Method for calculating molecular weight of biological macromolecular material based on AI algorithm

Abstract

The invention relates to a method for calculating molecular weight of a biological macromolecular material based on an AI algorithm. Part of biopolymer materials are difficult to form monomolecular chains after being dissolved into an aqueous solution, the ionic liquid is selected as a dissolution system, the monomolecular chains can be well formed after the macromolecular materials are dissolved, and then rheological data after biomacromolecules are dissolved in the ionic liquid are tested by using a rheological method. According to the method, a Rouse model is selected as a polymer solution model, and key parameter values are obtained by measuring a large amount of data, training the model by using an AI algorithm and carrying out nonlinear fitting, so that the molecular weight value of the measured material is calculated. The calculation method has the characteristics of accuracy and rapidness, and can be well popularized to the field of computer research of high polymer materials.

Description

Calculation method of molecular weight of biomacromolecule materials based on AI algorithm Technical Field

[0001] The present invention relates to a method for measuring the molecular weight of biomacromolecule materials, and in particular to a method for calculating materials whose molecular weight is difficult to determine based on a rheological model, and solving the molecular weight of the biomaterial using AI nonlinear fitting. Background Art

[0002] With the continuous development of biomolecular science, biomacromolecules have garnered increasing attention. Materials such as silk protein, hyaluronic acid, collagen, recombinant collagen, and sericin have been used as important medical and cosmetic ingredients. The biological properties of many biomacromolecules are closely related to their relative molecular weight. Materials with different relative molecular weights exhibit varying physiological activities, some even exhibiting completely opposite effects. For example, hyaluronic acid with a molecular weight greater than 2000 kDa exhibits excellent moisturizing, viscoelasticity, lubrication, and anti-inflammatory properties, making it suitable for ophthalmic surgery and bone and joint injections. Hyaluronic acid with a molecular weight of 10 kDa to 80 kDa can be taken orally and absorbed in the intestine, exerting cosmetic and skin-beautifying effects. Similarly, the swelling and transmittance of silk protein hydrogels decrease with increasing molecular weight. Therefore, accurately determining the molecular weight of polymer materials allows for more effective utilization of their properties.

[0003] Traditional methods for determining protein molecular weight include gel electrophoresis (SDS-PAGE), light scattering (SLS), small-angle neutron diffraction (SANS), and gel permeation chromatography (GPC). However, these methods often have limitations, including the inability to accurately measure the molecular weight of certain polymers that exhibit aggregation in solution. For example, the unique structure of silk protein molecules (a quasi-multiblock copolymer structure with alternating hydrophilic and hydrophobic segments) causes regenerated silk protein to easily aggregate into micelles and other aggregates in solution, rather than existing as "free" single molecular chains. Therefore, it is necessary to first find a good solvent in which silk protein can exist as "free" single molecular chains. Furthermore, a suitable test method is needed to quantitatively describe the chain state and viscoelastic behavior of polymer chains in solution.

[0004] Furthermore, finding a suitable model for polymer solutions is crucial. To investigate the structural and dynamic characteristics of polymer solutions, researchers have developed several coarse-grained geometric models, such as the bead-stick model, the bead-spring model, the pearl-necklace model, and the reptation model. Further exploration is needed to determine which model is most suitable for calculating the molecular weight of silk protein.

[0005] Molecular weight calculations based on the aforementioned models are complex, requiring extensive data testing and computational effort. Manual calculations are inefficient and lack precision. Introducing AI algorithms into traditional polymer molecular weight calculations can address timeliness and accuracy challenges, significantly improving the efficiency of researchers and researchers and furthering innovation in biopolymer computation.

[0006] Summary of the Invention

[0007] The problem to be solved by the present invention is to address the difficulties in measuring the molecular weight of existing biopolymer materials and invent a method for calculating the molecular weight of biopolymer materials based on an AI algorithm.

[0008] This paper proposes a rheology-based method for measuring the molecular weight of biomacromolecules. Using ionic liquids as the dissolution system and the Rouse model as the polymer solution model, the method measures large amounts of data and trains the model using an AI algorithm to obtain key parameter values, thereby calculating the molecular weight of the measured material.

[0009] First, the present invention screens out ionic liquids as good solvents for dissolving biomacromolecules by testing different solvent systems.

[0010] Ionic liquids refer to ionic compounds in liquid form, or with a melting point below a certain temperature. In recent years, ionic liquids have attracted significant attention worldwide as a new type of green solvent. They possess numerous excellent properties, including excellent thermal stability, structural designability, low vapor pressure, non-toxicity, and recyclability, benefiting both environmental protection and operator health. Ionic liquids are not only widely used as solvents and reaction media in organic synthesis reactions, but are also considered green solvents for some natural macromolecules. Ionic liquids also exhibit excellent properties in dissolving biopolymers, enabling the unique structure of silk protein molecules to exist as single chains in them.

[0011] Secondly, the present invention screened out the Rouse model from a variety of polymer solution models through simulation calculations. This model is more suitable for calculating the molecular weight of biopolymers. The relationship between polymer viscoelasticity data and molecular weight function in the Rouse model is as follows:

[0012]

[0013]

[0014]

[0015] Then, rheological methods are used to test the rheological data of biomacromolecules dissolved in ionic liquids. Rheological methods can well reflect the chain state and viscoelastic behavior of polymer chains in solution. Rheology can be used to measure rheological data such as viscoelasticity of polymers in ionic liquids.

[0016] Finally, the present invention introduces AI algorithms to solve the problem of calculating large amounts of rheological data, which improves timeliness and accuracy, increases the work efficiency of R&D personnel, and further opens up innovations in the field of biopolymer computing.

[0017] The present invention is carried out through the following technical solutions:

[0018] A method for calculating the molecular weight of biomacromolecule materials based on an AI algorithm comprises the following steps:

[0019] S1. Sample preparation: Take a biopolymer sample to be tested and dissolve it in an ionic liquid to obtain a sample solution to be tested;

[0020] S2. Sample testing: The sample prepared in step S1 is placed on the rheometer for sample testing and the required data is calculated;

[0021] S3. Establish the Rouse model for AI algorithm evaluation;

[0022] S4. Establish the Rouse model to fit the biopolymer system;

[0023] S5. Calculate molecular weight related data using the Rouse model optimized by AI algorithm;

[0024] Preferably, in step S1, the specific method for preparing the sample is:

[0025] a) Dissolution: Add the dry biopolymer material to the ionic liquid and mix, and dissolve at room temperature;

[0026] b) drying and removing water;

[0027] c) Dissolve under reduced pressure.

[0028] Preferably, in step S1, the ionic liquid is 1-allyl-3-methylimidazolium chloride ([AMIm]Cl) and 1-hexyl-3-methylimidazolium hydrogen sulfate ([Hmim]HSO4).

[0029] Preferably, in step S1, the biopolymer material comprises: silk protein, hyaluronic acid, collagen, recombinant collagen, and sericin. Preferably, the silk protein is selected from mulberry silk protein, spider silk protein, and tussah silk protein. The mulberry silk protein is the portion remaining after the sericin is removed from the silkworm cocoon, also known as silk fibroin. Preferably, in step S1, an ionic liquid solution of the biopolymer silk protein is prepared at a concentration of 0.1%-50%, preferably 1%-25%, and more preferably 5-20%.

[0030] Preferably, in step S1, freeze drying is selected for drying and removing water: the sample is placed in liquid nitrogen for freezing to fully solidify the ionic liquid, and then placed in a freeze dryer for fully dehumidifying, taken out, and sealed and thawed at room temperature.

[0031] Preferably, in step S1, the specific method of dissolving under reduced pressure is: heating the silk protein ionic liquid mixture prepared in step b in an oil bath while stirring, and performing reduced pressure distillation with an oil pump to remove any trace moisture that may remain in the mixture and eliminate bubbles, and heating until the biopolymer is completely dissolved.

[0032] Preferably, the silk protein ionic liquid mixture prepared in step S1(b) is heated in an oil bath at a temperature of 0-180°C while stirring, and is subjected to reduced pressure distillation using an oil pump with a vacuum degree ranging from -0.01 MPa to -0.5 MPa to remove any trace moisture that may remain in the mixture and eliminate bubbles, and is heated until the biopolymer is completely dissolved; heating and stirring are stopped, and the mixture is cooled to obtain a uniform and transparent solution.

[0033] More preferably, the obtained biopolymer ionic liquid solution is stored in a sealed and dry environment at room temperature for future use.

[0034] Preferably, in step S2, the specific method of testing the sample with a rheometer is:

[0035] a) Storage modulus and loss modulus test

[0036] Parallel clamps were selected and the test was protected by nitrogen purge through a temperature-controlled hood during the test.

[0037] The linear dynamic elastic test was used, that is, the strain amplitude of the oscillation mode was controlled below 50% to ensure that the strain amplitude of the oscillation mode was within the frequency scanning range (1×10 2 rad / s~30×10 -2 rad / s) and the storage modulus and loss modulus are linear.

[0038] Frequency sweeps were performed at the following temperatures (0°C, 10°C, 20°C, and 30°C) to obtain the storage modulus and loss modulus curves of the samples at different temperatures.

[0039] b) Viscosity test

[0040] Parallel clamps were selected and the test was protected by nitrogen purge through a temperature-controlled hood during the test.

[0041] Steady-state test experiment: the shear rate is scanned from low to high, the shear rate range is: 10 -5 ~10 5 s -1 . Record the viscosity value of the platform curve.

[0042] Preferably, in step S3, the specific method of establishing an AI algorithm to evaluate the Rouse model is: based on a commonly used AI model optimization algorithm, an algorithm is established to evaluate the quality of the Rouse model prediction results. This algorithm describes the degree of fit between the Rouse model prediction and the experimental results. The fitting process requires that the theoretical prediction results of the Rouse model are close to the experimental results. Therefore, the following algorithm parameters need to be set in advance:

[0043] a) Set a reasonable error threshold ε th , when the prediction error ε of the Rouse model is less than ε th , then it is believed that the accuracy of the Rouse model is high enough to truly describe the situation of the experimental polymer system;

[0044] b) Set a reasonable initial learning rate α0 to control the iteration speed of the algorithm;

[0045] C) Use the gradient descent method to continuously modify the model parameters until the prediction error is less than 0.01.

[0046] Preferably, in step S3, the AI ​​model optimization algorithm includes gradient descent, conjugate gradient method, Adam, AdamGrad, and RMSProp.

[0047] Preferably, in step S3, the error threshold ε th The selection range is 0.001-0.25.

[0048] Preferably, in step S3, the learning rate α0 is selected in the range of 0.0001-0.1.

[0049] Preferably, in step S4, the specific method of establishing the Rouse model to fit the biopolymer system is as follows: a) Assuming that the structure of the biopolymer system satisfies the Rouse model and the molecular weight distribution of the system satisfies the normal distribution. Model the experimental system and initialize the average value of the molecular weight distribution. and standard deviation ΔM0.

[0050] b) Use and ΔM0 to construct the normal distribution of molecular weight [Formula (4)], sampling is performed based on this distribution to establish the initial simulated physical system.

[0051]

[0052] c) Based on the molecular weight distribution of the simulated physical system, the relaxation time τ under mode p is calculated according to formula (3): ip .

[0053] d) Considering the contribution of all vibration modes p to the storage modulus G′ and dissipation modulus G″, calculate G′ and G″ of the simulated physical system according to formulas (1) and (2). The density ρ of the polymer in the solution, the zero shear viscosity η0 of the solution, and the viscosity η of the solvent used in the calculation are s This is consistent with the experimental data.

[0054] e) Based on the simulation data obtained from the Rouse theoretical model, the relationships between log G′ and log ω and log G″ and log ω of the simulated physical system can be calculated. The difference ε between the predicted values ​​of log G′ and log G″ from the Rouse model and the experimental values ​​measured in step S2 can be obtained by solving the root mean square formula.

[0055] f) The difference ε between the calculated log G′ and log G″ data describes the degree of agreement between the Rouse model and the experimental results. The fitting process requires that the predicted results of the Rouse model are close to the experimental measurement results. Therefore, the difference ε must be as small as possible. The fitting of the Rouse model evolves into an optimization problem, and the optimization goal is

[0056]

[0057] The difference ε varies with the distribution of polymers. The AI ​​algorithm described in step S3 is used to optimize the objective function of formula (5). Compare the actual ε with the error threshold ε of the AI ​​algorithm. th , such as ε≥ε th , then the normal distribution parameters are calculated by AI algorithm Optimize and update ΔM, return to process b) after updating, and recalculate the Rouse model; if ε<ε th, it means that the Rouse model at this time can accurately describe the experimental results. Complete this step and proceed to the next step.

[0058] Preferably, in step S5, when the molecular weight is calculated using the Rouse model optimized by the AI ​​algorithm, the molecular weight-related data are the weight-average molecular weight, the number-average molecular weight and the molecular weight distribution.

[0059] Furthermore, in step S5, the optimized Rouse model obtained in step S4 is used to simulate the molecular weight distribution of the physical system. Complete weight average molecular weight M w , number average molecular weight M n and molecular weight distribution, i.e., M w / M n Calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 shows the molecular weight calculation process of biopolymer systems based on the Rouse model.

[0061] 2-5 show the relationship between the storage modulus and loss modulus of silk protein ionic liquid solutions with different molecular weights and shear frequency in oscillation mode according to Examples 1-4 of the present invention, respectively.

[0062] FIG6 is the calculation results of different silk protein molecular weights in Examples 1-4 of the present invention. DETAILED DESCRIPTION

[0063] The present invention is described in the following specific embodiments. However, those skilled in the art will appreciate that the following specific embodiments do not limit the scope of protection of the present invention.

[0064] Example 1:

[0065] S1: Sample preparation

[0066] a) Dissolution: mulberry silk protein fibers prepared by degumming with sodium carbonate for 15 minutes were selected, and the mulberry silk protein was mixed with the ionic liquid 1-allyl-3-methylimidazolium chloride, with the concentration of the silk protein being controlled at 20%.

[0067] b) Freeze drying to remove water: freeze the sample in liquid nitrogen to fully solidify the ionic liquid, then place it in a freeze dryer to fully dehumidify it, take it out, and thaw it in a sealed container at room temperature.

[0068] c) Dissolution under reduced pressure: The silk protein ionic liquid mixture prepared in step b is heated in an oil bath at 150°C while stirring. The mixture is then subjected to reduced pressure distillation using an oil pump (vacuum range: -0.01 MPa to -0.5 MPa) to remove any remaining trace moisture and eliminate air bubbles. The mixture is heated until the silk protein is completely dissolved. Heating and stirring are stopped, and the mixture is cooled to obtain a uniform, transparent solution. The resulting silk protein ionic liquid solution is stored in a sealed, dry environment at room temperature until further use.

[0069] S2: Rheological method test:

[0070] 1) Storage modulus and loss modulus test

[0071] Parallel clamps were selected and the test was protected by nitrogen purge through a temperature-controlled hood during the test.

[0072] The linear dynamic elastic test was used, that is, the strain amplitude of the oscillation mode was controlled below 50% to ensure that the strain amplitude of the oscillation mode was within the frequency scanning range (1×10 2 rad / s~30×10 -2 rad / s) and the storage modulus and loss modulus are linear.

[0073] Frequency sweeps were performed at the following temperatures (0°C, 10°C, 20°C, and 30°C) to obtain the storage modulus and loss modulus curves of the samples at different temperatures.

[0074] 2) Viscosity test

[0075] Parallel clamps were selected and the test was protected by nitrogen purge through a temperature-controlled hood during the test.

[0076] Steady-state test experiment: the shear rate is scanned from low to high, the shear rate range is: 10 -5 ~10 5 s -1 . Record the viscosity value of the platform curve.

[0077] S3: Modeling with the Rouse model: Based on commonly used AI model optimization algorithms, an algorithm is established to evaluate the quality of the Rouse model prediction results:

[0078] a) Set the error threshold ε th = 0.01, when the prediction error ε of the Rouse model is less than ε th , it is believed that the Rouse model is accurate enough to truly describe the situation of the experimental polymer system.

[0079] b) Set the initial learning rate α0 = 0.05 to control the iteration speed of the algorithm.

[0080] C) Use the gradient descent method to continuously modify the model parameters until the prediction error is less than 0.01.

[0081] S4: The specific method of establishing the Rouse model to fit the biopolymer system is as follows: a) Assume that the structure of the biopolymer system satisfies the Rouse model and the molecular weight distribution of the system satisfies the normal distribution. Model the experimental system and initialize the average value of the molecular weight distribution. and standard deviation ΔM0.

[0082] b) Use and ΔM0 to construct the normal distribution of molecular weight [Formula (4)], sampling is performed based on this distribution to establish the initial simulated physical system.

[0083]

[0084] c) Based on the molecular weight distribution of the simulated physical system, the relaxation time τ under mode p is calculated according to formula (3): ip .

[0085] d) Considering the contribution of all vibration modes p to the storage modulus G′ and dissipation modulus G″, calculate G′ and G″ of the simulated physical system according to formulas (1) and (2). The density ρ of the polymer in the solution, the zero shear viscosity η0 of the solution, and the viscosity η of the solvent used in the calculation are s This is consistent with the experimental data.

[0086] e) Based on the simulation data obtained from the Rouse theoretical model, the relationships between log G′ and log ω and log G″ and log ω of the simulated physical system can be calculated. The difference ε between the predicted values ​​of log G′ and log G″ from the Rouse model and the experimental values ​​measured in step S2 can be obtained by solving the root mean square formula.

[0087] f) The difference ε between the calculated log G′ and log G″ data describes the degree of agreement between the Rouse model and the experimental results. The fitting process requires that the predicted results of the Rouse model are close to the experimental measurement results. Therefore, the difference ε must be as small as possible. The fitting of the Rouse model evolves into an optimization problem, and the optimization goal is

[0088]

[0089] The difference ε varies with the distribution of polymers. The AI ​​algorithm described in step S3 is used to optimize the objective function of formula (5). Compare the actual ε with the error threshold ε of the AI ​​algorithm. th , such as ε≥ε th , then the normal distribution parameters are calculated by AI algorithm Optimize and update ΔM, return to process b) after updating, and recalculate the Rouse model; if ε<ε th , it means that the Rouse model at this time can accurately describe the experimental results. Complete this step and proceed to the next step.

[0090] S5: Calculate molecular weight related data using the Rouse model optimized by AI algorithm: The optimized Rouse model obtained in step S4 is used to simulate the molecular weight distribution of the physical system. Complete weight average molecular weight M w The number average molecular weight is 266kDa, M n The molecular weight distribution is 253 kDa and 1.05.

[0091] Example 2

[0092] S1: Sample preparation:

[0093] a) Dissolution: mulberry silk protein fibers prepared by degumming with sodium carbonate for 30 minutes were selected, and the mulberry silk protein was mixed with the ionic liquid 1-allyl-3-methylimidazolium chloride, with the concentration of the silk protein being controlled at 20%.

[0094] b) Freeze drying to remove water: freeze the sample in liquid nitrogen to fully solidify the ionic liquid, then place it in a freeze dryer to fully dehumidify it, take it out, and thaw it in a sealed container at room temperature.

[0095] c) Dissolution under reduced pressure: The silk protein ionic liquid mixture prepared in step b is heated in an oil bath at 150°C while stirring. The mixture is then subjected to reduced pressure distillation using an oil pump (vacuum range: -0.01 MPa to -0.5 MPa) to remove any remaining trace moisture and eliminate air bubbles. The mixture is heated until the silk protein is completely dissolved. Heating and stirring are stopped, and the mixture is cooled to obtain a uniform, transparent solution. The resulting silk protein ionic liquid solution is stored in a sealed, dry environment at room temperature until further use.

[0096] S2: Rheological method test: same as step S2 in Example 1.

[0097] S3: Modeling using the rouse model: the same as step S3 in Example 1.

[0098] S4: Establishing a Rouse model to fit the biopolymer system: Same as step S4 in Example 1.

[0099] S5: Calculate molecular weight related data using the Rouse model optimized by AI algorithm: The optimized Rouse model obtained in step S4 is used to simulate the molecular weight distribution of the physical system. Complete weight average molecular weight M w The number average molecular weight is 181kDa, M nThe molecular weight distribution is 97 kDa and 1.87.

[0100] Example 3

[0101] S1: Sample preparation:

[0102] a) Dissolution: Soluble high molecular weight spray-dried mulberry silk protein powder prepared by sodium carbonate degumming for 45 minutes is selected and mixed with the ionic liquid 1-allyl-3-methylimidazolium chloride. The concentration of the silk protein is controlled to be 20%.

[0103] b) Freeze drying to remove water: freeze the sample in liquid nitrogen to fully solidify the ionic liquid, then place it in a freeze dryer to fully dehumidify it, take it out, and thaw it in a sealed container at room temperature.

[0104] c) Dissolution under reduced pressure: The silk protein ionic liquid mixture prepared in step b is heated in an oil bath at 120°C while stirring. The mixture is then subjected to reduced pressure distillation using an oil pump (vacuum range: -0.01 MPa to -0.5 MPa) to remove any remaining trace moisture and eliminate air bubbles. The mixture is heated until the silk protein is completely dissolved. Heating and stirring are stopped, and the mixture is cooled to obtain a uniform, transparent solution. The resulting silk protein ionic liquid solution is stored in a sealed, dry environment at room temperature until further use.

[0105] S2: Rheological method test: same as step S2 in Example 1.

[0106] S3: Modeling using the rouse model: the same as step S3 in Example 1.

[0107] S4: Establishing a Rouse model to fit the biopolymer system: Same as step S4 in Example 1.

[0108] S5: Calculate molecular weight related data using the Rouse model optimized by AI algorithm: The optimized Rouse model obtained in step S4 is used to simulate the molecular weight distribution of the physical system. Complete weight average molecular weight M w The number average molecular weight is 145kDa and the n The molecular weight distribution is 71 kDa and 2.05.

[0109] Example 4

[0110] S1: Sample preparation:

[0111] a) Dissolution: Soluble high molecular weight freeze-dried mulberry silk protein powder prepared by degumming with sodium carbonate for 60 minutes is selected and the mulberry silk protein powder is mixed with the ionic liquid 1-allyl-3-methylimidazolium chloride. The concentration of the silk protein is controlled to be 20%.

[0112] b) Freeze drying to remove water: freeze the sample in liquid nitrogen to fully solidify the ionic liquid, then place it in a freeze dryer to fully dehumidify it, take it out, and thaw it in a sealed container at room temperature.

[0113] c) Dissolution under reduced pressure: The silk protein ionic liquid mixture prepared in step b is heated in an oil bath at 120°C while stirring. The mixture is then subjected to reduced pressure distillation using an oil pump (vacuum range: -0.01 MPa to -0.5 MPa) to remove any remaining trace moisture and eliminate air bubbles. The mixture is heated until the silk protein is completely dissolved. Heating and stirring are stopped, and the mixture is cooled to obtain a uniform, transparent solution. The resulting silk protein ionic liquid solution is stored in a sealed, dry environment at room temperature until further use.

[0114] S2: Rheological method test: same as step S2 in Example 1.

[0115] S3: Modeling using the rouse model: the same as step S3 in Example 1.

[0116] S4: Establishing a Rouse model to fit the biopolymer system: Same as step S4 in Example 1.

[0117] S5: Calculate molecular weight related data using the Rouse model optimized by AI algorithm: The optimized Rouse model obtained in step S4 is used to simulate the molecular weight distribution of the physical system. Complete weight average molecular weight M w The number average molecular weight is 115kDa and the n The molecular weight distribution is 58 kDa and 1.97.

Claims

1. A method for calculating the molecular weight of a biopolymer material based on an AI algorithm, characterized in that, it includes the following steps: S1: Prepare a sample: Take a sample of the biopolymer material to be measured, dissolve it in an ionic liquid to obtain a sample solution to be measured, that is, an ionic liquid solution of the biopolymer material; S2: Sample testing: Place the sample prepared in step S1 on a rheometer for sample testing to calculate the required data; S3: Establish an AI algorithm to evaluate the Rouse model: Based on the commonly used AI model optimization algorithm, establish an algorithm to evaluate the quality of the prediction results of the Rouse model; S4: Establish a Rouse model to fit the biopolymer system; S5: Use the Rouse model optimized by the AI algorithm to calculate the molecular weight.

2. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to claim 1, characterized in that, the specific method for preparing the sample in step S1 is: a) Dissolution: Add the dried biopolymer material sample to the ionic liquid and mix, dissolve at room temperature; b) Dry to remove water; c) Dissolve under reduced pressure. Preferably, in step S1, the biopolymer material includes: silk fibroin, hyaluronic acid, collagen, recombinant collagen, sericin. Preferably, the silk fibroin is selected from mulberry silk fibroin, spider silk fibroin, tussah silk fibroin. Preferably, the ionic liquid in the step S1 is selected from [AMIm]Cl and [HMIm]HSO 4 .

3. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to claim 1 or 2, characterized in that, in the ionic liquid solution of the biopolymer material silk fibroin prepared in step S1, the concentration of silk fibroin is 0.1%-50%, preferably 1%-15%, and more preferably 5%-20%.

4. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to claim 2, characterized in that, the freeze-drying method is selected for drying and water removal in step S1. Specifically, the sample is placed in liquid nitrogen for freezing to fully solidify the ionic liquid, and then placed in a freeze-dryer to fully dehumidify and then taken out, and thawed in a sealed state at room temperature.

5. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to claim 2 or 4, characterized in that, the specific method for dissolving under reduced pressure in the sample preparation in step S1 is: Under stirring, heat the silk fibroin ionic liquid mixture prepared in step S1(b) in an oil bath, and use an oil pump for vacuum distillation to remove possible residual trace moisture in the mixture and eliminate air bubbles, and heat until the biopolymer is completely dissolved. Preferably, under stirring, heat the silk fibroin ionic liquid mixture prepared in step S1(b) in an oil bath at a temperature of 0-180°C, and use an oil pump for vacuum distillation with a vacuum range of -0.01 MPa to -0.5 MPa to remove possible residual trace moisture in the mixture and eliminate air bubbles, and heat until the biopolymer is completely dissolved; stop heating and stirring, and cool down to obtain a uniform and transparent solution. More preferably, store the obtained biopolymer ionic liquid solution at room temperature in a sealed and dry environment for later use.

6. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to any one of claims 1-5, characterized in that, The specific method for sample testing in step S2 is as follows: a) Storage modulus and loss modulus testing: Select parallel clamping plates, and protect the testing by purging nitrogen through the temperature control cover during the testing process; Linear dynamic elasticity testing is adopted, that is, the strain amplitude in the oscillation mode is controlled below 50% to ensure that the storage modulus and loss modulus are linear within the frequency sweep range (1×10 2 rad / s - 30×10 -2 rad / s). Perform frequency scanning at the following different temperatures (0 °C, 10 °C, 20 °C, and 30 °C) to obtain the storage modulus and loss modulus curves of the sample at different temperatures; b) Viscosity testing: Select parallel clamping plates, and protect the testing by purging nitrogen through the temperature control cover during the testing process; Steady-state test experiments were carried out: the shear rate was scanned from low to high, and the shear rate range was: 10 -5 -10 5 s -1 , and the viscosity values of the plateau curve were recorded.

7. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to any one of claims 1-6, characterized in that The specific method for establishing the AI algorithm to evaluate the Rouse model in step S3 is as follows: a) Set a reasonable error threshold ε th , when the prediction error ε of the Rouse model is less than ε th , it is considered that the accuracy of the Rouse model is high enough to truly describe the experimental polymer system; b) Set a reasonable initial learning rate α 0 , to control the iteration speed of the algorithm; c) Continuously correct the model parameters using the gradient descent method until the prediction error is less than 0.

01. Preferably, the AI model optimization algorithms in step S3 include the gradient descent method (Gradient Descent), the conjugate gradient method (Conjugate Descent), Adam, AdamGrad, and RMSProp.

8. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to any one of claims 1-7, characterized in that The error threshold ε in the step S3 th is selected in the range of 0.001 - 0.25, and the learning rate α 0 is selected in the range of 0.0001 - 0.

1.

9. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to any one of claims 1-8, characterized in that The steps for establishing the Rouse model to fit the biopolymer system in step S4 are as follows: a) Assume that the structure of the biopolymer system satisfies the Rouse model and the molecular weight distribution of the system satisfies the normal distribution; model the experimental system and initialize the average value of the molecular weight distribution with standard deviation ΔM 0 ; b) Use and ΔM 0 Construct a normal distribution of molecular weights Sample according to this distribution to establish an initial simulated physical system; c) Calculate the relaxation time τ in mode p according to the molecular weight distribution of the simulated physical system using Equation (3). ip ; d) Consider the contributions of all vibrational modes p to the storage modulus G′ and the loss modulus G″, and calculate G′ and G″ of the simulated physical system according to formulas (1) and (2). Preferably, the density ρ of the polymer in the solution, the zero-shear viscosity η of the solution 0 and the viscosity η of the solvent s are consistent with the experimental data; e) According to the simulated data obtained from the Rouse theoretical model, the relationships between log G′~logω and log G″~logω of the simulated physical system can be calculated. The difference ε between the predicted values log G′ and log G″ of the Rouse model and the measured experimental values in step S2 can be obtained by solving using the root mean square formula; f) The difference ε between the calculated log G′ and log G″ data describes the degree of agreement between the Rouse model and the experimental results. Preferably, the fitting process requires that the prediction results of the Rouse model approach the experimental measurement results; Therefore, the difference ε must be as small as possible, and the fitting of the Rouse model evolves into an optimization problem, and the optimization objective is The difference ε varies with the distribution of the polymer. Using the AI algorithm described in step S3, the objective function of formula (5) is optimized; the actual ε is compared with the error threshold ε of the AI algorithm th , if ε ≥ ε th , then the normal distribution parameters are adjusted through the AI algorithm Optimize and update with ΔM, and after the update, return to process b) and recalculate the Rouse model; if ε < ε th , it indicates that the Rouse model at this time can accurately describe the experimental results.

10. The method for calculating the molecular weight of a biopolymer material based on an AI algorithm according to claim 1, characterized in that In step S5, the molecular weight data are the weight-average molecular weight, the number-average molecular weight, and the molecular weight distribution. Preferably, the method for calculating the molecular weight using the Rouse model optimized by the AI algorithm in step S5 is as follows: using the optimized Rouse model obtained in steps S1 - S4, the molecular weight distribution of the simulated physical system is used Completion of the calculation of the weight-average molecular weight M w , number-average molecular weight M n and molecular weight distribution, i.e., M w / M n .

Images