the functional compound. Therefore, it is particularly preferred that the physico-chemical parameters include the molar mass of the functional compound. The possible physico-chemical parameters are defined in more detail below.

[0025] The compositional descriptor can refer to any one of electric potential, molecular weight, charge, spin, boiling point, melting point, enthalpy of fusion, dissociation constant, Hansen parameters, proton, polar and dispersion contributions, Abraham parameters, retention index, total polar surface area, receptor binding constant, Michaelis-Menten constant, inhibitor constant, mutagenicity, LD50, bioconcentration, toxicity, biodegradation profile, and viscosity.

[0026] The count descriptor can refer to any one of the sum of atomic electronegativities, the sum of atomic polarizabilities, the number of atoms and non-H atoms, the number of H, B, C, N, O, P, S, Hal, and heavy atoms, the number of H-donor and H-acceptor atoms, the number of bonds, the number of non-H or multiple bonds, the number of double bonds, triple bonds, and aromatic bonds, the number of functional groups, the ratio of functional groups, the sum of bond orders, the aromatic ratio, the number of rings or circuits, the number of unpaired electrons, the number of rotatable bonds, the rotatable bond fraction, and the number of conformational isomers.

[0027] The physicochemical parameter related to the list of structural fragment descriptors can refer to at least one of the list of molecular fractions, the list of functional groups, the list of bonds, and the list of atoms. The fingerprint descriptor preferably includes at least one of MACCS keys, preferably in bit format or total format, Morgan and other circular fingerprints, preferably in bit format or total format, topological torsion, atom pairs, infrared and related spectra, fingerprint counts, PubChem fingerprints, substructure fingerprints, and Klekota-Roth fingerprints. The graph invariant / topological index descriptor preferably includes at least one of topological structure indices and topological chemistry indices. In addition, the molecular graph representation can be used as a descriptor.

[0028] In a preferred embodiment, the physicochemical parameters of the functional compound are 3D descriptors, including at least one of the following: molecular volume, average volume per atom, molecular area, average area per atom, solvent-accessible surface, dispersion energy, dielectric energy, H-donor, H-acceptor, polar and non-polar surface areas, atom-resolved H-donor, H-acceptor, polar and non-polar surface areas, shape, sphericity, dipole and higher electric moments, polarizability, dielectric energy, proton, polar and non-polar surface areas, orbital energy and orbital gap, ionization energy, electron affinity, hardness, electronegativity, electrophilicity, excitation energy and intensity, infrared and ultraviolet absorption bands, reactivity measurements, redox potential, bond critical points, partial charges, charge surface area, atomic orbital contributions, bond order, atomic radius. In particular, preferably, the physicochemical parameters of the functional compound refer to 3D descriptors, including at least one of the following: molecular volume, average volume per atom, molecular area, average area per atom, solvent-accessible surface, dispersion energy, dielectric energy, H-donor, H-acceptor, polar and / or non-polar surface areas, atom-resolved H-donor, H-acceptor, polar and / or non-polar surface areas, shape, sphericity, cone angle, polarizability, dielectric energy, proton, polar and / or non-polar surface areas, excitation energy and intensity, infrared and / or ultraviolet absorption bands, reactivity measurements, particle charge and / or charge surface area. Higher-dimensional descriptors that may preferably be utilized may include at least one of the following: conformational partition function, solubility, vapor pressure, activity coefficient, diffusion coefficient, partition coefficient, interfacial activity, rotational constant, moment of inertia, radius of gyration, density, viscosity, conformation-weighted volume and area, conformation-weighted H-donor, H-acceptor, proton, polar and / or non-polar surface areas, charge distribution, conformational dipole moment, and molecular refraction. Higher-dimensional descriptors are preferably used, which include at least one of solubility, vapor pressure, and activity coefficient, interfacial activity, conformation-weighted H-donor, H-acceptor, proton, polar and non-polar surface areas, and charge distribution.

[0029] The physicochemical parameters of the functional compound can be encoded in a molecular graph representation, e.g., a molecular fingerprint such as an extended connectivity fingerprint, one-hot encoding, word embedding, or graph-level embedding. For example, a machine learning feature model can be used to generate a predetermined molecular fingerprint including a predetermined quantity of characterization parameters. Such a machine learning model can be trained in supervised or unsupervised training using a corresponding molecular database. Preferably, a graph neural network is used as the feature model. Preferably, the molecular graph representation reflects the similarity of the properties of the functional compound. In an embodiment, a machine learning-based feature model can be used as part of a biodegradation model to determine a corresponding molecular graph representation from a digital representation of the functional compound, which can then be used as an input to a second machine learning model, as part of the biodegradation model, and then biodegradation is determined based on the molecular graph representation.

[0030] The method further includes providing a biodegradation habitat, where the biodegradation habitat indicates a habitat descriptor value of a habitat descriptor that affects the biodegradation of a functional compound in the corresponding habitat. In particular, providing may refer to receiving the biodegradation habitat from an input by a user using, for example, a corresponding input unit. Additionally, providing may also refer to accessing a storage unit on which the biodegradation habitat has been stored. Further, providing may also refer to a preset of the biodegradation-capable habitat. For example, if the method is used in a very specific context that is only sensitive to a specific biodegradation habitat, the corresponding biodegradation habitat may be preset and thus does not have to be provided as a specific input. Additionally, the providing may further include, for example, directly receiving the habitat descriptor value of the habitat descriptor from other sources via a network connection and providing the received habitat descriptor value of the habitat descriptor as the biodegradation habitat. The provided biodegradation habitat may refer to a general habitat, for example, may refer to a marine habitat, where the corresponding habitat descriptor value of the habitat descriptor of the habitat has been stored in an accessible corresponding storage device. However, the provided biodegradation habitat may also directly include the corresponding habitat descriptor value of the biodegradation habitat to provide additional specifications of the biodegradation habitat. Additionally, providing the biodegradation habitat may include providing a digital representation of the biodegradation habitat, where the digital representation may then indicate the corresponding habitat descriptor value of the habitat descriptor that affects the biodegradation of the functional compound in the corresponding habitat. Additionally, in a preferred embodiment, the habitat is derived from a digital representation of a potential target synthesis specification. For example, the biodegradation habitat may be indicated by the aggregation of the corresponding potential target functional compound under normal conditions.

[0031] Generally, a habitat descriptor indicates the environmental characteristics of a habitat. In particular, the environmental characteristics of the biodegradation habitat may affect the biological activity in the corresponding habitat. For example, it may affect the presence, growth, or absence of specific bacteria. Thus, the environmental characteristics defined by the habitat descriptor also indirectly affect the biodegradation of the functional compound in the corresponding habitat. For example, if a functional compound can be biodegraded by specific bacteria that require a specific salt concentration, the functional compound will be rapidly biodegraded in a habitat that provides such a salt concentration (such as a marine habitat), but in a habitat with an inappropriate salt concentration (such as wastewater), the biodegradation will be much slower.

[0032] Preferably, the biodegradation habitat refers to any one of a marine habitat, a wastewater habitat, a lake habitat, a composting habitat, or a soil habitat. In a preferred embodiment, the biodegradation habitat refers to a marine habitat, and wherein the habitat descriptor refers to at least one of the following: salt concentration, sedimentation type, oxygen level, location, sample depth, water temperature, nutrient concentration, pH value, environmental type, and microbial community. In another preferred embodiment, the biodegradation habitat refers to a limnetic habitat, and wherein the habitat descriptor refers to at least one of salt concentration, sedimentation type, oxygen level, location, sample depth, water temperature, nutrient concentration, pH value, environmental type, and microbial community. In a particularly preferred embodiment, the biodegradation habitat refers to wastewater, and the habitat descriptor refers to at least one of the following: water temperature, microbial community, sludge concentration, nutrient concentration, pH value, test duration, and enzyme environment. Preferably, for the wastewater habitat, the biodegradability model is trained based on biodegradability determined using standard tests defined by OECD 301 and OECD 302 specifications. Additionally, preferably, the habitat descriptor values for the habitat refer to the habitat descriptor values defined by OECD 301 and OECD 302 specifications. In another preferred embodiment, the biodegradation habitat refers to soil, and the habitat descriptor refers to at least one of the following: temperature, sand content, pH value, water content, nutrient concentration, microbial community, and enzyme environment. In another preferred embodiment, the biodegradation habitat refers to compost, and the habitat descriptor refers to at least one of the following: temperature, compost activity, pH value, water content, humidity, degree of compost maturity, compost composition, compost source, nutrient concentration, microbial community, and enzyme environment. For the ocean, the following parameters can affect biodegradation: salt concentration, sediment, water temperature, bacterial culture, etc. In some examples, the marine habitat descriptors can be stored in a database together with the geographical location. Generally, the habitat can also refer to the habitat of a standard test used to determine the biodegradability of a functional compound. For example, standard tests defined by ISO13432, ISO14852, ISO14855, ISO17556, OECD 301, and OECD 302 also define the specific habitats where biodegradation occurs. Thus, the provision of a biodegradation habitat can also include providing, for example, selecting one of the standard tests via user input, wherein the habitat descriptor then refers to the specific characteristics of the test, i.e., the test environment and thus the specific characteristics of the test habitat. Additionally, the habitat can also be determined by referring to the biodegradation of a functional compound or other reference chemicals. In this case, the habitat can be provided by providing the reference and its biodegradation. In this case, the reference and its biodegradation indicate the habitat descriptor.

[0033] The method further includes providing a biodegradation model based on the provided biodegradation habitat. In particular, preferably, providing a biodegradation model means selecting a biodegradation model based on the provided biodegradation habitat. For example, multiple biodegradation models may be stored in a biodegradation storage device, where each biodegradation model has been trained for one or more different biodegradation habitats. Preferably, in particular, each biodegradation model is trained for different values or value ranges of habitat descriptor values for the biodegradation habitat. Based on the provided biodegradation habitat indicating the habitat descriptor value, an appropriate corresponding biodegradation model can be selected from the multiple biodegradation models. For example, if the indicated habitat descriptor value falls within the range of habitat descriptor values for which the biodegradation model has been trained, the biodegradation model is appropriate. For example, a corresponding lookup table may be provided that allows for easy comparison between the indicated habitat descriptor value and the descriptor value ranges for which the biodegradation models stored on the storage device have been trained, such that an appropriate biodegradation model can be directly selected. However, in another embodiment, providing a biodegradation model based on the provided biodegradation habitat may also mean that the user selects a biodegradation model. For example, preselected biodegradation models that refer to the provided biodegradation habitat may be provided to the user, and then the user is allowed to select the corresponding biodegradation model that should be used. Generally, the biodegradation models that may be stored are those that have been parameterized based on corresponding training dataset parameters for one or more habitats. Since the training dataset used to parameterize the biodegradation model is historical data, as described in more detail below, the biodegradation model can thus be trained and thus generated at any time before determining the specific biodegradation of a particular functional compound, and stored in the corresponding database after training. However, the biodegradation model that is trained and thus generated can of course also be executed when a specific biodegradation model is needed, for example for a specific habitat.

[0034] The biodegradation model can be parameterized based on a training dataset that includes the measured biodegradation in the corresponding habitat associated with the corresponding formulation in the training dataset. The measured biodegradation can be measured relative to the corresponding habitat using a predetermined biodegradation test method (e.g., any of the test methods described above). Thus, the biodegradation model represents the measured biodegradability of the training formulation.

[0035] Then, the provided biodegradation model is adapted to determine the biodegradability of a functional compound in a corresponding biodegradation habitat. In particular, the biodegradation model is a data-driven model parameterized with respect to biodegradation habitat parameters such that it determines the biodegradability of a functional compound based on a digital representation. Preferably, the biodegradation model is trained to determine biodegradation based on the characterization parameters of the functional compound indicated by the digital representation. Additionally or alternatively, the biodegradation model can be trained to determine biodegradation based on the chemical structure as described above, preferably based on the chemical formula and the amount of more than one structural formula present in the functional compound or the ratio of the corresponding chemical formulas. The term "such that" is here interpreted as parameterization adaptation and thus enables the biodegradation model to provide information on the biodegradability of the habitat when provided with the physicochemical parameters of the functional compound as input. For example, the biodegradation model correlates the physicochemical parameters of the historical digital representation of the functional compound and the historical digital representation of the habitat with biodegradability. This allows the digital representation of the functional compound to be determined based on the target biodegradability. The term "data-driven" is used here to emphasize that the model is mainly based on the corresponding data input and not on, for example, intuition, personal experience or knowledge. Preferably, the biodegradation model refers to a machine learning-based model that is based on known machine learning algorithms such as neural networks, regression models, classification algorithms, etc. It has been found that for most applications in this context, in particular, regression models based on neural networks, linear regression, Random Forests, Boosted Trees, Lasso, Ridge regression and MARS algorithms are suitable, while for classification models, in particular, Random Forests, Logistic regression and SVM algorithms are suitable. Generally, the biodegradation model is parameterized during the training process, where the digital representation of the functional compound as described above or one or more characterization parameters derived from the digital representation and / or the chemical structure are used together with the corresponding biodegradability of a specific biodegradation habitat. Based on such a training data set specific to the biodegradation habitat, for example, for a specific range and / or values of habitat descriptors, known training methods can be utilized to determine the corresponding parameters of the data-driven model such that the biodegradation model is also able to determine the biodegradability of functional compounds that are not part of the training data set.

[0036] In an embodiment, the biodegradation model is a two-stage machine learning model comprising two machine learning algorithms, where the output of the first stage is the input of the second stage, and the two machine learning algorithms are both trained simultaneously with the same training dataset. Preferably, the first stage is trained to determine one or more characterization parameters from a digital representation, whereupon the second stage is then trained to determine biodegradability using the one or more characterization parameters. In a preferred embodiment, the first stage is based on a graph neural network that is trained to determine a molecular graph representation, such as a molecular fingerprint, of the functional compound as described above based on the digital representation, and the second stage is based on a random forest algorithm and determines the corresponding biodegradation based on the molecular graph representation and optionally a habitat descriptor.

[0037] Furthermore, in a preferred embodiment, the biodegradation model may also be adapted to further determine the biodegradation of the functional compound based on the habitat descriptor value as input. In particular, the biodegradation model can be trained by using a training dataset that includes a) a digital representation of the functional compound, preferably including or indicating, for example, the physicochemical parameters, structural formula, etc. of the functional compound, and b) the relevant biodegradability of a specific habitat as described above, thereby producing a biodegradation model that indirectly takes into account the specific habitat. However, the training dataset may optionally also include the specific habitat descriptor values of the corresponding habitat. In this case, the biodegradation model can be trained such that, in addition to the above-mentioned functional compound and / or derivable characterization parameters, the habitat descriptor value can also be provided as input, whereupon the biodegradation model then further determines the biodegradability based on the habitat descriptor value. The advantage of this is that biodegradability can be determined more accurately, especially in cases where biodegradation strongly depends on the specific habitat descriptor values of the habitat. For example, in a marine habitat, the temperature or salt concentration can vary greatly in different regions of the world, and for some functional compounds, this can also result in different biodegradabilities. Therefore, for such cases, it may be advantageous to directly provide the habitat descriptor value as input to the biodegradation model. However, it is also possible not to provide the habitat descriptor value as input to the biodegradation model, but rather to train two different biodegradation models and indirectly consider different regions as different habitats.

[0038] In addition, the method includes determining the biodegradability of potential target functional compounds based on the provided biodegradation model and digital representation. In particular, as described above, a digital representation of a potential target compound provided by or derivable from the digital representation of the functional compound, the amount of chemical formula, and / or characterization parameters can be provided as input characterization parameters to the biodegradation model. The biodegradation model then provides the biodegradability of the potential target functional compound as output. If the digital representation does not directly include the corresponding input, such as physicochemical parameters, the determination of biodegradability can also include first determining the corresponding input, such as a structural formula, the amount of more than one structural formula present in the functional compound, and / or characterization parameters, as described above. Then the input thus determined can be provided to the biodegradation model.

[0039] Determining biodegradability using a biodegradation model can be regarded as a virtual measurement of biodegradability. In particular, the biodegradation model is based on measurement data, e.g., the measured biodegradability of functional compounds used to train the biodegradation model. Thus, the biodegradation model includes information provided by these previous measurements. In addition, physicochemical parameters can also refer to the measured properties of functional compounds in some cases. Therefore, the determined biodegradability of a new functional compound determined using the biodegradation model can also be regarded as being at least partially based on measurement results.

[0040] In the following step, the determined biodegradability of the potential target functional compound is compared with the target biodegradability. Based on this comparison, it is decided whether the potential target functional compound is determined to be a target functional compound, in which case the iteration can stop at this point. In addition, based on this comparison, new potential target functional compounds can also be determined, and the determination of this biodegradability is repeated using the new potential target functional compounds. In particular, as described above, the new potential target functional compounds are provided in the form of new digital representations. Thus, an iteration is performed at this point, where the biodegradation model and the characterization parameters of the potential target functional compounds are repeatedly used to determine biodegradability until a potential target functional compound is determined to be a target functional compound. In particular, the comparison can include determining whether the determined biodegradability of the potential target functional compound is within a predetermined range near the target biodegradability, in which case the target can be regarded as being met and the potential target functional compound is determined to be a target functional compound. If the determined biodegradability is outside the predetermined range near the target biodegradability, it is determined that the target is not met and new potential target functional compounds that may meet the target biodegradability are provided.

[0041] Typically, the iterations performed can refer to any search or directed search of the space of potential target functional compounds. For example, new potential target functional compounds can simply be arbitrarily selected from a large number of potential target functional compounds generated within a computer. However, specific rules for generating new potential target functional compounds can also be applied based on a comparison between the determined biodegradability and the potential target functional compounds, with or without considering the optimization of additional target properties of the functional compounds. Generally, known methods for generating new potential target functional compounds and / or new potential target functional compounds can be utilized. For example, a data-driven generation model for molecules can be used.

[0042] Then, by utilizing the biodegradation habitats and digital representations of the new potential target functional compounds as described above, iterations can be performed in the step of determining the biodegradability of the new potential target functional compounds. Optionally, if the characterization parameters have not been provided together with the digital description of the new potential target functional compounds, determining, for example, the characterization parameters from the digital representation of the new potential target functional compounds can also be part of the iteration. Additionally, it is preferred to use the same biodegradation model in all iteration steps for determining the biodegradation ability. However, in some cases, different biodegradation models can also be used for different iteration steps. For example, if other characterization parameters of the new potential target functional compounds are utilized, another biodegradation model may be more appropriate.

[0043] After the iterations have stopped, for example, after a potential target functional compound has been determined as a target functional compound, or if no new potential target functional compounds can be selected or generated, the results of the iterations can be provided to the user. For example, if no possible potential target functional compound has met the target biodegradability, the user can be notified that the determination of the target functional compound has failed. In cases where the target functional compound can be determined, the target functional compound can be provided to the user as an output. For example, the determined target functional compound can then be provided to an output unit or to a computational unit for further processing. Preferably, providing the target functional compound leads to further processing for determining the potential synthesis specifications of the target functional compound. This can be done, for example, by querying a database with stored synthesis specifications or by using a data-driven forward synthesis planning tool.

[0044] Preferably, the processing of the target functional compound includes determining the target synthesis specifications, such as those described above, and determining a control signal for controlling the production process based on the determined target synthesis specifications. Preferably, the production process refers to the production process of the target functional compound using the target synthesis specifications. Further, preferably, the target synthesis specifications refer to machine-executable synthesis specifications of the target functional compound such that the control signal can directly refer to the control of the corresponding laboratory or process equipment that allows the execution of the synthesis specifications to produce the functional compound. In an embodiment, providing the target synthesis specifications of the target functional compound includes providing a control signal that is suitable for controlling industrial equipment for producing the target functional compound according to the target synthesis specifications.

[0045] In a preferred embodiment, the method further includes providing connectivity information as a digital representation of the functional compound and determining model inputs, such as structural formulas, characterization parameters, based on the connectivity information. In particular, the connectivity information includes information about atoms, the chemical bonds between them, and the stereochemistry of the functional compound. The method then includes determining the chemical structure, structural formula, and / or characterization parameters based on the connectivity information.

[0046] In an embodiment, with reference to the intended application of the target functional compound, the target application of the functional compound is further provided, wherein a biodegradation habitat is provided based on the target application. The target application of the functional compound may refer to, for example, the intended application context of the functional compound. For example, if the functional compound is intended to be used as an excipient, plasticizer, stabilizer, inhibitor, odorant, aromatic component, nutritional component, catalyst, radiation absorber, lubricant, or surfactant. Such target applications indicate specific biodegradation habitats. For example, for a pesticide used in agricultural applications, if the functional compound biodegrades in the soil, it may be of interest. In another example, if the target application refers to using the functional compound as a surfactant in personal care products, it is likely that the functional compound will be found in the wastewater environment sooner or later. Thus, the corresponding target application indicates the corresponding biodegradation habitat. In this context, a predefined list may be provided on a storage device, on which the corresponding target applications and the corresponding biodegradation habitats are stored. Then, the target application of the functional compound may be provided, for example, by presenting a list of target applications to the user and allowing the user to select the corresponding target application, wherein the corresponding target application is associated with one or more biodegradation habitats. Then, the target functional compound may be determined for each of the biodegradation habitats associated with the target application, or the user may again select the corresponding biodegradation habitat associated with the target application. Additionally or alternatively, information indicating the end-of-life treatment of the functional compound may be provided. For example, the end-of-life treatment may indicate whether the functional compound is expected to biodegrade in a specific environment, or whether it should undergo a specific treatment, for example, in a bioreactor. Thus, as described above, the information on the end-of-life treatment may also be used to determine the biodegradation habitat of the functional compound.

[0047] In an embodiment, additional information indicating the accessible surface area of the functional compound in the intended form is provided, wherein the biodegradation model is further trained to determine biodegradability based on the accessible surface area, and wherein the method further includes determining the biodegradability on the accessible surface area. For example, the information may refer to whether the intended product is provided in a solid, powder, foam, granular, or any other form. Preferably, the information indicates the surface area per mass of the product or the geometry of the smallest independent part of the product. Generally, although the biodegradability of the functional compound is an inherent property of the functional compound, the exact timing of the biodegradability of the product containing the functional compound may also depend on, for example, the surface area accessible to the microbial components in the habitat responsible for biodegradation. Thus, further determining the biodegradability based on the surface area of the product containing the functional compound allows increasing the accuracy of the biodegradability prediction of the final product, and thus also increasing the accuracy of determining the suitable target functional compound for the final product.

[0048] In an embodiment, target technical application properties of a target functional compound are provided, and based on the provided target technical application properties, potential target functional compounds are provided such that the potential target functional compounds meet the provided target technical application properties. In particular, the technical application properties may refer to any property of a functional compound and / or a substance at least partially composed of the functional compound that allows assessment of the technical suitability of the corresponding functional compound provided after its synthesis. Preferably, the technical application properties include at least one of mechanical properties, optical properties, physicochemical properties, chemical properties, and biological properties. Generally, the mechanical properties may refer to any one of adhesion, tensile strength, stiffness, hardness, shrinkage rate, elongation rate, tear degree, tear strength, resilience, compressibility, wear, spillage amount, morphology, tactile properties, fracture stress, fracture elongation rate, particle size determination, and filling degree. The optical properties generally may include any one of coloring, turbidity, opacity, transparency, reflection, appearance, absorption, scattering, color intensity, cloud point, extinction, optical density, spectrum, refractive index. In addition, the physicochemical properties may refer to any one of the following: density, viscosity, K value, molar weight, dispersibility, particle size distribution, solubility, partition coefficient, interfacial properties, surface tension, dispersibility, storage stability, odor, segregation, coagulation, conductivity, capacitance, surface area, flow time, vapor pressure, VOC, solid content, hygroscopicity, miscibility, thixotropy, phase change properties, corrosion inhibition, solvent separation, aggregation, impact sensitivity, drying loss, contact angle, static charge, minimum film-forming temperature, and charge density. The chemical properties may include any one of the following: functional group count, atom type count, functional group density, atom type density, chemical resistance, reaction timing, crystallinity, reaction temperature, reaction pressure, decomposition, thermal decomposition, photodegradation, acidity, pK a , pH, moisture / water content, flammability, burning rate, spontaneous combustion, flash point, formation of combustible gas, reaction to fire, deflagration rate, by-product formation, salt content, temperature tolerance, oxidation properties, reduction properties, reactivity, ash content, stability, chelating ability, calorific value, saponification value. In addition, the biological properties may include any one of biodegradability, biological resistance, toxicity, biotransformation, ecotoxicology, sensitization, bacterial count, enzyme activity, distribution in the environment, bioaccumulation, and bioexposure. In a preferred embodiment, the technical application properties may also refer to biodegradability, such as biodegradability in another habitat. For example, the first target biodegradability may refer to a marine habitat, where the second target biodegradability, i.e., the technical application property in this case, may refer to wastewater.

[0049] Then potential target compounds are provided such that they meet the provided target technical application properties. For example, a database can be utilized where functional compounds and corresponding technical application properties have been stored, and target functional compounds that meet the provided target technical application properties can be selected from the database. Generally, functional compounds that meet the target technical application properties can be regarded as forming a potential target functional compound space, and this potential target functional compound space can be explored during an iterative process of searching for target functional compounds. Then, a first potential target functional compound can be selected from the selected target functional compounds that meet the target technical application properties.

[0050] In an embodiment, providing new potential functional compounds is based on modifying the provided target application properties and providing new potential target functional compounds such that the potential target functional compounds meet the modified target application properties. In particular, if new potential target functional compounds must be provided, the comparison of the determined biodegradability with the target biodegradability indicates that the determined biodegradability of the current potential target functional compounds does not meet the target biodegradability. In such a case, new potential target functional compounds can be provided such that if there are corresponding functional compounds, the new potential target functional compounds still meet the target technical application properties. However, in many cases, it is not possible to provide such new potential target functional compounds, or it may be technically unreasonable to provide such new potential target functional compounds that still meet the target technical application properties. In these cases, it is advantageous to modify the target technical application properties, for example, by using less stringent target technical application properties, such as modifying the target technical application properties such that it now refers not to a specific value but to a range of values, or if it refers to a range of values, to a wider range of values. Then new potential target functional compounds can be selected or generated such that they meet the modified target technical application properties.

[0051] In an embodiment, providing potential target functional compounds based on the provided nature of the target technical application includes using a determination model suitable for determining the nature of the technical application of a functional compound based on a digital representation of the functional compound, where the determination model is a data-driven model parameterized to determine the nature of the technical application associated with the functional compound based on a digital representation including characterization parameters of the functional compound. The determination model can refer to any known data-driven determination model that allows the nature of the technical application to be determined based on a digital representation of the functional compound including characterization parameters. Generally, preferably, the determination model follows the same principle as the biodegradability model described above. In fact, the determination model can be based on or utilize the same machine learning algorithms and training methods, only using different training data, i.e., the training data includes another corresponding nature of the technical application of the functional compound rather than biodegradability. Thus, all of the above embodiments regarding the biodegradation model can also be implemented regarding the determination model for determining the nature of the technical application. Using such a determination model has the advantage that not only can iterations be performed on the biodegradability of the functional compound, but also iterations can be performed on one or more additional natures of the technical application in a fast and computationally inexpensive manner, resulting in a target functional compound that not only meets the target biodegradability but also meets one or more additional target natures of the technical application.

[0052] In an embodiment, the method further includes providing a biodegradation test method, where the provided biodegradation test method indicates a standardized biodegradation test method for experimentally determining the biodegradation of a chemical, and a biodegradation model is further provided based on the provided biodegradation test method. Generally, there are multiple standardized biodegradation test methods for testing the biodegradation of chemicals. For example, such test methods can be found in DIN or ISO standards. Further providing the biodegradation test method and providing a biodegradation model that has been trained based on the provided biodegradation test method allows the determination of biodegradation that can be easily compared with, for example, the biodegradation measured separately using the corresponding test method. In particular, for this embodiment, preferably, the biodegradation model is trained based on a data set in which the test method based on which biodegradation has been determined is clearly specified, such that the biodegradation model can be specifically trained for one or more test methods.

[0053] In an embodiment, the habitat descriptor values of the habitat descriptor are stored in association with the corresponding geographical location, where the provided biodegradable habitat refers to the geographical location where the habitat is provided, and these habitat descriptor values of the geographical location are retrieved from the storage device. For example, the geographical location may refer to coordinates or other area identifiers. For example, the geographical location may refer to the name of a city, country, country region, sea area, geographical feature, etc. Based on such geographical locations, the corresponding habitats and / or habitat descriptors may be stored, for example, the average value or minimum and maximum values of the habitat descriptors. Thus, by providing the geographical location, the corresponding habitat descriptor values of the geographical location can be provided. This has the advantage that the user does not have to know the exact habitat or exact habitat descriptor values of the area. Thus, the user can simply provide the location where the expected target functional compound may be biodegraded in the area.

[0054] In an embodiment, the characterization parameters that can be indicated by the digital representation of the functional compound may refer to at least one of the following: formulation parameters from synthesis, composition descriptors, count descriptors, list of structural fragments, fingerprints, graph invariance, 3D descriptors, and / or higher-dimensional descriptors indicating the chemical nature of the functional compound. The corresponding associations (e.g., previously calculated) of the digital representation with the characterization parameters or further information about the functional compound may have been stored and associated with the corresponding digital representation. For example, if the digital representation refers to a brand name, the components corresponding to the brand name, the corresponding structural formula and quantity, and / or physicochemical parameters may have been stored on, for example, the storage device of the brand name owner.

[0055] In an embodiment, the method further includes providing a biodegradation test method, where the provided biodegradation test method indicates a standardized biodegradation test method for experimentally determining the biodegradation of a chemical, and a biodegradation model is further provided based on the provided biodegradation test method. For example, the test method may be any one of the following standardized test methods: ISO13432, ISO14852, ISO14855, ISO17556, OECD 301, and OECD 302.

[0056] In another aspect, an interface method for providing an interface is proposed, where the interface method includes a) receiving the target biodegradability, digital representation, and habitat as inputs via a user interface and providing the received target biodegradability, digital representation, and the habitat to a processor that executes the method as described above, and b) providing the target functional compound as a result, where the result is received from the processor that executes the method as described above.

[0057] In another aspect, a computer-implemented training method for training a data-driven biodegradation model for parameterizing the biodegradation model is proposed, wherein the training method comprises a) providing training data associated with a predetermined biodegradation habitat, wherein the training data includes i) digital representations of a plurality of training functional compounds, and ii) the biodegradability of the corresponding biodegradation habitat associated with each training functional compound, b) providing a data-driven trainable biodegradation model, c) training the provided data-driven biodegradation model based on the provided training data such that the trained biodegradation model is adapted to determine the biodegradation of a functional compound based on the digital representation of the functional compound, and d) providing the trained biodegradation model.

[0058] In yet another aspect, a device for determining a target functional compound having a target biodegradability is proposed, wherein the device comprises a) a target biodegradability providing unit for providing the target biodegradability, wherein the biodegradability indicates the biodegradation characteristics of a functional compound, b) a digital representation providing unit for providing a digital representation of a potential target functional compound, c) a habitat providing unit for providing a biodegradation habitat, wherein the biodegradation habitat indicates habitat descriptor values of habitat descriptors that affect the biodegradation of a functional compound in the corresponding habitat, wherein the habitat descriptor indicates environmental characteristics of the habitat, d) a model providing unit for providing a biodegradation model based on the provided biodegradation habitat, wherein the biodegradation model is adapted to determine the biodegradability of a functional compound in the corresponding biodegradation habitat, wherein the biodegradation model is a data-driven model parameterized with respect to the biodegradation habitat such that the biodegradation model determines the biodegradability of a functional compound based on the digital representation, e) a biodegradability determining unit for determining the biodegradability of the potential target functional compound based on the selected biodegradation model and the digital representation, and f) an iterative control unit for comparing the determined biodegradability of the potential target functional compound with the target biodegradability and, based on the comparison, i) determining the potential target functional compound as the target functional compound, or ii) providing a new potential target functional compound and repeating the determination of the biodegradability with the new potential target functional compound.

[0059] In another aspect, an interface device for providing an interface is proposed, wherein the interface device comprises a) an input interface unit for receiving a target biodegradability, a starting digital representation, and a habitat as inputs via a user interface and for providing the received target biodegradability, starting digital representation, and habitat to the device as described above, and b) a result interface for providing a habitat descriptor value of a functional compound as a result, wherein the result is received from the device as described above.

[0060] In another aspect, a training device for training a data-driven biodegradation model for parameterizing the biodegradation model is proposed, wherein the training device comprises a) a training data providing unit for providing training data associated with a predetermined biodegradation habitat, wherein the training data comprises i) digital representations of a plurality of training functional compounds, and ii) the biodegradability of the corresponding biodegradation habitat associated with each training functional compound, b) a trainable model providing unit for providing a data-driven trainable biodegradation model, c) a training unit for training the provided data-driven biodegradation model based on the provided training data such that the trained biodegradation model is adapted to determine the biodegradation of a functional compound based on the digital representation, and d) a trained model providing unit for providing the trained biodegradation model.

[0061] In another aspect of the present invention, the use of the method as described above is proposed, wherein the method is used to determine a target functional compound having a target biodegradability for any one of the following: i) a functional compound related to a nutritional component, ii) a functional compound related to a UV absorber for skin protection, iii) a formulation additive for personal care applications, iv) a functional compound for fragrance applications, v) a functional compound used as a plasticizer, vi) a functional compound used as a lubricant, and vii) a functional compound used as an active ingredient.

[0062] In another aspect of the present invention, a system is proposed, wherein the system comprises i) a control signal comprising the synthesis specifications of a functional compound indicating one or more components for producing the functional compound, wherein the control signal is generated according to the method described above, and ii) one or more components indicated by the synthesis specifications in the control signal.

[0063] In another aspect of the present invention, the use of the control signal generated according to the method described above for controlling a production process, particularly a production process including the production of a functional compound, is proposed.

[0064] In another aspect of the present invention, a control signal is provided, wherein the control signal is generated according to the above method. Preferably, the control signal includes machine-executable synthesis specifications for producing a target functional compound.

[0065] In another aspect, a computer program product for determining a target functional compound having a target biodegradability is provided, wherein the computer program product includes program code means for causing the device as described above to perform the method as described above.

[0066] In another aspect, a computer program product for training a biodegradation model is provided, wherein the computer program product includes program code means for causing the device as described above to perform the method as described above.

[0067] It should be understood that the methods as described above, the devices as described above, and the computer program products as described above have similar and / or identical preferred embodiments, in particular the preferred embodiments as defined in the dependent claims. Furthermore, the training methods as described above, the training devices as described above, and the training computer program products as described above also have similar and / or preferred embodiments, in particular the preferred embodiments as defined in the dependent claims.

[0068] It should be understood that the preferred embodiments of the present invention may also be any combination of the dependent claims or the above embodiments with corresponding dependent claims.

[0069] With reference to the embodiments described hereinafter, these and other aspects of the present invention will become apparent and be elucidated. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In the following drawings:

[0071] Figure 1 Embodiments of a system are schematically and exemplarily shown, the system including a device for determining target synthesis specifications that indicate a target functional compound having a target biodegradability;

[0072] Figure 2 A flowchart of a method for determining target synthesis specifications that indicate a target functional compound having a target biodegradability is schematically and exemplarily shown.

[0073] Figure 3 A flowchart of a method for training a biodegradation model for determining the biodegradability of a functional compound is schematically and exemplarily shown,

[0074] Figure 4 and Figure 5A flowchart schematically and exemplarily shows a preferred more detailed embodiment of a method for determining a target synthesis specification, which target synthesis specification indicates a target functional compound having a target biodegradability, and

[0075] Figures 6 to 8 A block diagram of a system architecture of a system and device for determining a target synthesis specification, which target synthesis specification indicates a target functional compound having a target biodegradability. Detailed description of the invention

[0076] Figure 1 A schematic and exemplary embodiment of system 100 is shown. The system includes a device 110 for determining a target synthesis specification, which target synthesis specification indicates a target functional compound having a target biodegradability. In addition, system 100 includes a training device 130 for training a biodegradation model utilized in device 110, a database 140 on which the results of determining the target synthesis specification can be stored, and a production system 120 for producing a product, which product particularly includes a target functional compound that can be controlled by the determined target synthesis specification of the target functional compound.

[0077] Device 110 includes a target biodegradability providing unit 111, a digital representation providing unit 112, a habitat providing unit 113, a model providing unit 114, a biodegradability determining unit 115, an iteration control unit 116, and optionally an output and / or control unit 117, which output and / or control unit is adapted to output the determined target functional compound and to determine the synthesis specification and optionally is also adapted to provide a control signal for controlling the production process of production system 120 based on the determined synthesis specification.

[0078] The target biodegradability providing unit 111 is adapted to provide a target biodegradability indicating the desired biodegradation characteristics of the functional compound. The target biodegradability providing unit 111 may refer to, for example, an input unit into which a user can input the corresponding target biodegradability. In addition, the target biodegradability providing unit 111 may refer to a user interface or may be part of a user interface that allows the user to interact with device 110 for providing the target biodegradability. However, the target biodegradability providing unit 111 may also refer to a storage unit or be communicatively coupled to a storage unit on which, for example, target biodegradabilities for specific applications have been stored.

[0079] The digital representation providing unit 112 is adapted to provide a digital representation indicative of a potential target functional compound. The digital representation providing unit 112 may refer to, for example, an input unit into which a user may input the corresponding digital representation. Additionally, the digital representation providing unit 112 may refer to a user interface or a part of the user interface that allows a user to interact with the device 110 and / or the database 140. However, the digital representation providing unit 112 may also refer to a storage unit on which a digital representation of the functional compound has been stored or is communicatively coupled to the storage unit. Generally, the digital representation may directly include, for example, the chemical structure, structural formula, and corresponding amounts and / or characterization parameters of the functional compound. However, instead of directly providing, for example, the characterization parameters, only the known synthesis specifications and / or molecular structure of the functional compound may be provided. In such a case, preferably, the digital representation providing unit 112 is also adapted to determine the corresponding other information and / or parameters, such as the chemical structure, structural formula, and corresponding amounts of more than one structural formula present in the functional compound and / or the characterization parameters from the synthesis specifications and / or molecular structure. In particular, the digital representation providing unit 112 may be adapted to determine, for example, by accessing a database on which the corresponding information of a plurality of functional compounds has been stored, the chemical structure, structural formula, and corresponding amounts of more than one structural formula present in the functional compound and / or the characterization parameters. Then, the digital representation providing unit 112 is adapted to provide the digital representation including the separately determined other information (such as the characterization parameters) to, for example, the biodegradability determination unit 115.

[0080] The habitat providing unit 113 is adapted to provide a biodegradation habitat. The habitat providing unit 113 may refer to, for example, an input unit into which a user may input the corresponding biodegradation habitat. For example, a user interface may be provided that allows the user to select from a number of predefined biodegradation habitats. In a preferred embodiment, the habitat providing unit 113 may be communicatively coupled to or refer to a user interface that allows the indication of a geographical location, for example, by marking a location on a map, by indicating coordinates, or by providing the name of a region, such as a political or geological region, where the habitat providing unit may then be adapted to provide a biodegradation habitat based on the geographical location. For example, if the geographical location indicates a specific sea area, such as the North Sea or the Atlantic Ocean, the habitat providing unit may be adapted to determine a marine habitat as the biodegradation habitat.

[0081] Typically, biodegradation habitat descriptors affect the habitat descriptor values of habitat descriptors that influence the biodegradation of functional compounds in the corresponding habitat. In particular, habitat descriptors indicate the environmental characteristics of the habitat. For example, for marine habitats, the salt concentration can strongly affect the biodegradation of functional compounds in the marine habitat. Typically, since biodegradation is determined, the chemical effect of the habitat descriptor on the functional compound is not important for this application. Thus, it is the effect of the habitat descriptor on habitat biology, particularly on the habitat microbial population, that indirectly affects biodegradation. The typical habitat descriptor values for the corresponding habitat can be stored in a database. However, the user can also input the corresponding specific habitat descriptor values, for example, if the habitat descriptor values of the corresponding habitat are known to deviate from the typical habitat descriptor values.

[0082] The model providing unit 114 is adapted to provide a biodegradation model based on the provided biodegradation habitat. In particular, preferably, the model providing unit 114 is adapted to select a biodegradation model from a plurality of biodegradation models already stored in the database. For example, a biodegradation model can be trained with respect to training data corresponding to one or more specific biodegradation habitats. These specific biodegradation habitats can be defined with respect to specific habitat descriptor values or value ranges that define for which biodegradation habitats the corresponding biodegradation model is applicable. For example, a lookup table can be provided that allows the model providing unit to select which biodegradation model is appropriate based on the biodegradation habitat, for example, based on the habitat descriptor values of the biodegradation habitat. However, the model providing unit 114 can also include or refer to an input unit to which a biodegradation model can be provided, for example, by a user selection or user input indicating which biodegradation model should be used.

[0083] The biodegradation model is a data-driven model that is parameterized to determine the biodegradability of a functional compound based on digital representations, particularly based on the chemical structure, structural formula, and the amounts of more than one structural formula present in the functional compound and / or the characterization parameters of the functional compound. Optionally, the biodegradation model can also be trained to further utilize the provided habitat descriptor values as inputs. In a preferred embodiment, the data-driven model refers to a machine learning model, for example, using an algorithm based on a regression model or an algorithm based on a classifier model. The algorithm based on a regression model can be based on any one of a neural network algorithm, a linear regression algorithm, a LASSO algorithm, a ridge regression algorithm, a MARS algorithm, a random forest algorithm, and a boosting tree algorithm. The algorithm based on a classifier model can be based on any one of a random forest algorithm, a logistic regression algorithm, and an SVM algorithm. The present inventors have found that for most applications, particularly neural networks, linear regression, random forests, and MARS-based algorithms are suitable.

[0084] For example, the training device 130 can be utilized to train the biodegradation model. In particular, the training device 130 includes a training data providing unit 131, which is used to provide training data for training the data-driven biodegradation model. The training data includes a) digital representations of a variety of training functional compounds, and b) biodegradabilities associated with each training functional compound in one or more different habitats. Optionally, the training data set may also include habitat descriptor values for a specific habitat, for which the corresponding biodegradabilities of the functional compounds in the specific habitat have been determined. Preferably, in the training data, the biodegradability provided for each training functional compound refers to the biodegradability measured according to the same measurement method. However, biodegradabilities can also be provided for different measurement methods, in which case it is preferred to clearly indicate which biodegradabilities are associated with which measurement methods, so that the biodegradability model can be trained to distinguish different measurement methods. Generally, the training data can be designed to cover the predetermined habitat space of the biodegradation model to be trained, where the habitat space is defined by the value range of the corresponding habitat descriptors for which the biodegradation model will be trained. For example, the training data can be designed to cover a predetermined type of functional compound in a predetermined habitat. Known methods for designing and optimizing the training data for the predetermined habitat space can be utilized, such that the habitat space is well covered by the training data and random outliers are avoided.

[0085] In addition, the training device 130 includes a model providing unit 132 that is adapted to provide a data-driven trainable biodegradation model, for example, a biodegradation model including parameters that can be set during a training process for training the biodegradation model. For example, the trainable biodegradation model may already be stored on a storage unit, and the model providing unit 132 can access the storage unit for providing the model. Further, the training device 130 includes a training unit 133 that is used to train the provided data-driven biodegradation model based on the provided training data. In particular, training may refer to changing the parameters of the biodegradation model based on the corresponding training data until the biodegradation model is adapted to determine the biodegradability of a functional compound based on digital representations, in particular input parameters of the functional compound (such as any one of a chemical structure, a structural formula, and the amounts of more than one structural formula present in the functional compound) and / or characterization parameters. Generally, any known training algorithm for data-driven training, in particular a machine learning-based model, can be utilized. Preferably, during the training of the biodegradation model, the input parameters of the functional compound that have the most influence on biodegradability in the corresponding habitat are also determined, and then the model is trained based on these most influential input parameters, for example, based on the most influential characterization parameters. To determine these most influential input parameters, for example, clustering analysis or PCA analysis tools can be utilized. In addition, a learning representation based on a molecular graph can be used as an input to the prediction model. In particular, the input parameters can be used to determine the application space of the training data, where the application space is then defined by the input parameters of the functional compound and the habitat descriptors covered by the data. Then, the determination of the most influential input parameters and / or habitat descriptors can be performed as a dimensionality reduction of the application space. An algorithm for optimizing the training data in the application space can then be applied, for example, covering the application space with as little training data as possible.

[0086] Then, the training device 130 includes a trained model providing unit 134 that is adapted to provide the trained biodegradation model to, for example, a storage unit on which biodegradation models trained respectively for different habitats and / or different types of functional compounds and / or characterization parameters are stored. However, the trained model providing unit 134 can also be adapted to directly provide the trained biodegradation model, for example, to the biodegradation model providing unit 114 of the device 110.

[0087] In all cases, the biodegradation model provides unit 114 and is then adapted to provide a suitable trained biodegradation model to the biodegradability determination unit 115. The biodegradability determination unit 115 can then utilize the biodegradation model and the provided digital representation for determining biodegradability. In particular, the biodegradability determination unit 115 can be adapted to utilize the input parameters indicated by the digital representation (e.g., any one of a chemical structure, a structural formula, and the amounts of more than one structural formula present in a functional compound and / or a characterization parameter) as the input to the biodegradation model, which, as described above, has been trained to then provide a determination of biodegradability as an output, and which has been trained for that biodegradability.

[0088] In addition, the device includes an iterative control unit 116, which is adapted to control an iterative process for determining a target synthesis specification. In particular, the iterative control unit 116 is adapted to compare the determined biodegradability of a potential target functional compound with a target biodegradability. Based on this comparison, the iterative control unit 116 is then adapted to decide whether additional iterative steps are needed for determining the target functional compound, or whether the iteration has ended, in particular, whether the potential target functional compound can be set as the target functional compound. Preferably, the comparison of the determined biodegradability of the potential target functional compound with the target biodegradability refers to determining whether the determined biodegradability is within a predetermined range near the biodegradability, e.g., by determining whether the difference between the determined biodegradability and the target biodegradability is below a predetermined threshold. However, the comparison can also refer to a more complex mathematical function, and the condition for determining the potential target functional compound as the target functional compound can refer to any condition based on the comparison of the determined biodegradability with the target biodegradability. Generally, if a predetermined condition is met, e.g., if the determined biodegradability is within a predetermined range near the target biodegradability, the iterative control unit 116 determines that the potential target functional compound is the target functional compound.

[0089] If the above conditions are not met, for example, if the determined biodegradability is not within a predetermined range near the target biodegradability, the iterative control unit 116 is adapted to determine that additional iterative steps are required. In such a case, the iterative control unit 116 is adapted to provide a new potential target functional compound and repeat the determination of biodegradability using the new potential target functional compound, particularly based on the digital representation of the new potential functional compound. For example, the new potential target functional compound (particularly in the form of the corresponding digital representation) can be provided on a database on which a plurality of potential target functional compounds have been stored, and the iterative control unit 116 can optionally or according to a predetermined rule select the new potential target functional compound from the database. For example, such a rule can be a function of the comparison between the determined biodegradability and the target biodegradability of the potential target functional compound. For example, the function can refer to the magnitude of the difference between the determined biodegradability and the target biodegradability, where the smaller the difference, the more similar the new potential target functional compound is to the similar part among the potential target functional compounds. In such cases, these rules can cause the iterative control unit 116 to be adapted to select a new potential target functional compound that is more similar to the potential target functional compounds if the determined biodegradability of the potential target functional compound is already similar to the target functional compound, and less similar to the potential target functional compounds if the difference between the determined biodegradability and the target biodegradability is high. However, completely different rules can also be applied. In addition, the iterative control unit 116 can also be adapted to generate, for example, based on the potential target functional compound and a predetermined rule or arbitrarily, a new potential target functional compound. Also in such a case, for the rule, the same principle as described above can be applied.

[0090] In addition, the iterative control unit 116 can also be adapted to apply a termination criterion to the iteration, which indicates that no suitable target functional compound can be found for the corresponding target biodegradability. For example, the iterative control unit 116 can be adapted to apply a termination criterion that refers to a predetermined number of iterative steps, that is, to determine a predetermined number of new potential target functional compounds. However, other termination criteria can also be used.

[0091] For example, an output unit related to a display may then be adapted to output the determined target functional compound, for example, in the form of a visual representation of the functional compound, an identification of the functional compound, a chemical formula representing the functional compound, a chemical structure of the functional compound, etc. In addition, the output unit may alternatively or additionally be adapted to provide the determined target functional compound to the database 140 for storing the corresponding determined target functional compound associated with the corresponding target biodegradability for future use. Optionally, the device 110 may include a control unit 117 that is adapted to provide a control signal based on the determined target functional compound to control the production process of the production system 120. In particular, if the synthesis specifications are not known, the control unit 117 may be adapted to determine the synthesis specifications of the target functional compound based on the corresponding known methods. Then preferably, the control signal indicates the machine-executable synthesis specifications of the target functional compound, and the machine-executable synthesis specifications are generated based on the determined synthesis specifications for producing the target functional compound that meets the target biodegradability. However, the control unit 117 may also be adapted to control the production process of another product based on the determined target functional compound, for example, by providing a control signal indicating the machine-executable synthesis specifications of another product that utilizes or includes the corresponding target functional compound.

[0092] Figure 2A flowchart of a method for determining a target functional compound having a target biodegradability is schematically and exemplarily shown. Method 200 includes a first step 210 of providing a target biodegradation ability. Additionally, in step 220, a digital representation of a potential target functional compound is provided. In particular, providing the target biodegradation ability and providing the digital representation can be respectively based on the principles regarding the target biodegradation ability providing unit 111 and the digital representation providing unit 112 described above. Further, in step 230, a biodegradation habitat is provided, and the biodegradation habitat indicates a habitat descriptor value of a habitat descriptor that affects the biodegradation of the functional compound in the corresponding habitat. Similarly for this step 230, the principles regarding the habitat providing unit 113 described above can be applied, for example. Additionally, in step 240, a biodegradation model is provided, and the biodegradation model is adapted to determine the biodegradability of the functional compound based on the digital representation. As has been discussed in more detail above, providing the biodegradation model can also refer to selecting a biodegradation model based on the provided biodegradation habitat. Further, the biodegradation model is a data-driven model parameterized with respect to the biodegradation habitat such that it can preferably determine the biodegradability of the functional compound based on the characterization parameters of the functional compound. Generally, steps 210, 220, 230, and 240 can be performed in any order or even simultaneously. In the following step 250, the biodegradability is determined based on the provided digital representation of the potential target functional compound and the biodegradation model. In step 260, the determined biodegradability of the potential target functional compound is then compared with the target biodegradability. Based on this comparison, the potential target functional compound is determined as the target functional compound, or a new potential target functional compound is provided, and the determination of the biodegradability using the new potential target functional compound is repeated. In an optional step 270 after the target synthesis specifications have been determined using the above steps, the determined target functional compound and the target biodegradability can be provided to the user via an output unit. Additionally, in step 270, the synthesis specifications of the target functional compound can be determined, and then it can be used to generate a control signal that allows control of the production process of the product, such as the production process of the target functional compound or a product containing the target functional compound, as has been described in detail above.

[0093] Figure 3 A flowchart of a method for training a data-driven biodegradation model is schematically and exemplarily shown, and the model is used, for example, in the method 200 discussed regarding Figure 2 As discussed. Generally, method 300 can be, for example, by as regarding Figure 1performed by corresponding units of the described training device 130. Method 300 includes a step 310 of providing training data for training a data-driven biodegradation model. The training data includes a) digital representations of a plurality of training functional compounds, and b) biodegradabilities associated with each of the training functional compounds in corresponding biodegradation habitats, e.g., for a specific habitat descriptor value. Optionally, the training data set may also include corresponding specific habitat descriptor values. In particular, the training data may be provided according to the principles described for the training data providing unit 131 with respect to Figure 1 The method further includes a step 320 of providing a data-driven trainable biodegradation model, e.g., a machine learning-based biodegradation model such as a neural network. Generally, step 310 and step 320 may be performed in any order, or even simultaneously. Method 300 then further includes a step 330 of training the provided data-driven biodegradation model based on the provided training data, e.g., by varying parameters in the data-driven trainable biodegradation model such that the trained biodegradation model is adapted to determine the biodegradability of a functional compound based on the digital representation of the functional compound. In step 340, the trained biodegradation model may then be provided, e.g., by storing the trained biodegradation model on a storage device or by directly providing the trained biodegradation model to device 130, as described with respect to Figure 1 described.

[0094] In the following, more detailed preferred examples of the above methods and corresponding devices will be described. Figure 4A schematic and exemplary flowchart of exemplary and preferred embodiments of the method is provided. In this exemplary embodiment, the method starts with a request for a target value of a target application, in particular a target biodegradation capacity, via a user interface, for example. Furthermore, in a next step, the optimization is initialized by providing potential target functional compounds. Optionally, for example, if the user provides such constraints, the constraints on the functional compounds can be taken into account during the process. These constraints can refer to, for example, constraints in the production of functional compounds, constraints in the starting materials that should be used for synthesizing the functional compounds, etc. Furthermore, additional application conditions that particularly indicate the biodegradation habitat of the target functional compound can be requested. Additionally, the additional application conditions can also indicate additional information about the target functional compound that should be met. For example, the requested additional application conditions can refer to a geographical location that indicates where the expected functional compound can be biodegraded, where based on these geographical locations, the biodegradation habitat and the corresponding habitat descriptors can be determined, for example, by using a database on which the corresponding associated biodegradation habitats and biodegradation physicochemical parameters have been stored. Based on the above steps, the optimization for determining the target functional compound (i.e., the target synthesis specification) can be initialized. In a first step of the optimization, the characterization parameter values can be derived from the provided functional compounds. However, the derivation of the characterization parameters can also refer to accessing a storage device on which the corresponding characterization parameter values of the corresponding potential target functional compounds have been stored. Furthermore, if the provided digital representation of the potential target functional compound already includes the characterization parameters, this step can also be omitted. Based on the requested additional application conditions, in particular based on the biodegradation habitat, a corresponding determination model, i.e., a biodegradation model, can be provided. Based on the provided determination model and the digital representation of the potential target functional compound, the target application value of the potential target functional compound, i.e., the biodegradability, can be provided. In a next step, it is determined whether the determined performance value, i.e., the determined biodegradation capacity, meets the target value, i.e., the target biodegradation capacity, within a predetermined limit. If this is not the case, i.e., if the condition is not met, the functional compound is modified and, optionally, the previously provided constraints are taken into account to determine a new potential target functional compound. Then, the iteration can be restarted for the new potential target functional compound. If at a certain point, the determined performance value meets the target value within the limit, i.e., if the corresponding condition is met, the potential target functional compound is determined as the target functional compound and is provided to, for example, a user or a control unit for producing the corresponding determined target functional compound.

[0095] Figure 5An additional preferred embodiment of the above-described method for determining a target synthesis specification with a predetermined target biodegradability is schematically and exemplarily shown, wherein in this embodiment, in addition to the target biodegradability, it is desired that the target functional compound also meets additional target values, namely target technical application properties. The additional target technical application properties can refer to any technical application property, for example, also to an additional biodegradation ability in another habitat, or any other technical application property. Generally, the method follows the same principle as described above with respect to Figure 4 as described. However, due to the additional target values, additional conditions must be met during optimization. Therefore, hereinafter, only the main differences from the method described above will be pointed out. In particular, in this preferred embodiment, the optimizer module optimizes not only on the first target value, i.e., on the target biodegradation ability, but also on the second target value. Preferably, also for the second target value, a determination model suitable for determining the value of the technical application property based on the characterization parameters is utilized. Therefore, in addition to the method for the second target application described above, a second determination model is provided, which allows the determination of the application property value based on the characterization parameters of the second target application. The second determination model can, for example, be based on the same algorithm as the biodegradation model and is only trained with different data sets such that it determines another property of the functional compound. Then, the comparison not only refers to determining whether the determined biodegradation ability meets the target biodegradation ability within the limits, but also refers to whether the determined second application property value meets the target second application property value within the limits. Predetermined rules can be used to determine in which cases to continue the iteration, i.e., to provide new functional compounds, and under which conditions to determine a potential target functional compound as the target functional compound. For example, the user can predetermine weights for weighting which conditions must be met. For example, for the user, it may be more important to meet the biodegradation ability, while other target application properties are less important. In this case, the limit within which the second target application property can be satisfied can be set wider, or the satisfaction of this condition can be weighted less strongly. In this context, the Pareto optimization method can also be used to find the best trade-off between different goals. If at a point in the iteration, it is determined that the conditions are met and the predetermined rules are satisfied, then the corresponding potential target functional compound can be determined as the target functional compound and provided to the user as an output, or can be used to generate a control file for the production of the corresponding target functional compound.

[0096] Figure 6A block diagram illustrating an exemplary system architecture of an automated laboratory system 1000 for synthesizing functional compounds, the automated laboratory system having a laboratory equipment control device 1102, a network 1150, and a synthesis specification (i.e., recipe) module 1100 / 1110 and a client device 1108. The automated laboratory system includes a laboratory equipment control device layer 1152 as part of the laboratory equipment control device 1102, and a synthesis specification module layer 1154 associated with the synthesis specification module and a remote control or client layer 1156 associated with the client device 1108. The laboratory equipment control device layer can be divided into several hierarchical layers: a hardware layer, a middleware layer, and an interface layer. The hardware layer relates to the hardware resources specifically for controlling the synthesis of functional compounds, such as sensors and actuators. The middleware relates to any known middleware for laboratory or equipment synthesis operations. An example is LABS / QM, which provides different abstractions for the hardware, network, and operating system, such as low-level device control and messaging. The communication layer relates to communication protocols, and one of the protocols in the protocol can be REST, which can be implemented on different transport protocols (i.e., UDP, TCP, telemetry), allowing messages to be exchanged between the laboratory equipment control device and the laboratory equipment device. Such a software architecture allows the control and monitoring of laboratory equipment without having to interact with the hardware.

[0097] The synthesis specification module layer 1154 can include: a mass storage layer, a computing layer, and an interface layer. The storage layer is configured to provide a mass storage device for a data-driven biodegradation model in order to provide a recipe (i.e., synthesis specification) of a functional compound that meets the target biodegradability, as described in detail above. In particular, as described above, the functions performed by the device can be provided as program code components stored on the mass storage device. In addition, the synthesis specifications of multiple functional compounds can be stored in the mass storage device. Such data can be stored in a structured database, such as an SQL database, or stored in a distributed file system, such as HDFS, a NoSQL database, such as HBase, MongoDB. The computing layer can include an application layer that allows customization of the functions provided by standard cloud services to perform computational processes based on the target properties. Such functions can include determining a digital representation of a target functional compound based on the target biodegradability and the biodegradation model, generating a synthesis specification from the digital representation of the target functional compound, and providing the synthesis specification as control data to the laboratory equipment control device.

[0098] The interface layer can implement web services, a network interface as UDP or TCP, or a Websocket interface. To communicate with the laboratory equipment control device, a REST API is implemented.

[0099] The client layer 1156 provides an interface for end users. For end users, the client layer 1156 may run a client-side network application that provides an interface to the synthetic specification module layer 1154 or the laboratory equipment control device layer 1152. A UI may be provided to the user for selecting a target biodegradability and a biodegradation habitat for the target biodegradability, and the target biodegradability may also include a range of biodegradability values. In other examples, a UI may be provided to the user for selecting more than one target biodegradability and corresponding values. The application may be configured for the user to remotely monitor and control the laboratory equipment control device and operations. In other examples, the client device layer and the synthetic specification module layer may be integrated into one device. The alternatives described herein are for illustrative purposes only and should not be considered limiting.

[0100] Figure 7 A block diagram illustrating an exemplary system architecture of a system and apparatus, network 2150, and model generation module 2100 / 2110 for generating a biodegradation model for determining biodegradability, where the model generation module may be regarded as or include a training model device, a synthetic specification module 1100 / 1110, and a client device 2108. The system for generating a biodegradation model includes a model generation module layer 2154 that is part of the model generation module and a client layer 2156 associated with the client device 2108.

[0101] The model generation module layer 2154 may include: a mass storage layer, a computing layer, and an interface layer. The storage layer is configured to provide a mass storage device for the data-driven biodegradation model as described above. Additionally, the mass storage device is configured to store the synthesis specifications of functional compounds and the measured biodegradability of one or more habitats. Such data may be stored in a structured database, such as an SQL database, or in a distributed file system, such as HDFS, a NoSQL database, such as HBase, MongoDB. The computing layer may include an application layer that allows customization of the functions provided by standard cloud services to perform the computational processes for generating a biodegradation model for determining the biodegradability of functional compounds. Such functions may include receiving the respective digital representations associated with the synthesis specifications of at least two previously measured functional compounds, measurement data of the at least one biodegradability of each of the at least two previously measured functional compounds in at least one habitat; receiving the digital representation of at least one unmeasured functional compound at the model generation module; training the model based on the digital representations of at least two previously measured functional compounds according to the training principles described above, the measurement data of the at least one biodegradability of each of the at least two previously measured functional compounds in at least one habitat, and preferably, the measure of similarity between the digital representation associated with the synthesis specification of each of the at least two previously measured functional compounds and the corresponding digital representation associated with the synthesis specification of at least one unmeasured functional compound, and providing the biodegradation model of biodegradability via an output interface. The model generation module layer may be configured to deploy the generated model and the synthesis specification database to the synthesis specification module layer. This may include storing the generated model and the synthesis specification database in the mass storage device associated with the synthesis specification module.

[0102] The model generation module layer may also be configured to determine the digital representation of a functional compound associated with the synthesis specification based on the synthesis specification and / or the molecular structure. The digital representation may include a set of characterization parameters associated with the synthesis specification and / or the molecular structure of each measured functional compound. One way to derive these characterization parameters may be to apply the SMILES algorithm or any other principle described above. In the case of generating the model based on the digital representation derived from the formulation, the relationship between the synthesis specification and / or the molecular structure and the characterization parameters may be stored in the mass storage device associated with the model generation module. In this case, deploying the model includes providing this relationship.

[0103] The interface layer can implement web services, a network interface as UDP or TCP, or a Websocket interface. To communicate with the client device, a REST API is implemented in this example. The client layer 2156 provides access to a mass storage device that contains the synthesis specifications of functional compounds and the biodegradability of at least one of at least two functional compounds. The client layer also provides an interface for the end user. For the end user, the client layer 2156 can run a client-side web application that provides an interface to the model generation module layer 2154 or the mass storage device associated with the client layer. A UI can be provided to the user for selecting a test method and / or a habitat for which biodegradability should be determined. A UI for selecting synthesis specification data can also be provided to the user. The user interface can also provide an option to upload the selected data to the model generation module layer and, optionally, an option to initiate model generation.

[0104] Figure 8 An exemplary system 700 for producing a chemical product based on synthesis specifications generated according to the present invention is shown. In this example, the system includes a user interface 710 and a processor 720 associated with a control unit 740. The user interface 710 and the processor 720 can be associated with or implemented according to the above principles and, in particular, can be adapted to execute a computer-implemented method to determine a target functional compound and / or synthesis specifications based on the determined biodegradability, as described above. The control unit 740 is configured, for example, to receive control data generated according to the present invention as described above, in particular, to receive control data generated based on synthesis specifications of functional compounds including a target biodegradability. In this example, the control data is provided by a database 730. However, in other examples, the control data can also be provided by a server or any other computing unit for distributing data. Containers 750, 752 each contain components of the chemical product, such as components, catalysts, etc. There are typically more than two containers. However, in this example, only two are shown for illustrative purposes. Valves 760, 762 are associated with containers 750, 752. According to the synthesis specifications, valves 750 and 752 can be controlled to incorporate appropriate amounts of each component into reactor 770. The motor 800 of mixer 780 can also be controlled by the control unit according to the synthesis specifications. An optional heater 790 can also be controlled according to the synthesis specifications. Finally, an outlet valve 810 in fluid communication with the reactor can be controlled by the control unit to provide the chemical product to a container or a test system 820.

[0105] In the following, a more detailed example of a possible biodegradation model is provided. In this example, a machine learning feature model is first used to determine the features of a molecule, which determine the biodegradability of the molecule. For example, the feature model can be a graph neural network (GNN). The GNN interprets the molecule as a graph, where the atoms of the molecule are nodes and the atomic bonds of the molecule are the edges of the graph. Then, the GNN is configured to use this graph to find relevant substructures of the molecule. Then the feature model is trained to transform the molecule into a vector of a predetermined amount of features (e.g., 20 features) such that similar molecules are similar in their representation in these predetermined features. The similarity can be determined based on the Tanimoto similarity based on the Morgan Fingerprint. More details about examples of GNNs that can be utilized in this context can be found in the following articles: "Molecular contrastive learning of representations via graph neural networks.", Wang, Y., Wang, J., Cao, Z. et al., Nat Mach Intet 4, 279-287 (2022) and "Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrast", Wang Y., Magar R., Liang C. and Farimani A.B., J. Chem. Inf. Model. 62, 11, 2713–2725 (2022). Then, the feature vector including the values of the corresponding features of the molecule can be used as the digital representation of the molecule. Then a random forest model is used as the biodegradation model and is trained to determine the biodegradation of the corresponding habitat based on the digital representation.

[0106] Then, training data from two public databases (e.g., the NITE database) can be used to train two models together, and biodegradation measurements of corresponding molecules in corresponding habitats can be performed separately. In this example, more than 3000 data points are provided below. The measurements are performed in the corresponding test habitats using a standardized OECD 301 (A-F) test setup and have a measurement error uncertainty of approximately 10%. The molecules used for training in this exemplary training data are organic compounds with a molecular mass of less than 500 g / mol for more than 90% of the molecules. As is common in machine learning, various parameters of the two models are set such that they perform best on the available data, i.e., biodegradability can be predicted as accurately as possible based on the data at hand. Standard machine learning training protocols are followed to ensure and test that the models can also be generalized to other molecules. Since the two models are trained simultaneously based on the same training data, the biodegradation model can also be considered to include two models and utilize, for example, a molecular graph or a numerical representation of an indicative molecular graph as the input numerical representation.

[0107] To evaluate the performance of the trained models, especially the cascade of the GNN model and the random forest model that map molecules to their biodegradation values, a cross-validation method can be utilized. Thus, in this example, the training data utilized is divided into a predetermined number of data parts, e.g., five parts. From the data parts, multiple random data parts are each used to train the model based on the corresponding data part. Then at least one data part is used to apply the trained model to the molecules in that data part, and the results of the determined biodegradation are compared with the known values of biodegradation. This technique allows simulating the extent to which the model can determine the biodegradability of molecules it has not been trained on. For the exemplary models in this article, considering a 10% measurement uncertainty and a median error of 14%, the average error reaches 19%.

[0108] In addition, as an optional extension to the model, a method for quantifying the uncertainty determined by the model can be implemented based on the degree of similarity of the corresponding molecules to the molecules of the training data utilized. Furthermore, molecules can also be visualized in a two-dimensional graph based on the numerical representation from the GNN model. For example, 20 eigenvalue projections are made into a 2D space for visualization while fairly well preserving the similarity. For the models trained as above, especially for the 838 molecules used as test inputs from the NITE database, the average prediction error determined by cross-validation is 19%, and the median error is 13%. An example is given in the table below:

[0109]

[0110] Typically, possible training data sources, i.e., data sources from which training data for deriving a corresponding biodegradation model can be obtained, are, for example, data from the NITE database, data retrieved from literature papers, and data retrieved from the Aropha dataset. Additionally, corresponding measurements of the biodegradation of training molecules can be performed on test samples on a large scale, for example, using the corresponding standard protocols and methods as described in OECD 301 (A - F).

[0111] Although the above - described biodegradation model is only described as a random forest model, where the numbers are determined by a GNN model, in other embodiments, the numbers can be considered as the structure or graph of a molecule or any representation indicating the structure or graph, where in this case, the biodegradation model includes both a random forest model and a GNN, and thus refers to a two - stage machine - learning model, where the output of the first model is the input of the second model.

[0112] In the following, some additional details regarding some of the above - described embodiments are provided. Typically, in some applications, it is desirable to find functional compounds that meet a certain technical application property (such as tensile strength) and also meet the requirements regarding biodegradability. For this application, a method is proposed, for example, as described regarding Figure 5 In an example of the embodiment applicable to this application, a target requirement for biodegradability can be provided, and a target application property can be further provided. Based on the target application property, a determination model is selected, where the determination model preferably relates the characterization parameters associated with the synthesis specifications to the application property. Additionally, another model is selected based on the habitat of the functional compound. The biodegradation model preferably relates the habitat information and characterization parameters associated with the synthesis specifications to biodegradability. Based on the target application requirements, the characterization parameters based on the synthesis specifications are determined. In an optional step, based on the selected biodegradation model used, additional descriptor values for the habitat are requested. Based on the biodegradation model, the biodegradation ability can be determined. Then, the determined biodegradation ability is compared with the target biodegradation ability. Additionally, the determination model is used to determine the application property of the functional compound, and the determined application property is compared with the target property. If the determined biodegradability meets the target biodegradability and the determined application property meets the target property, the synthesis specifications of the functional compound are provided. The synthesis specifications can also refer to or include control data for controlling the equipment for producing the functional compound. In the case where the target biodegradation ability and / or the target application property are not met, the target application property can be reduced, and the process can be re - run with the reduced target application requirements until the biodegradation requirements are met. An acceptable range for the target application property can be provided. If no functional compound that meets the required targets of biodegradability and target application performance is found, the process can be stopped, and the user can be notified.

[0113] Potential indicators of biodegradability can be one or more of the following: mineralization degree, which refers to whether a functional compound is completely mineralized or the time to reach mineralization; biotransformation, which refers to the change in chemical structure that results in the loss of specific properties (such as toxicity) of a functional compound, or the time to reach this point; half-life, which refers to the time when 50% of the functional compound decomposes. The main habitats are the ocean, wastewater, and soil. For the ocean, the following parameters can affect biodegradation: salt concentration, sediment, water temperature, bacterial culture, etc. In some examples, ocean habitat descriptors can be stored in a database together with the geographical location. In this case, the geographical location can be input, and the parameter values related to this geographical location can be retrieved from the database. For wastewater, the following parameters can affect biodegradation: temperature, bacterial population, bacterial type, enzyme concentration, enzyme. For soil, the following parameters can affect biodegradation: temperature, bacterial population, bacterial type, enzyme concentration, enzyme.

[0114] By studying the drawings, the disclosure, and the appended claims, those skilled in the art can understand and implement other variations of the disclosed embodiments when practicing the claimed invention.

[0115] For the processes and methods disclosed herein, the operations performed in the processes and methods can be implemented in a different order. In addition, the outlined operations are provided only as examples, and some of these operations can be optional, combined into fewer steps and operations, supplemented with further operations, or extended to additional operations without departing from the essence of the disclosed embodiments.

[0116] In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality.

[0117] A single unit or device can perform the functions of several items recited in the claims. The fact that certain measures are recited only in mutually different dependent claims does not mean that a combination of these measures cannot be used advantageously.

[0118] Procedures such as providing a digital representation and a biodegradation model, determining biodegradability, providing biodegradability, etc., which are performed by one or several units or devices, can be performed by any other number of units or devices. These procedures can be implemented as program code components of a computer program and / or dedicated hardware.

[0119] A computer program product can be stored / distributed on a suitable medium (such as an optical storage medium or a solid-state medium), supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.

[0120] Any unit described herein can be a processing unit as part of a classical computing system. The processing unit can include a general-purpose processor and can also include a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or any other specialized circuit. Any memory can be a physical system memory, which can be volatile, non-volatile, or some combination of both. The term "memory" can include any computer-readable storage medium, such as a non-volatile mass storage device. If the computing system is distributed, then the processing and / or memory capabilities can also be distributed. The computing system can include multiple structures as "executable components". The term "executable component" is a term for a structure in the computing field that is well understood to be a structure that can be software, hardware, or a combination thereof. For example, when implemented in software, those of ordinary skill in the art will understand that the structure of an executable component can include software objects, routines, methods, etc. that can be executed on a computing system. This can include executable components in the heap of a computing system or on a computer-readable storage medium. The structure of an executable component can exist on a computer-readable medium such that when interpreted by one or more processors (e.g., by a processor thread) of a computing system, the computing system is caused to perform a function. Such a structure can be directly computer-readable by a processor, for example, as in the case where the executable component is binary, or it can be structured such that it is interpretable and / or compilable, for example, either in a single stage or in multiple stages, to produce such binary that can be directly interpreted by a processor. In other instances, the structure can be hard-coded or hard-wired logic gates that are implemented exclusively or near-exclusively in hardware, such as within a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), or any other specialized circuit. Thus, the term "executable component" is a term for a structure well understood by those of ordinary skill in the computing field, whether implemented in software, hardware, or a combination. Any implementation described herein is described with reference to actions performed by one or more processing units of a computing system. If such actions are implemented in software, one or more processors, in response to having executed computer-executable instructions that make up an executable component, direct the operation of the computing system. The computing system can also include a communication channel that allows the computing system to communicate with other computing systems, for example, via a network. A "network" is defined as one or more data links that enable the transfer of electronic data between computing systems and / or modules and / or other electronic devices. When information is passed or provided to a computing system via a network or another communication connection (e.g., hardwired, wireless, or a combination of hardwired or wireless), the computing system appropriately treats that connection as a transmission medium. The transmission medium can include a network and / or a data link that can be used to carry the desired program code components in the form of computer-executable instructions or data structures and that can be accessed by a general-purpose or specialized computing system or combination.While not all computing systems require a user interface, in some embodiments, a computing system includes a user interface system for interfacing with a user. The user interface serves as an input or output mechanism to the user, for example, via a display.

[0121] Those skilled in the art will appreciate that at least portions of the present invention may be practiced in a network computing environment having many types of computing system configurations, including personal computers, desktop computers, laptop computers, messaging processors, handheld devices, multiprocessor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, mobile phones, PDAs, pagers, routers, switches, data centers, wearable devices (such as glasses), and the like. The present invention may also be practiced in a distributed system environment where local and remote computing systems that are linked, for example, by a hardwired data link, a wireless data link, or a combination of hardwired and wireless data links, both execute tasks over a network. In a distributed system environment, program modules may be located in both local and remote memory storage devices.

[0122] Those skilled in the art will also appreciate that at least a portion of the present invention may be practiced in a cloud computing environment. A cloud computing environment may be distributed, although this is not required. When distributed, a cloud computing environment may be distributed internationally within an organization and / or have components owned by multiple organizations. In this specification and the following claims, "cloud computing" is defined as a model for enabling on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage devices, applications, and services). The definition of "cloud computing" is not limited to any of the many other advantages that may be obtained from such a model at the time of deployment. The computing systems of the figures include various components or functional blocks that may implement the various embodiments disclosed herein as explained. The various components or functional blocks may be implemented on a local computing system or on a distributed computing system that includes elements residing in the cloud or implementing aspects of cloud computing. The various components or functional blocks may be implemented as software, hardware, or a combination of software and hardware. The computing systems shown in the figures may include more or fewer components than those shown in the figures, and some of the components may be combined when the environment permits.

[0123] Any reference numerals in the claims should not be construed as limiting the scope.

[0124] The present invention relates to a method for determining a synthetic specification including a target biodegradability. A target biodegradability indicating the biodegradation characteristics of a functional compound is provided. A digital representation of a potential target synthetic specification is provided, which potential target synthetic specification indicates the physicochemical characteristics of the functional compound. A habitat is provided, which habitat indicates a habitat descriptor value of a habitat descriptor. A model based on the habitat is provided, which model is adapted to determine the biodegradability of the functional compound in the habitat. The biodegradability of the potential target functional compound is determined based on the provided model and the digital representation. The determined biodegradability is then compared with the target biodegradability, and i) the potential target functional compound is determined as the target functional compound, or ii) a new potential target synthetic specification of the potential target functional compound is provided, and the determination of the biodegradability is repeated using the new potential target synthetic specification.

Claims

1. A computer-implemented method for determining a target functional compound having a target biodegradability, wherein the method (200) comprises: Providing (210) the target biodegradability, wherein the biodegradability indicates the biodegradation characteristics of a potential target functional compound, Providing (220) a digital representation of the potential target functional compound, Providing (230) a biodegradation habitat, wherein the biodegradation habitat indicates habitat descriptor values of habitat descriptors that affect the biodegradation of the functional compound in the corresponding habitat, and wherein the habitat descriptors indicate environmental characteristics of the habitat, Providing (240) a biodegradation model based on the provided biodegradation habitat, wherein the biodegradation model is adapted to determine the biodegradability of the functional compound in the corresponding biodegradation habitat, and wherein the biodegradation model is parameterized with respect to the biodegradation habitat such that it determines the biodegradability of the functional compound based on the digital representation of the functional compound, and the biodegradation model is a data-driven model, Determining (250) the biodegradability of the potential target functional compound based on the provided biodegradation model and the digital representation, and Comparing (260) the determined biodegradability of the potential target functional compound with the target biodegradability, and based on the comparison, i) determining the potential target functional compound as the target functional compound, or ii) providing a new potential target functional compound and repeating the determination of the biodegradability using the new potential target functional compound.

2. The method according to claim 1, wherein the functional compound has a molecular mass of less than 10,000 g / mol.

3. The method according to any one of the preceding claims, wherein the functional compound has at least one of the following properties: having an impact on living organisms, being adapted to affect the structure of living organisms or being adapted to affect the function of living organisms.

4. The method according to any one of the preceding claims, wherein the functional compound comprises at least one of the following functional groups: a carboxylic acid group, a carboxylic acid derivative group, an ether group, a hydroxyl group, a carbonyl group, an amine group, an ammonium group, an alkyl group, an alkylene group, a phenyl group, an acetal group, a ketal group, a thiol group, a sulfide group, a phosphate group, a halide group, or a combination thereof.

5. The method according to any one of the preceding claims, wherein the digital representation comprises or indicates the chemical structure of the functional compound, and wherein the chemical structure is associated with and / or indicates the ratio of one or more structural formulas and the number of more than one structural formula.

6. The method according to claim 5, wherein the at least two structural formulas correspond to structural formulas related via chemical equilibrium and / or stereochemistry.

7. The method according to any one of the preceding claims, wherein the method further comprises providing a biodegradation test method, wherein the provided biodegradation test method indicates a standardized biodegradation test method for experimentally determining the biodegradation of a chemical, and wherein the biodegradation model is further provided based on the provided biodegradation test method.

8. The method according to any one of the preceding claims, wherein the habitat descriptor values of the habitat descriptors are stored in association with corresponding geographical locations, wherein providing a biodegradation habitat means providing the geographical location of the habitat and retrieving the habitat descriptor values of the geographical location from a storage device.

9. An interface method for providing an interface, wherein the interface method comprises: Receiving, via a user interface, a target biodegradability, a starting digital representation, and a habitat as inputs, and providing the received target biodegradability, digital representation, and the habitat to a processor that executes the method (200) according to any one of claims 1 to 8, and Providing the starting digital representation of the functional compound as a result, wherein the result is received from the processor that executes the method (200) according to any one of claims 1 to 8.

10. A computer-implemented training method for training a data-driven biodegradation model for parameterizing the biodegradation model, wherein the training method (300) comprises: Providing (310) training data associated with a predetermined biodegradation habitat, wherein the training data comprises a) digital representations of a plurality of training functional compounds, and b) the biodegradability of the corresponding biodegradation habitat associated with each training functional compound, Providing (320) a data-driven trainable biodegradation model, Training (330) the provided data-driven biodegradation model based on the provided training data such that the trained biodegradation model is adapted to determine the biodegradation of a functional compound based on the digital representation of the functional compound, and Providing (340) the trained biodegradation model.

11. A device for determining a target functional compound having a target biodegradability, wherein the device (110) comprises: A target biodegradability providing unit (111) for providing a target biodegradability, wherein the biodegradability indicates the biodegradation characteristics of a potential target functional compound, A digital representation providing unit (112) for providing a digital representation of a potential target functional compound, A habitat providing unit (113) for providing a biodegradation habitat, wherein the biodegradation habitat indicates the habitat descriptor values of habitat descriptors that affect the biodegradation of a functional compound in the corresponding habitat, and wherein the habitat descriptor indicates the environmental characteristics of the habitat. A model providing unit (114) for providing a biodegradation model based on the provided biodegradation habitats, wherein the biodegradation model is adapted to determine the biodegradability of a functional compound in the respective biodegradation habitat, and wherein the biodegradation model is parameterized with respect to the biodegradation habitat such that it is a data-driven model for determining the biodegradability of the functional compound based on the digital representation of the functional compound. A biodegradability determining unit (115) for determining the biodegradability of the potential target functional compound based on the selected biodegradation model and the digital representation, and An iterative control unit (116) for comparing the determined biodegradability of the potential target functional compound with the target biodegradability, and based on the comparison, i) determining the potential target functional compound as the target functional compound, or ii) providing a new potential target functional compound and repeating the determination of the biodegradability using the new potential target functional compound.

12. An interface device for providing an interface, wherein the interface device comprises: An input interface unit for receiving a target biodegradability, a starting digital representation, and a habitat as inputs via a user interface, and for providing the received target biodegradability, starting digital representation, and the habitat to the device according to claim 11, and A result interface for providing the habitat descriptor value of the functional compound as a result, wherein the result is received from the device according to claim 11.

13. A training device for training a data-driven biodegradation model for parameterizing the biodegradation model, wherein the training device (120) comprises: A training data providing unit (121) for providing training data associated with a predetermined biodegradation habitat, wherein the training data comprises a) digital representations of a plurality of training functional compounds, and b) the biodegradability of the respective biodegradation habitat associated with each training functional compound, A trainable model providing unit (122) for providing a data-driven trainable biodegradation model, A training unit (123) for training the provided data-driven biodegradation model based on the provided training data such that the trained biodegradation model is adapted to determine the biodegradation of a functional compound based on the digital representation of the functional compound, and A trained model providing unit (124) for providing the trained biodegradation model.

14. A computer program product for determining a target functional compound having a target biodegradability, wherein the computer program product comprises program code means for causing the apparatus according to claim 11 to perform the method according to any one of claims 1 to 8.

Images