Platform for microbially producing colorants

Abstract

A method, system, computer program, and computer platform for AI-guided design, production, and application of biosynthesized colorants. The invention enables a computer platform system to receive user-defined input data such as a target colorant, functional attributes, and environmental impact constraints, and to generate a customized recipe for biosynthesized production and / or use of the colorant molecules. The recipe may include selection of microbial strains, pigment molecules or precursors, fermentation parameters, extraction and post- processing techniques, and application instructions. The system can monitor process parameters during deployment, either in a simulation or physically, determine real-time changes, and generate updated recipes accordingly. The invention supports precursor-based workflows (e.g., Tryptone conversion to violacein), chemical or enzymatic post-processing, and predictive modeling of color, biosynthetic feasibility, and compliance. This disclosed system provides a scalable, sustainable, and programmable production of colorants across textile, cosmetic, plastic, food, and biomedical industries.

Description

PLATFORM FOR MICROBIALLY PRODUCING COLORANTSTECHNICAL FIELD

[0001] The present disclosure generally relates to processes for biosynthetic production of microbial colorants, and more particularly related to a computer platform for controlling digital or physical production and use of microbial colorants.BACKGROUND OF THE INVENTION

[0002] In recent times, the production of dyes from microorganisms has gained prominence as a sustainable alternative to traditional chemical processes. This method of microbial dye production is celebrated for its reduced environmental footprint, leveraging the natural processes of microorganisms to create colorants with less waste and pollution. The process not only aligns with the growing demand for eco-friendly products but also offers innovative solutions to longstanding environmental issues associated with the dye and pigment industry.

[0003] However, the microbial production of dyes is not without its challenges. There are numerous microorganisms that produce pigment molecules and each may have numerous variants with unique characteristics and pathways to produce the pigment molecules. Moreover, taking pigment-generating microorganisms through laboratory-scale or production-scale processes requires selecting and tuning many biological, chemical and process variables and parameters for optimized colorant production.

[0004] Getting to a desired microbially produced colorant is challenging and requires careful selection of many parameters such as the right microorganisms, molecular pathways, chemical parameters, and process parameters, to name a few. This is a very time consuming and costly selection and optimization process.

[0005] Al-assisted computer programs that use curated databases, biosynthetic modeling, and real-time simulations can be leveraged to accelerate optimization processes. Such computer programs can assist users in generating custom solutions, and predicting properties for molecules, and are currently used in various processes, such as new drug discovery.

[0006] There is a need in the art of microbial colorant production for improved systems and methods which allow for leveraging automatic or Al-assisted optimization for identifying and optimizing production of colorants, as well as identification of novel pigments or pigment precursors.SUMMARY

[0007] In accordance with one disclosed aspect, provided herein is a method for discovery and optimization of production and / or use of microbial colorants from microorganisms by a computer platform system, the method including receiving, through a user interface of an input data module of the computer platform system, user input data, from a user, related to a desiredcolorant; generating, by at least one recipe generating engine of the computer platform, a recipe for production and / or use of the desired colorant using microorganisms, based on the user input data; communicating, by an output data module of the computer platform, the recipe to a deployment system for implementing the recipe; and monitoring, by at least one processor of the computer platform system, a plurality of parameters of the deployment system during the implementation of the recipe.

[0008] The method may further include determining, by the processor, at least one change in the recipe, generating, by the processor, a modified recipe, and communicating, by the processor, the modified recipe to the deployment system for implementation.

[0009] The at least one change may be notified to the user, by the processor, and the modified recipe is generated by the processor once the user approves the change.

[0010] The recipe may be generated by the processor by selecting, at the processor, one or more microorganisms from a plurality of microorganisms, the one or more microorganisms capable of producing pigment molecules or precursor molecules that are an ingredient of the desired colorant; selecting, by the processor, a set of biomanufacturing processes from a plurality of sets of biomanufacturing processes, the set of biomanufacturing processes to yield a desired amount of the desired colorant; selecting, by the processor, a set of process parameter values related to the set of the biomanufacturing processes for causing the set of biomanufacturing processes to yield a desired amount of the desired colorant; and generating the recipe, wherein the recipe is the combination of the selected one or more microorganisms, set of biomanufacturing processes, and set of process parameter values.

[0011] The method may further include selecting the pigment molecules from a plurality of pigment molecules, and wherein the recipe further includes instructions on engineering the pigment molecules.

[0012] The selected microorganisms may produce pigment precursor molecules, and the pigment precursors may be polymerized or chemically modified by one or more chemical agents external to the microorganism to form the final colorant. The pigment precursor may include tryptophan, glutamine, phenazines, tyrosine, or a catechol compound.

[0013] The method may further include selecting an application technique from a plurality of application techniques to yield a desired use of the desired colorant, and wherein the recipe further includes the selected application technique.

[0014] The selected one or more microorganisms may include one or more microorganisms that may need to be genetically modified or engineered.

[0015] The recipe may be transmitted by the processor for display to the user.

[0016] The deployment system may be a digital biomanufacturing process simulation system, a physical biomanufacturing process facility, or a combination thereof.

[0017] The deployment system may include a local controller, wherein communicating, by the processor, the recipe to the deployment system includes transmitting the recipe to the localcontroller.

[0018] Physical process data may be collected by one or more sensor of the physical biomanufacturing process facility and the physical process data may be transmitted to the local controller or the processor.

[0019] The user input data may include one or more of: desired color value, a desired property value for the desired colorant, a desired application value for the desired colorant, and a desired impact value for the desired colorant.

[0020] In accordance with one disclosed aspect, provided herein is a computer program to control production and / or use of biosynthesized colorants from microorganisms, the computer program once executed by a processor causes a processor to receive user input data from a user related to a desired colorant, generate a recipe for production and / or use of the desired colorant using microorganisms, communicate the recipe to a deployment system for implementing the recipe, and monitor a plurality of parameters of the deployment system during the implementation of the recipe.

[0021] The processor may be further configured to determine at least one change in the recipe, generate a modified recipe, and communicate the modified recipe to the deployment system for implementation.

[0022] In accordance with one disclosed aspect, provided herein is a system to manage biosynthetic production and / or use of microbial colorants from microorganisms, the system comprising a computer platform including a processor and a memory storing instructions executable by the processor to cause the computer platform to receive user input data from a user related to a desired colorant, generating a recipe for production and / or use of the desired colorant using microorganisms, communicate the recipe to a deployment system for implementing the recipe, and monitor a plurality of parameters of the deployment system during the implementation of the recipe.

[0023] The computer platform may be further configured to determine at least one change in the recipe, generate a modified recipe, and communicate the modified recipe to the deployment system for implementation.

[0024] In accordance with one disclosed aspect, provided herein is a computer platform system for discovery and optimization of production and / or use of microbial colorants from microorganisms, the computer platform system including at least one memory storing processor executable instructions, and at least one processor, configured to execute the processor executable instructions for an input data module, wherein the input data module receives user input data related to a desired colorant through a user interface, a recipe generating engine, wherein the recipe generating engine receives the user input data, queries at least one data system for related data to the user input data, and generates a recipe for production and / or use of the desired colorant using microorganisms, based on the user input data and the related data, an output data module, wherein the output data modules outputs the recipe including a pluralityof parameters for implementation of the recipe, and a deployment system, wherein the deployment system receives the recipe from the output data module and implements the recipe.

[0025] Other aspects and features will become apparent to those ordinarily skilled in the art upon review of the following description of specific disclosed embodiments in conjunction with the accompanying figures.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In the following, embodiments of the present disclosure will be described with reference to the appended drawings. However, various embodiments of the present disclosure are not limited to the arrangements shown in the drawings.

[0027] Figure 1 is a schematic diagram pertaining to a computer platform for controlling production and use of microbial colorants, according to an embodiment;

[0028] Figure 2 is a block diagram of a processor circuit for implementing the computer platform of Figure 1 , according to another embodiment;

[0029] Figure 3 is a graphical user interface depicting an implementation of the user interface of Figure 2, according to one embodiment;

[0030] Figures 4A to 4C are flowcharts depicting blocks of code for directing the processor circuit of Figure 2 to control deployment of microbial production of Figure 1 ;

[0031] Figures 5A to 5C are schematic process flow diagrams showing various elements pertaining to the colorant production process of Figure 1 .DETAILED DESCRIPTION

[0032] Various processes, systems, compositions, and methods will be described below to provide an example of each claimed aspect. No embodiment described below limits any claimed aspect and any claimed aspect may cover processes, systems, compositions, or methods that differ from those described below. The claimed aspects are not limited to processes, systems, compositions, or methods having all of the features of any one process, system, composition, or method described below or to features common to multiple or all of the processes, systems, compositions, or methods described below.

[0033] Unless otherwise specified, platform referred to herein refers to a system of software or hardware components, or a combination of both.

[0034] The following abbreviations are used throughout this disclosure:Abbreviation MeaningSMB Spent Microbial BiomassGHG Greenhouse GasAl Artificial IntelligenceANN Artificial Neural Network

[0035] According to various aspects, the disclosed process pertains to a platform forgenerating a recipe for production and use of microbial colorants. Key benefits of the disclosed process include improving scalability and productivity of microbial dye production by providing an artificial expert knowledge base and streamlining the steps pertaining to biomanufacturing process design and colorant application. The disclosed platform provides detailed instructions regarding biomanufacturing processes of microbial colorants and thus facilitate a user to obtain instructions to custom-make different variations of microbial colorants according to the customized needs and requirements of the user. Further, the disclosed platform facilitates saving time and resources in finding the right microorganisms, used chemicals, process steps (from strain selection and design all the way to the Dye Application stage), and related process parameters.

[0036] Referring to Figure 1 , a computer platform system for production and use of microbial colorants is shown generally at 100. The platform system 100 includes a recipe generating engine 110 that is configured to receive desired parameters of a target microbial colorant from an input data module 230, and outputs instructions pertaining to biomanufacturing of the target microbial colorant to an output data module 240. The recipe generating engine 110 can be trained on a curated and dynamically expandable database of microbial pigments, biosynthetic pathways, bioreactor parameters, and downstream application contexts. The input data module 230 can help guide a user to input requirements, specifications, and constraints related to a microbial colorant.

[0037] The computer platform system 100 includes at least one processor and at least one memory storing processor executable instructions wherein the at least one processor is configured to execute the instructions to perform the functions of the recipe generating engine 110, input data module 230, and output data module 240, and to access information for data systems 207.

[0038] The instructions for the computer platform system 100 may be stored locally, stored at other computing devices (e.g., servers), or stored in a cloud system.

[0039] The information in the data system 207 may be stored locally, stored at other computer device (e.g., servers) and / or stored in a cloud system.

[0040] The output data module 240 stores and presents the recipe. A general recipe for production and use of a colorant from a microorganism can include elements of strain design (or selection), variant molecule design, upstreaming, downstreaming (e.g. pigment extraction), and colorant application as is shown generally in Figurel , according to an embodiment. Below, more details about some examples and embodiments of these four steps are provided. Exemplary process flows of various microbial colorant production processes are depicted in Figures 5A to 5C.

[0041] Referring to Figure 5A, a schematic process flow is generally shown at 500, illustrating various steps in a bioproduction facility for production of a microbial colorant. The illustrated process flow includes adding a seed culture 510 and a fermentation medium 512 to a fermenteror reactor 514. The resulting fermentation broth is processed through a pigment extraction process 520 to produce extracted pigment 530 and spent microbial biomass 540. The extracted pigment 530 may then undergo further processing with additives and conditioning 560 to yield the final colorant 590. The platform system 100 further includes a data system 207 in communication with the recipe generating engine 110 to enable the recipe generating engine 110 to look up or retrieve from thousands, millions, billions, or even more of data types and datapoints related to the variables defining production and use of colorants from microbial sources, to select values for process variables that can generate a colorant according to the requirements defined by a user through the input data module 230.

[0042] The upstreaming step may include preparing a microbial seed culture (e.g., seed culture 510) generally including one or more microbial agents, such as microbes, bacteria, fungi, or yeast that include a gene pathway responsible for creating a pigment molecule or a precursor molecule for a pigment molecule. The microbial agents may include bacterial strains such as bacteria from one or more of the genera Janthinobacterium, Chromobacter, Duganella, Collimonas, Massilia, Pseudoalteromonas, Escherichia, Citrobacter, Corynebacterium, and Streptomycese. Additionally, the upstreaming step may include engineered microbial strains, which can be optimized through, for example, codon usage adjustment, pathway balancing, and secretion system tuning to improve pigment yield and simplify downstream extraction.

[0043] Various microbial agents along with their corresponding produced pigment molecules are studied and may be known or readily available to a person skilled in the art from publicly or privately available sources such as scientific publications. For example, bacteria from the genera Janthinobacterium and Chromobacter may produce violacein (C20H13N3O3) presenting a purple color. In a further example, bacteria from the genera Pseudoalteromonas and Streptomycese may produce prodigiosin (C20H25N3O) presenting a red color. Other examples of pigment molecules produced by microorganisms include melanin, flexirubin, cartenoids, indigoidine, and riboflavin.

[0044] Additionally, microbial agents that can produce precursor molecules to pigments can be known as well. For example, microorganisms such as Streptomyces, Halomonas venusta, or engineered E. coli can be used to produce, for example, tryptophan or tyrosine, known precursors in biosynthesis pathways of melanin. Tryptophan, as a precursor molecule, can undergo oxidation, non-enzymatic, and enzymatic transformations to produce violacein (generally bluepurple), indigogine (generally blue), and / or deoxyviolacein (generally purple). By controlling the formation of colorants from tryptophan (e.g, controlling temperature, time, pH, added enzymes) and integration of additives such as amino acids into the created molecule, a wide range of colorants can be produced, including unconventional green, blue, and purple. Santarcangelo et. al., “Generation and structure elucidation of a red colorant formed by oxidative coupling of chlorogenic acid and tryptophan” Food Chemistry. 2023, showed combining tryptophan with chlorogenic acid can yield a red colorant for use on textile substrates, food products and moreunder eco-friendly conditions.

[0045] A precursor molecule, such as tryptophan, may be used as a primary molecule extracted from microorganisms that can then undergo further treatments and produce various range of colorants.

[0046] Referring to Figure 5B, a process flow for producing target colorants using a precursor molecule (e.g., tryptophan) is generally shown at 500b. The precursor molecule is extracted and may be purified (e.g., using centrifugation or filtration) at step 516, and may or may not be dried. The extracted precursor molecule 518 undergoes a cell-free synthesis (e.g..condensation) process which may or may not include additional additives (e.g., small molecules, amino acids, etc.) to produce the pigment molecule 530 (e.g., violacein). The pigment molecule can be conditioned at step 560 using various additives (e.g. linker molecules, surfactants) to yield the final colorant 590.

[0047] In other embodiments, an extracted precursor molecule may undergo other processes, such as polymerization to produce the pigment molecule 530. In some embodiments, the extracted precursor molecule 518 can be stored for later use. For example, the precursor may be combined with one or more stabilizing agents to facilitate storage under ambient or controlled conditions. At a subsequent time, a catalytic agent, such as an oxidant, may be introduced to the stored precursor molecule 518 to trigger its conversion into the pigment molecule 530. This conversion may occur via a cell-free synthesis or chemical transformation process, such as oxidative condensation or polymerization. The transformation can take place in proximity to, or directly within, a substrate application environment (e.g., a dye bath containing a textile material). Such an approach may be advantageous in scenarios where in situ pigment formation enhances substrate binding, improves color durability, or enables controlled timing of pigmentation in downstream applications.

[0048] In other embodiments, the seed culture 510 may be prepared from other microorganisms such as yeast and fungi, such as from the genera Yarrowia, Saccharomyces and Pichia, capable of producing or metabolizing pigment molecules.

[0049] The microbial agents may include natural or modified (i.e., engineered) microorganisms. Natural microorganisms are naturally occurring and may be found or extracted from nature. Engineered microorganisms are created synthetically, for example, through genetic engineering. According to one example, a particular yeast such as Pichia Pastoris may be engineered to include a gene pathway to produce prodigiosin. Preferably the microorganism is engineered to result in production of pigment molecules in high quality and high yield. Quality metrics may include chemical purity, color intensity, spectral absorbance, or stability across application pH or temperature ranges. Yield optimization may include simulations of metabolic flux, cofactor balance, and oxygen transfer efficiency, among other parameters.

[0050] The microbial agents may also have been engineered to produce new pigment molecules that are a variant of naturally occurring pigment molecules. For example, newmolecular variations of prodigiosin may be designed with improved UV tolerance. The new molecular variations may then be engineered into modified microorganisms with pathways to produce the new pigment molecule. These variants may be proposed by an embedded generative model (e.g., GAN or transformer-based), trained on pigment structure and color function data, and screened for biosynthetic feasibility.

[0051] Preparing the microbial seed culture at the upstreaming step may include taking small amounts of microbial agents from a head sample and growing the small amounts in a suitable culture medium including complex organic and inorganic sources, for example, that provide optimal multiplication and reproduction to create healthy microbial agents. The system recipe can include recommendations for appropriate culture media and growth conditions specific to the microbial agents, such as carbon or nitrogen ratios, trace metals, and pH buffering systems.

[0052] The recipe may also include instructions on a downstreaming step. The downstreaming step may include transferring the microbial seed culture, prepared in the upstreaming step, to a fermenter or a bioreactor in a nutrient-rich fermentation medium. In an embodiment, the recipe provides instructions for preparing a fermentation medium that includes one or more carbon sources (e.g., glucose, glycerol, starch), nitrogen sources (e.g., yeast extract, peptone, ammonium salts), trace minerals, buffers, optional additives to modulate redox conditions or secretion profiles, and different macro- and micro-nutrients such as salts and amino acids. The nutrient may be sourced from industrial by-products, such as glycerol, or waste resources, such as agricultural waste obtained as waste source from farms or produce refineries. In an embodiment, beet pulp is used as a fermentation nutrient including both nitrogen and carbon for the Pichia Pastoris yeasts to metabolize during the fermentation.

[0053] The seed culture and the fermentation medium are mixed in the fermenter. The fermentation may be performed in a bioreactor under aerobic or anaerobic conditions, in batch, fed-batch, or continuous mode.

[0054] The internal environment of the fermenter may be configured during the fermentation to yield optimal and efficient pigmentation. The recipe may specify the fermentation mode based on productivity, cost, or pigment stability considerations. In an embodiment, further fermentation conditions such as temperature, pH, dissolved oxygen levels, aeration, agitation, and fermentation duration are considered, controlled and monitored by the platform system 100 for optimal production yield. The platform system 100 may simulate oxygen transfer rates (kl_a), shear sensitivity, and foaming behavior to optimize the fermenter’s configuration.

[0055] The selection of the fermentation medium and the fermentation conditions may largely depend on the microbial strain used in the seed culture. For example, some microbial strains are acid-fermenting while others are base-fermenting, and some are anaerobic while some are aerobic. A person skilled in the art can appreciate that some of the production processes, methods, and compositions in reaching a desired colorant are largely dependent on specific microbial agents used for bioproduction of colorants, and thus the platform disclosed herein canhave a pivotal advantage in facilitating generating wide range of recipes according to a wide selection of desired colorant requirements.

[0056] Generally, the harvested materials at the end of fermentation include a fermented broth that is a liquid, rich in microbial biomass and produced intracellular or extracellular pigment or precursor molecules.

[0057] The extraction step may include processing the liquid fermented broth to extract and purify intracellular or extracellular pigment molecules produced from the fermented broth. The colorant extraction and purification, for example, includes filtering and separating pigment molecules from residual biomass and unutilized nutrient medium in the fermented broth. Typically, the colorant extraction and purification may depend on the intracellularity or extracellularity of the pigment molecules. Solvents or surfactants may be used to help extract pigment molecules from cells and improve the properties of the final colorant.

[0058] Non-aqueous colorant extraction methods may be preferred to aqueous methods so that minimal or no water is wasted during the extraction step. Accordingly, further wastewater treatment may not be necessary, for example, to comply with ever tightening environmental regulations (e.g., this simplifies compliance with local discharge regulations and may reduce total water footprint of the production process). Accordingly, non-aqueous colorant extraction methods may result in simple and economic downstreaming processes that can immensely reduce the overall cost of the colorant production.

[0059] In an example, the colorant extraction and purification includes a solvent-based extraction method in which a solvent, particularly an organic solvent, is added to the fermented broth to create a solution mixture for convenient and efficient separation and isolation of pigment molecules from undesired products of the fermented broth. The solution mixture may be further processed. The solution mixture may undergo mechanical agitation, sonication, or enzymatic digestion to facilitate cell disruption and dissolution of pigment molecules in the solution mixture while leaving insoluble residual biomass in a precipitated layer. The precipitated residual biomass may be separated after a gravity separation method, for example.

[0060] In some embodiments, surfactant-based extraction methods may be used instead of or in conjunction with solvent-based extraction methods. The surfactant-based methods may have similar steps to the solvent-based extraction methods.

[0061] In other embodiments, one or more other cell disruption, extraction, and recovery methods such as bead milling, high pressure homogenization, freeze-thaw, ultrasonication, chemical-based extraction, centrifugation, and filtration may be used alone, with the solventbased extraction, or surfactant-based extraction methods. The optimal extraction method may be determined based on intracellular or extracellular pigment production, among other factors.

[0062] In another embodiment, the fermented broth, or a combination of the fermented broth and the solvent, is passed through one or more filter systems, such as microfiltration with 0.1 to 1 pm sized filter and nanofiltration with 1 to 10 nm sized filter, trapping particles larger than thepigment molecule size. The filtered liquid or slurry may be a mix of solvent, nutrient medium, biomass, and pigment which may be purified through silica gel column chromatography, for example, to obtain a crude pigment paste. The generated recipe may recommend sequential filtration and chromatography of the filtered liquid or slurry, based on estimated molecular size, distribution, and viscosity.

[0063] The extraction step may further include dehydrating the crude pigment (or colorant) solution or medium, for example using lyophilization techniques (for example using a VirTis industrial lyophilizer), spray drying, drum drying, or oven drying to achieve a completely dry and crude pigment in powder form. Alternatively, the crude pigment may remain in a semi-solid or liquid paste form and may be used in liquid or paste format. The resulting colorant may be further processed according to a colorant mixing formula to obtain a final colorant which then could be deposited, embedded, or applied in a variety of applications such as use in textile dyeing, plastic coloring, food and beverage production, pharmaceutical production, and cosmetic production. The colorant mixing formula may involve adding certain additives to the dehydrated pigment powder or pigment past, for example, to increase pigment bonding to a target substrate for longer- lasting and more intense colorant application, or improve other functional properties of the final colorant such as UV tolerance, heat resistance, anti-bacterial and antioxidant properties, for example. The additive(s) may be obtained using proprietary and naturally driven formulations. In some embodiments, such as the embodiment of Figure 5C, other pigments or colorants are added to the extracted pigment to create a final colorant with a desired hue and / or darkness level. The recipe generating engine 110 may compute additive-to-pigment ratios and sequence of addition to ensure color blending stability, for example.

[0064] The output data module 240 may present the recipe as instructions of the designed colorant production and use process. The output data module 240 may also present or communicate the generated or designed recipe to a deployment module 250. The deployment module 250 is a module that can implement the designed colorant production and use process in part or in its entirety by simulation and / or physical execution (e.g. in a lab-scale, pilot-scale, or commercial-scale production facility). The deployment (either simulation or physical) can be controlled by a local controller 252.

[0065] The deployment module 250 (when not a physical execution deployment module) and local controller 252 may be executed by the same or different at least one processor and at least one memory as the recipe generating engine 110, the input data module 230, and the output data module 240.

[0066] In some embodiments of the computer system platform 100 or 200, the recipe generating engine 110, input data module 230, output data module 240, data systems 207, local controller 252, and deployment module 250 comprise the computer platform elements of the computer platform system, particularly when the deployment module is a simulation module. While in embodiments where the deployment module 250 is a physical execution module, thedeployment module may not be included within the ’’computer platform” but is part of the computer platform system.

[0067] In an embodiment, the deployment module 250 may be one or more software systems that can digitally simulate gene engineering, new pigment molecule variation design, biomanufacturing production of the desired microbial colorant and use of the desired colorant according to the generated colorant recipe. Such software systems may be in the form of a digital process simulation software with physics simulation engines that a user can interact with, for example using a graphical user interface. Additionally, such simulations may include color rendering previews, process cost estimates, and compliance checks such as compliance with environmental, social ,and governance (ESG) regulations or application specific standards, guidelines, or requirements (e.g., REACH, ISO, FDA). The user may be able to see visualization of various embodiments of the colorant production and use, and provide feedback and / or tune various parameters of the generated process or adjust the desired input data. The generated results from the simulation software system may be fed back to the data system 207 through the graphical user interface and the input data module 230 or through the local controller 252 of the simulation software for reinforcement learning and recipe refinement, using user-in-the-loop active learning workflows. In some embodiments, the software systems may also include capabilities to simulate strain selection and design and genetic engineering steps of the biomanufacturing process.

[0068] In another embodiment, the deployment module 250 may be a physical lab-scale, pilotscale, or commercial-scale facility that can physically implement the production and use of the desired colorant according to the generated colorant recipe. The physical implementation may be manual, semi-automated, or fully-automated, for example using robotic actuators and monitoring systems. The physical execution may also be modularized by unit operations, such as fermentation, pigment separation, purification, and formulation, which can be separately monitored and controlled. Similar to the digital simulator case, the physical process steps and their results can be displayed to the user in real-time or asynchronously through a graphical user interface, for example.

[0069] The results generated from the physical implementation may be collected (for example using sensors such as sensors 180 to 183) and fed back to the data system 207 or the local controller 252 to improve future outputs of the data system 207 by updating or retraining models with the real results of previous predictions. The sensors 180 to 183 can collect and monitor physical parameters and physical results, such as fermentation process parameters (e.g. nutrient composition and volume, fermenter temperature, pH, dissolved oxygen levels, volume of fermentation harvest), mass or volume of extracted pigments, intracellular versus extracellular pigment ratios, color composition of the produced pigments (e.g. RGB levels, darkness levels), mass or volume of various waste (waste water, waste heat, waste chemicals, waste biomass) generated during the upstreaming and downstreaming steps, amount of GHG emissions, andamount of electrical or thermal energy used during various steps, just to mention few. These measured data can be used to retrain predictive models in the data system 207 or ultimately generate new recipes with improved scores.

[0070] Referring to Figure 2 now, a block diagram depicts, generally at 200, another implementation of the computer platform system 100 of Figure 1. The platform 200 may be implemented using an embedded processor circuit such as a Linux-operated computer. Referring to Figure 2, the platform system 200 includes a microprocessor 202 (or processor), a memory 204, and an input output (I / O) 208, all of which are in communication with the microprocessor 202. The I / O 208 includes a wireless interface 216 (such as an IEEE 802.11 interface) for wirelessly receiving and transmitting data communication signals between the platform system 200 and a network 218. The I / O 208 also includes a wired network interface 210 (such as an Ethernet, USB, I2C, and CAN interface) for connecting the microprocessor 202 to the data system 207 and the input data module 230.

[0071] Referring to Figure 3 now, a graphical user interface depicting an implementation of the user interface 232 is shown. The graphical user interface shown in Figure 3 represents a user interface that receives inputs from the user 238 requesting a desired colorant. The user interface 232, includes fields to receive a desired colorant (or color) and the use application of the colorant. The field for receiving the desired color can include a menu of a list of pre-defined colors or could be any other alternative way to define a color such as a menu of a color palates for the user 238 to choose from. Alternatively, spectral values such as CIELAB or HEX codes may be used to define the target hue. The desired color field may also include a desired darkness field 292 through which the user 238 can choose a darkness level or value for the desired color.

[0072] The desired application field may also include a pre-determined list of applications for the user to choose from, such as, textile, plastic, food and beverage, cosmetics, and pharma. This field defines the desired use case of the colorant and it can impact the designed colorant and its production and use process. For example, a pigment selected for food use may require higher purity, allergen-free substrates, and GRAS-certified microbial strains. In some embodiments, depending on the selected application, further selection options and fields may be presented to the user to further refine application-specific requirements (e.g., allergen-free). The user interface can also include fields to receive additional information from the user 238 related to desired functional or nonfunctional properties of the colorant, such as UV tolerance, heat resistance, surface properties (e.g. hydrophobicity), antibacterial and antioxidant properties. The user may choose one or more of the listed properties. The colorant properties field may also include a desired level field 294 through which the user 238 can choose a qualitative or quantitative level for the desired colorant properties. The user may adjust the desired level for each selected property. In some embodiments, the computer platform system 100 may also suggest properties, to the user through the user interface, based on inferred use cases or complementary market needs.

[0073] The user interface can further include fields to receive additional information from the user 238 related to desired impacts of the production and use of the desired microbial colorant. For example, the user can choose from a pre-defined list of impacts such as sustainability impacts including GHG emissions, energy consumption, and generated waste (e.g. wastewater, chemical, material, and energy) during the production and use of the desired colorant. Other selections may include use of sustainable feedstock (e.g., as a nutrition source for fermentation media), biodegradability of the produced pigment, and avoidance of known allergens or heavy metals. The user may select one or more of the impacts and each impact may further ask for a value for the impact (e.g. amount of GHG emissions indicated in CO2tons per kg of the produced or used colorant). The user may also choose compliance with a regulation or standard related to the production and use of the colorant. A few examples of related regulations and standards include but is not limited to non-GMO regulations, FDA compliance, Europe’s REACH regulation, and OKEO-TEX standard 100 certification. The user interface 232 may also provide a field 296 to the user to receive inputs regarding the importance of the selected desired impacts, for example the importance of the GHG emissions may be higher than the importance of energy consumption levels. Overall, through the user interface, the user may be able to select tradeoffs between yield vs. sustainability, or cost vs. time to market, for example. Such selections can further provide instructions for multi-objective optimization routine conducted by the computer platform system 100 to generate the colorant recipe.

[0074] Referring back to Figure 2, the computer platform system 200 further includes the data system 207 that is configured to store various databases related to the production and use of microbially produced colorants. The data system 207 may include databases of pigment molecules, pigment precursor molecules, molecular pathways for pigment production, pigment producing microorganisms, biomanufacturing process steps, process parameters, color mixing formulas, natural and chemical additives, and colorant application techniques, for example. These databases may be static or may be dynamic and frequently updated.

[0075] The databases in the data system 207 may be receiving dynamic data from input data module 230. For example, the input data module 230 may monitor or crawl 3rdparty databases 234 related to the above-mentioned databases of the data system 207 and fetch new or updated relevant data thereon. Examples of 3rdparty databases 234 include public databases, such as scholarly articles, journals, papers, patent, and other publicly available publications, and private databases such as proprietary and commercial databases.

[0076] The microprocessor 202 can be programmed to run the recipe generating engine 110 and can be configured to generate recipes for microbial dye production and use. In other words, the recipe generating engine 110 can be implemented or realized using the microprocessor 202. In some embodiments, the recipe generating engine 110 can be separate modules that is in data communication with the microprocessor 202. The generated recipe comprises ingredients and instructions to produce and use the desired microbial colorants including data related to thedesired pigment molecule, the pigment precursor molecule, the pigment producing pathways, the microorganism to host the pathways and produce the pigment or precursor molecules, the process steps, the process parameters, the natural and chemical additives during the production process and for addition to the produced colorant for creating the desired formula of the final colorant, the mixing formula of the final colorant, and the application technique of the final colorant (e.g. techniques for depositing or fixating the final colorant on material such as textile, plastic, or food material). The generated recipe can also include calculated impacts which would be created during the production and use of the colorant. Calculated impacts can include qualitative or quantitative amount pertaining to generated GHG emissions, energy consumption, and generated waste (e.g. water, chemical, material, and energy waste) for example.

[0077] The generated recipe may be generated by executing programs or instructions 206 from the memory 204, in the microprocessor 202. The executed programs 206 in the microprocessor 202 may include retrieving desired input data from the user 238 communicated through the user interface 232 and the input data module 230, retrieving and using data from the databases in the data system 207, and using Al algorithms such as pre-trained or evolving machine learning and ANN models to generate the recipe.

[0078] In some embodiments, the recipe generating engine 110 includes or communicates with a predictive Al model trained to identify and evaluate new colorant candidates derived from microbial or chemical precursors. Such predictive Al models (e.g., model structure and parameters) or instructions for their training, retraining, and inference could be stored in the memory 204 and the programs 206. The Al models and the use instructions can be executed by the microprocessor 202. The predictive Al models may be based on one or more machine learning (ML) architectures such as decision trees, deep neural networks, support vector machines, or ensemble models. They may be pre-trained on datasets comprising known pigment molecules and their associated color values, spectrophotometric properties, biosynthetic pathways, host organisms, and production conditions. The models can then infer likely color outputs and functional properties for unseen or hypothetical molecules based on their molecular descriptors.

[0079] In some embodiments, the predictive Al model uses input features such as 2D and 3D molecular descriptors (e.g., from RDKit or Dragon), molecular fingerprints, topological indices, and / or quantum mechanical properties derived from computational chemistry simulations. These descriptors can be fed into classification or regression models to predict color attributes such as hue, darkness, chroma, and color stability under pH, heat, or UV stress. In an embodiment, the platform system 100 leverages a pretrained deep neural network or random forest model similar to that disclosed by Zhang et al., “Artificial intelligence deciphers codes for color and odor perceptions based on large-scale chemoinformatic data” GigaScience, 2020, which demonstrated high-accuracy color prediction across thousands of pigment molecules.

[0080] The predictive Al model may also simulate combinations of precursors and pathway variants, including unnatural biosynthetic reactions or gene modifications, to explore a broaderchemical space. For example, rather than only evaluating violacein, the system may simulate how co-feeding or pathway engineering using precursors like tryptophan, catechol, glutamine, or phenazines may result in structurally novel and aesthetically distinct pigments. In some embodiments, predictive models trained on reaction data can estimate transformation likelihoods and reaction conditions required for successful pigment synthesis. In some embodiments, the Al model performs combinatorial screening and virtual experimentation by generating and testing hypothetical pigment variants, modifying functional groups, oxidation states, or substitution patterns. Each variant is analyzed in silico for its predicted absorbance spectrum, biosynthetic pathway complexity, host compatibility, and application-relevant performance metrics. These predictions may be made in batch or active-learning loops to efficiently converge on topperforming candidates.

[0081] In some embodiments, the predictive Al model is used for optimization of colorants, either instead of or in addition to discovery of colorants. For example, once a promising microbial pigment candidate is identified, reinforcement learning or Bayesian optimization may be applied to iteratively suggest improvements to the gene expression levels, culture media composition, process parameters (e.g., pH, temperature), or extraction methods that improve pigment yield, reduce cost, or enhance color stability. These optimized recipes may be dynamically updated in the output data module 240 and shared with the deployment module 250 for implementation and feedback. Feedback from the physical or simulated deployment module 250 may be looped back into the predictive Al model to further refine their predictions. For instance, discrepancies between predicted and observed color outputs, yield, or pigment solubility may be used to retrain or recalibrate the models. The platform system 100 may employ transfer learning to adapt pretrained color models to specific microbial hosts, fermentation conditions, or regulatory requirements, thereby increasing the accuracy and relevance of the recipe predictions for real- world applications.

[0082] Once colorant predictions are obtained by the predictive Al model, one or more recipes may be generated using a ranking or scoring procedure executed by the microprocessor 202. The microprocessor 202 may rank candidate pigment molecules or pigment-producing pathways based on their match with the user’s desired color value or target specifications. For example, if the user selects a target green hue with antioxidant properties and pH stability, the microprocessor 202 can generate a list of candidate molecules or pathways that meet or approximate these traits. In one embodiment, candidate ranks or scores are weighted by predicted color similarity, expected production yield, functional property match, environmental impacts, and predicted compliance with desired application constraints (e.g., food safety, textile adhesion, REACH compliance).

[0083] In some embodiments, the user may be able to select (e.g., through the user interface 232) weights assigned to favor specific dimensions (e.g., regulatory compliance over cost). The microprocessor 202 may further use the selected weights and score the candidate generatedrecipes.

[0084] In addition to molecular color prediction, the platform system 100 may include or interface with models that predict biosynthetic feasibility and microbial compatibility. These models can assess whether a given pigment candidate can be feasibly synthesized by a known microbial chassis (e.g., E. coli, Pichia pastoris, Streptomyces) by matching the molecular structure of the candidate to known or predicted biosynthetic gene clusters (BGCs) or using retrosynthetic algorithms. In some embodiments, biosynthetic feasibility is assessed using curated databases such as MIBiG, KEGG, or antiSMASH, or through proprietary machine learning models trained on engineered strain design data.

[0085] In another embodiment, the recipe generating engine 110 may include a user-guided interface that allows the user to explore Al-suggested pigment candidates, pathway variants, or process modifications. The interface may provide interactive visualizations of the molecular structures, predicted color swatches, associated biosynthetic routes, host organisms, and functional attributes (e.g., UV absorption curves). The user may select preferred candidates and instruct the controller to generate a full production recipe or to simulate the selected candidates within a virtual biomanufacturing environment.

[0086] The platform system 100 may further include a feedback refinement module (not shown in figures), which takes in results from the deployment module 250 (e.g., real fermentation yields, color stability) and updates parameters of the recipe generating engine 110 accordingly. Over time, this closed-loop learning improves the predictive accuracy of recipe generation. In industrial use, the platform may become increasingly customized to specific equipment, local resources, or host strain capabilities.

[0087] The incorporation of the predictive Al model can enable the platform system 100 to transcend traditional limitations in colorant discovery and rather than relying solely on known pigment pathways, the system can propose entirely new, biologically feasible pigment molecules and recipes optimized for user-defined constraints. This empowers users to create custom colorant solutions with novel hues, improved environmental profiles, and application-specific performance while transforming microbial colorant production from a trial-and-error process into a rational and algorithmically driven design framework.

[0088] In some cases, the platform may suggest novel pigments not found in nature, using generative models trained on molecular color rules. For example, a generative adversarial network (GAN) embedded in the platform can propose new pigment molecules with desired hue and functional groups. These molecules are evaluated for synthesizability using retrosynthesis prediction engines and then matched to potential microbial hosts via pathway mining tools. In one instance, a newly generated catechol-indole hybrid molecule is predicted to produce a rare teal pigment with antioxidant properties. The system then builds a hypothetical biosynthetic gene cluster, simulates codon optimization, and estimates production feasibility in Streptomyces coelicolor.

[0089] An example of such programs is illustrated in Figure 4A. Referring to Figure 4A now, a flowchart depicting blocks of code or instructions for directing a processor, such as the microprocessor 202, to generate a recipe related to a desired microbial colorant, is shown at 400. The flowchart 400 generally outlines a systematic approach for generating a recipe for microbial colorant production and use. The flowchart 400 starts at block 402 by displaying input fields to the user 238 in the user interface 232. The input fields are pertaining to a desired colorant, its related properties, its application, and related impact constraints. The displayed input fields may be similar to the input fields as shown in the user interface 232 under Figure 3.

[0090] At block 404 the flowchart instructs the microprocessor 202 to receive the user 238 inputs to the input fields of the user interface 232. In some embodiments, the flowchart 400 may include a feedback loop from block 404 to block 402 to facilitate dynamic display input fields. For example, once the user 238 chooses a certain desired colorant and a certain desired impact, a new input field, such as a menu with several suggesting process steps, may be displayed on the user interface 232 to provide further options to the user 238 to further define its needed colorant.

[0091] At block 406, the microprocessor 202 is instructed to select a pigment molecule, a precursor molecule, pigment-producing pathways, and the corresponding pigment-producing or precursor-producing microorganisms. The microprocessor 202 may select these strain design variables (which may be dependent or independent variables) by querying related databases from the data system 207 and selecting one or more values for these strain design variables such that the values can fulfil the desired color indicated in the input data by the user 238. In some embodiments, block 406 may involve an Al models to select these values. In some other embodiments, block 406 may involve resorting to gene engineering and suggesting creation of new molecules, new pathways, and new genetically engineered microorganisms, for example using Al models or 3rdparty expert systems, as further candid values for the strain design variables to fulfil the user-defined inputs and requirements.

[0092] At block 408, the microprocessor 202 is instructed to select a plurality of process steps and the related parameters of these process steps. Selecting a plurality of process steps may involve selecting an archetype of the process steps (e.g. gene-engineering step, culturing step, fermentation step, pigment extraction step, etc.), a process sub-step for each archetype (e.g. gene extraction, gene cloning, preparing seed culture, preparing fermentation medium, pigment purification, biomass processing, etc.), and an instance or value for the steps or sub-steps (e.g. silica gel based gene extraction techniques, batch fermentation, solvent purification of pigments, etc.), upon determining instances or values for these production process variables, process parameters related to these variables can be identified and selected (e.g. for batch fermentation, selecting related process parameters may include selecting carbon versus nitrogen sources as medium, selecting temperature range of the fermenter, selecting the pH range of the fermented broth, selecting the duration of the fermentation). The microprocessor 202 may select these production variables (which may be dependent or independent variables) by retrieving relateddatabases from the data system 207 and selecting one or more values for these process variables such that the values can fulfil the desired color, or the cumulative effect of the desired color, it desired properties, its application, and the desired impact constraints (e.g. complying to certain regulations and fulfilling certain impact levels) indicated in the input data by the user 238. In some embodiments, block 408 may involve an Al models to select these values.

[0093] At block 410, the microprocessor 202 is instructed to select none, one, or more chemical additives and colorant composition formulations to obtain the final colorant which should match the desired colorant indicated by the user 238 (for example, in case of food coloring no chemical additives may be added in the formulation of the final colorant). The microprocessor 202 may select these colorant composition variables (which may be dependent or independent variables) by retrieving related databases from the data system 207 and selecting one or more values for these colorant composition variables such that the values can fulfil the desired colorant requirements indicated in the input data by the user 238. In some embodiments, block 410 may involve an Al model to select these values.

[0094] In some embodiments, the flowchart 400 further includes instructions for the microprocessor 202 to select one or more values related to techniques for applying the final colorant to a material as defined by the user 238 in the desired application field of the user interface 232. Additionally, the flowchart 400 includes instructions for the microprocessor 202 to calculate impact metrics related to desired impacts as defined by the user 238.

[0095] The combination of the selected values in blocks 406 to 410 can form the instructions or recipe for production and use of a microbially produced colorants according to the input requirements by the user 238.

[0096] The generated recipe can be communicated to an output data module 240, which could be a separate computer or a memory to store the generated colorant recipe data, through the I / O module 208. The output data module 240 can communicate the generated output data to the user interface for display to the user 238, for example. The output data module 240 may further communicate the generated output data to the deployment module 250. The deployment module 250 is configured to simulate and / or physically execute the generated colorant recipe in part or in its entirety. As mentioned earlier, the deployment module 250 may be a software system that can digitally simulate the gene engineering, new pigment molecule variation design, the biomanufacturing, and the use of the desired colorant according to the generated colorant recipe. In this example, the software system may be in the form of a digital process simulation software with physical simulation engines that the user 238 can interact with through the user interface 232. The user 238 may be able to see visualization of various embodiments of the colorant production and use, and provide feedback and / or tune various parameters of the generated process or adjust the desired input data. The generated results from the simulation software system may be fed back to the data system 207 through the input data module 230. In some embodiments, the software system may also include capabilities to simulate strain selection anddesign and genetic engineering steps of the biomanufacturing process.

[0097] Additionally, the deployment module 250 may be a physical lab-scale, pilot-scale, or commercial-scale facility that can physically implement the production and use of the desired colorant according to the generated colorant recipe. The physical implementation may be manual, semi-automated, or fully-automated, for example using robotic actuators and monitoring systems. Again, the process steps and their results can be displayed to the user 238 through the user interface 232. The generated results from the physical implementation may be collected (for example using sensors such as sensors 180 to 183) and fed back to the data system 207 through the input data module 230.

[0098] In some embodiments, the deployment module 250 may include both a software simulation system and the physical production facility in conjunction or simultaneously implementing the generated recipe. In other embodiments, the deployment module 250 may include one of the simulation or physical systems at a time. The deployment module 250 may further be in communication with a local controller 252 such as a local computer or processor circuit to oversee and control the deployment of the generated recipe. In one embodiment the local controller 252 is a local computer circuit of a biomanufacturing facility. The generated recipe may be loaded to the local controller 252 to control the deployment and ensure that the desired colorant is being produced according to the generated recipe. The local controller 252 may be in wired or wireless data communication with the microprocessor 202 of the computer platform system 200, for example through the I / O 208. In some embodiments, the local controller 252 may be implemented by the microprocessor 202 (i.e. be the same as the microprocessor 202).

[0099] The local controller 252 may be configured to monitor the implementation of the generated recipe and provide feedback to the user 238 or adjust the process parameters. Referring to Figure 4B, a flowchart depicting blocks of code or instructions for directing a processor, such as the microprocessor 202 and the local controller 252, is shown at 420. The flowchart 420 generally outlines a systematic approach for directing the deployment or implementation of a generated recipe for microbial colorant production and use. The flowchart 420 starts at block 422 by displaying input fields to the user 238 in the user interface 232. As mentioned before, the input fields may be pertaining to a desired colorant, its related properties, its application, and related impact constraints. The displayed input fields may be similar to the input fields as shown in the user interface 232 under FIG 3. At block 424, the microprocessor 202 is instructed to receive the user 238 inputs to the input fields of the user interface 232. At block 426, the microprocessor 202 is instructed to generate a recipe for colorant production and use. This step may include similar steps as to blocks 406 to 410 in Figure 4A for generating the recipe.

[0100] At block 428, the microprocessor 202 is instructed to communicate the generated recipe for deployment on a physical bioproduction facility. For example, the generated recipe may be displayed or presented to an operator of a bioproduction facility for execution. The execution may be manual, semi-automated, or fully automated. The generated recipe is loaded to the localcontroller 252. In case of fully automated facility, the local controller 252 may be configured to control the automated steps for executing the generated recipe and the monitoring thereof. In case of manual execution, the local controller 252 may be configured to monitor the execution of the generated recipe. In any case, the local controller 252, as instructed by block 430, is instructed to monitor a plurality of physical parameters of the physical bioproduction facility. The monitoring data can be passed on the microprocessor 202 or the data system 207.

[0101] At block 432, the microprocessor 202 or the local controller 252, may determine a required change in the recipe and are instructed to generate a modified recipe. In some embodiments, such as the flowchart 440 as shown in Figure 4C, the determined change in the recipe may be an optional change rather than a required change. For example, the optional change may involve a new process step or a new process parameter that not only complies with the user’s requirements but also can improve the yield or impact metrics beyond the user’s expectations. In this case, the optional change to the recipe is notified, by the microprocessor 202 or the local controller 252, to the user 238, e.g. through the user interface 232, at block 442. The user 238 may approve or reject the notified optional changes to the recipe. At block 444, if the optional changes are approved by the user 238, the changes will then proceed to block 434 for instructing the changes to the deployment module 250.

[0102] At block 434, the modified recipe is communicated with the deployment module 250 and the local controller 252 to adjust the physical bioproduction process. The required change in the recipe and the communication of the modified recipe to the deployment module 250, are described in several examples below, wherein Figures 5A and 5C are referred to.

[0103] Referring to Figure 5A, a weight sensor (not shown in figures) may monitor the amount of extracted pigment 530. The weight sensor measurement data may be transmitted to the local controller 252 or the microprocessor 202. The original generated recipe may have instructed production parameters to produce 10 kg of extracted pigment at step 530. According to one embodiment, however, the mass of the extracted pigment at 530 is 8kg and lower than the desired amount required by the recipe. In this case, the microprocessor 202 determines the discrepancy (as instructed at block 432) and may generate a modified recipe to instruct the deployment module 250 (the bioproduction facility) to produce another smaller batch to produce a 2kg deficit (as instructed by block 434). The new recipe is passed on to the facility for deployment.

[0104] In another example, at block 434, instead of instructing the facility to introduce another batch of pigment production, the recipe may be altered to introduce additives to the extracted pigments at step 560 such that the total mass of the final produced colorant 590 matches the desired 10kg amount, while maintaining other required properties and impacts requested by the user.

[0105] Referring to Figure 5C now, a process flow for using resulting microbial biomass as a darkening agent, is generally shown at 500c. According to Figure 5C, a precise color detection sensor (not shown in figures) may measure the darkness levels of the extracted pigments 530,and have reported lower darkness levels compared to the desired darkness levels required by the user 238 or instructed by the original generated recipe, and thus a required change in the recipe is determined (as instructed by block 432) for increasing the darkness levels of the final colorant. According to instructions generated at block 434 the microprocessor 202 may generate a modified recipe by introducing a biochar production process step 550 from spent microbial biomass (biomass that remains after pigment extracting process 520) and adding the produced biochar as a darkening additive to the extracted pigments 530 to create final colorant 590 that matches the desired properties required by the user 238.

[0106] In some embodiments, similar steps explained in Figures 4A to 4C, may be used for deployment into a simulation software.

[0107] Examples

[0108] Example 1. Violet color for plastic coloring

[0109] In one embodiment, a user may input a request through the graphical user interface 232 for a “deep violet colorant suitable for plastic coloring that also exhibits high UV absorption and heat resistance”. Upon receiving the input, the controller accesses the data system 207 and queries the pigment molecule database for known violet-producing compounds such as violacein and derivatives. Simultaneously, the predictive Al model generates a set of modified tryptophan- based structures through in silico mutation and calculates their color output and functional performance using trained deep neural networks. The top-ranked molecule (e.g., a methylated violacein analog) is identified as having higher thermal stability and comparable color output to natural violacein. The microprocessor 202 then selects Chromobacterium violaceum as the native producer but also suggests an engineered E. coli strain with codon-optimized vioABCDE operon to improve scalability and safety. The fermentation process is simulated and optimized for yield and cost under defined bioreactor conditions.

[0110] Example 2. Orange-red colorant for cosmetics

[0111] The user selects a bright orange-red pigment for cosmetics with antioxidant functionality and REACH compliance. The microprocessor 202 invokes the predictive Al model to search for prodigiosin variants and evaluates their color and bioactivity profile. A novel prodigiosin derivative with enhanced radical scavenging potential is identified by the Al model using a pretrained functional activity predictor. The generated recipe suggests Pichia pastoris as the microbial chassis, engineered to express a hybrid gene cluster constructed from Serratia and Hahella species. The generated recipe further recommends pH-controlled fermentation and an aqueous extraction method to minimize use of solvents. REACH-compliant safety data, including cytotoxicity predictions, are also included in the recipe, allowing the user to validate regulatory compatibility.

[0112] Example 3. black pigment for textile dying using a precursor

[0113] In yet another example, a user enters a query for a black pigment for textiles with high wash fastness and low-impact environmental solution. The platform’s microprocessor 202identifies eumelanin produced from a precursor (e.g., tyrosine or other intermediates upstream or downstream of the final pigment biosynthesis) as a viable candidate. It also models structurefunction relationships and suggests adding an iron ion co-factor during fermentation to increase pigment polymerization and bonding strength. The platform selects a recombinant E. coli strain with high enzyme expression (e.g. tyrosinase) and simulates a fed-batch fermentation with glucose-limited feed. For the downstream step, the system recommends a non-aqueous extraction using ethanol, followed by lyophilization. Application instructions are generated, including a mordanting step with natural tannins to improve wash fastness when used on cotton substrates.

[0114] In alignment with the user’s preference for low-impact environmental solution, the suggested upstreaming process includes a natural fermentation medium derived from agricultural waste streams. Specifically, sugar beet pulp hydrolysate is suggested as a carbon source (providing glucose and trace minerals), while corn steep liquor is suggested as a rich nitrogen source containing amino acids, vitamins, and growth factors. These substrates could be selected from a curated feedstock database within the data system 207, evaluated for both cost efficiency and sustainability impacts. The fermentation is conducted in a 10L stirred-tank bioreactor using a fed-batch mode. The suggested fermentation parameters include a fermentation temperature of 37°C, pH range of 6.8 ± 0.1 , and dissolved oxygen of at least 30% saturation (e.g., controlled via cascade airflow). The suggested feed strategy includes glycerol-based feed every 8 hours and the suggested total fermentation duration is 72 hours.

[0115] For pigment recovery, a non-aqueous surfactant-assisted extraction method is employed using Tween-80 at 1% concentration in ethanol. The mixture is stirred for 30 minutes and centrifuged at 5,000 rpm for phase separation. A secondary filtration using a 0.45 pm membrane removes residual biomass.

[0116] In the final formulation step, the pigment is suggested to be mixed with a naturally derived mordant (e.g., tannic acid) and spray-dried with starch-based encapsulants to improve dispersion in textile dye baths. The recipe also includes an application instruction for dye fixation: immersion at 90°C for 45 minutes in a 1 :50 pigment-to-fabric ratio bath, followed by cold rinsing. This results in superior wash fastness and UV resistance, and biodegradability that is in alignment with textile applications (e.g., ISO 105-C06, OEKO-TEX Standard 100).

[0117] The microprocessor 202 is also estimating a 35% reduction in GHG emissions compared to synthetic dye production, based on the selected natural feedstocks and energy input modeling.

[0118] Across all these examples, the platform system 100 may include a scoring system to rank candidate solutions by cost, performance, sustainability, and user-defined preferences. The scoring system may integrate Al predictions, experimental data, user feedback, and third-party databases. Users may also view detailed rationale for each selection, such as molecular descriptors contributing to predicted hue or factors influencing microbial host compatibility.

[0119] While specific embodiments and examples have been described and illustrated throughout this disclosure, such embodiments should be considered illustrative only and not as limiting the disclosed embodiments as construed in accordance with the accompanying claims.

Claims

CLAIMS1 . A method for discovery and optimization of production and / or use of microbial colorants from microorganisms by a computer platform system, the method comprising: receiving, through a user interface of an input data module of the computer platform system, user input data, from a user, related to a desired colorant; generating, by at least one recipe generating engine of the computer platform, a recipe for production and / or use of the desired colorant using microorganisms, based on the user input data; communicating, by an output data module of the computer platform, the recipe to a deployment system for implementing the recipe; and monitoring, by at least one processor of the computer platform system, a plurality of parameters of the deployment system during the implementation of the recipe.

2. The method of claim 1 , further comprising: determining, by the processor, at least one change in the recipe; generating, by the processor, a modified recipe; and communicating, by the processor, the modified recipe to the deployment system for implementation.

3. The method of claim 2, wherein the at least one change is notified to the user, by the processor, and the modified recipe is generated by the processor once the user approves the change.

4. The method of claim 1 , wherein the recipe is generated by the processor by: selecting, at the processor, one or more microorganisms from a plurality of microorganisms, the one or more microorganisms capable of producing pigment molecules or precursor molecules that are an ingredient of the desired colorant; selecting, by the processor, a set of biomanufacturing processes from a plurality of sets of biomanufacturing processes, the set of biomanufacturing processes to yield a desired amount of the desired colorant; selecting, by the processor, a set of process parameter values related to the set of the biomanufacturing processes for causing the set of biomanufacturing processes to yield a desired amount of the desired colorant; and generating the recipe, wherein the recipe is the combination of the selected one or more microorganisms, set of biomanufacturing processes, and set of processparameter values.

5. The method of claim 4, wherein the method further comprising selecting the pigment molecules from a plurality of pigment molecules, and wherein the recipe further includes instructions on engineering the pigment molecules.

6. The method of claim 4, wherein the selected microorganisms produce pigment precursor molecules, and the pigment precursors are polymerized or chemically modified by one or more chemical agents external to the microorganism to form the final colorant.

7. The method or system of claim 5, wherein the pigment precursor comprises tryptophan, glutamine, phenazines, tyrosine, or a catechol compound.

8. The method of any of claims 4 to 7, wherein the method further comprising selecting an application technique from a plurality of application techniques to yield a desired use of the desired colorant, and wherein the recipe further includes the selected application technique.

9. The method of any of claims 4 to 8, wherein the selected one or more microorganisms includes one or more microorganisms that may need to be genetically modified or engineered.

10. The method of any of claims 1 to 9, wherein the recipe is transmitted by the processor for display to the user.

11. The method of any of claims 1 to 10, wherein the deployment system is a digital biomanufacturing process simulation system, a physical biomanufacturing process facility, or a combination thereof.

12. The method of any of claims 1 to 11 , wherein the deployment system comprises a local controller and wherein communicating, by the processor, the recipe to the deployment system comprises transmitting the recipe to the local controller.

13. The method of claim 11 , wherein physical process data is collected by one or more sensor of the physical biomanufacturing process facility and the physical process data is transmitted to the local controller or the processor.

14. The method of any of claims 1 to 13, wherein the user input data includes one or more of: a desired color value; a desired property value for the desired colorant; a desired application value for the desired colorant; and a desired impact value for the desired colorant.

15. A computer program to control production and / or use of biosynthesized colorants from microorganisms, the computer program once executed by a processor causes a processor to: receive user input data from a user related to a desired colorant; generate a recipe for production and / or use of the desired colorant using microorganisms; communicate the recipe to a deployment system for implementing the recipe; and monitor a plurality of parameters of the deployment system during the implementation of the recipe.

16. The computer program of claim 15, wherein the processor is furthered configured to: determine at least one change in the recipe; generate a modified recipe; and communicate the modified recipe to the deployment system for implementation.

17. A system to manage biosynthetic production and / or use of microbial colorants from microorganisms, the system comprising a computer platform comprising a processor and a memory storing instructions executable by the processor to cause the computer platform to: receive user input data from a user related to a desired colorant; generating a recipe for production and / or use of the desired colorant using microorganisms; communicate the recipe to a deployment system for implementing the recipe; and monitor a plurality of parameters of the deployment system during the implementation of the recipe.

18. The system of claim 17 wherein the computer platform is further configured to: determine at least one change in the recipe; generate a modified recipe; and communicate the modified recipe to the deployment system for implementation.

19. A computer platform system for discovery and optimization of production and / or use of microbial colorants from microorganisms, the computer platform system comprising: at least one memory storing processor executable instructions;at least one processor, configured to execute the processor executable instructions for: an input data module, wherein the input data module receives user input data related to a desired colorant through a user interface; a recipe generating engine, wherein the recipe generating engine receives the user input data, queries at least one data system for related data to the user input data, and generates a recipe for production and / or use of the desired colorant using microorganisms, based on the user input data and the related data; an output data module, wherein the output data modules outputs the recipe including a plurality of parameters for implementation of the recipe; and a deployment system, wherein the deployment system receives the recipe from the output data module and implements the recipe.

Images