Protein solubility prediction method
By combining protein pre-training models and primary structural features, a protein solubility prediction model of an automatic machine learning framework is constructed, which solves the problem of low classification and regression accuracy of existing models, and achieves high-accuracy protein solubility prediction and variants filter.
Patent Information
- Application Number
- PCT/CN2024/092583
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-11
- Filing Date
- 2024-05-11
- Publication Date
- 2024-11-14
AI Technical Summary
The existing protein solubility prediction model is difficult to ensure the accuracy of classification and regression problems at the same time, and the accuracy needs to be improved.
The features extracted based on the protein pre-training model are combined with primary structural features, and a model that is compatible with classification and regression is constructed through an automatic machine learning framework, and hyperparameter screening is automatically selected to optimize model performance.
Highly accurate protein solubility prediction is achieved, which can simultaneously predict whether proteins are soluble and the probability of soluble in different variants, which helps to screen high-potential variants in advance, surpassing the performance of existing models.
Smart Images

Figure CN2024092583_14112024_PF_FP_ABST
Abstract
Description
Protein solubility prediction methods
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to Chinese patent application No. 202310533179.6 filed on May 11, 2023, the entire contents of which are incorporated herein by reference. Technical Field
[0003] The present disclosure relates to protein research, and more particularly to methods for predicting protein solubility. Background Art
[0004] Improving protein solubility plays a crucial role in increasing recombinant protein yields, conducting protein functional studies (including structural and functional proteomics), and completing the screening of enzyme engineering precursors. It directly impacts the functional performance and production costs of recombinant proteins. For example, during protein engineering, if the solubility of different protein variants can be accurately predicted in advance based on the protein's amino acid sequence, this can improve screening efficiency, identify high-potential proteins, and effectively reduce the scale of downstream experimental validation. Numerous methods exist to improve protein solubility during production. For example, in Escherichia coli expression systems, methods such as using weak promoters, specific protein tags, or altering signal peptides, denaturation and renaturation with strong denaturants, long-term expression at low temperatures, codon optimization, and expression of molecular chaperones can reduce inclusion bodies and increase effective protein yield (i.e., soluble protein). However, accurately predicting protein solubility in advance remains a crucial issue, which is why models and tools for predicting protein solubility have been continuously developed in recent years (see Table 1).
[0005] The primary sequence of a protein determines its higher-order structure after folding, and the higher-order structure of a protein determines its solubility. Therefore, both primary structure-based sequence features and higher-order structure-based features have been used to construct protein solubility prediction models, including protein sequence length, amino acid residue composition (e.g., the number of charged amino acid residues, the number of hydrophobic / hydrophilic amino acid residues, k-mer characteristics), protein secondary structure (e.g., the ratio of alpha helices to beta sheets), and protein tertiary structure (e.g., the angle between the alpha carbon atoms of two amino acids). (For detailed descriptions of these features, see the papers listed in Table 1.) Currently, statistical, machine learning, and deep learning methods used for protein solubility prediction include: Gaussian discriminant analysis models, linear regression, support vector machines, logistic regression, linear programming, random forests, gradient boosting machines, naive Bayes, neural networks, convolutional neural networks, dual-channel convolutional neural networks, bidirectional threshold recurrent unit neural networks, and graph neural networks. (For detailed descriptions of the models used, see the main text of the papers listed in Table 1.)
[0006] Table 1: Currently known methods for predicting protein solubility
[0007] However, these current protein solubility prediction models are not compatible with classification and regression problems, and their accuracy needs to be improved.
[0008] Summary of the Invention
[0009] Given that existing protein solubility classification prediction models struggle to simultaneously ensure high classification and regression accuracy, this paper proposes a protein solubility prediction method that combines features extracted from a protein pre-training model with protein primary structure features. Using an automated machine learning framework, this method constructs and trains a model that is compatible with both classification problems (predicting whether a protein is soluble) and regression problems (predicting the probability of different proteins being soluble, predicting the impact of amino acid mutations on solubility, and facilitating early screening of high-potential variants). Furthermore, automated hyperparameter screening within the automated machine learning framework can yield a hyperparameter combination that optimizes model performance.
[0010] According to a first aspect of the present disclosure, a method for predicting protein solubility is provided. The method may include: extracting features based on a protein sequence using a pre-trained protein model; extracting primary structural features of the protein based on the protein sequence; concatenating the features extracted by the pre-trained protein model with the primary structural features of the protein to obtain a feature vector for the protein; and inputting the feature vector into a protein solubility prediction model to obtain a prediction result for protein solubility.
[0011] In the method according to the first aspect of the present disclosure, preferably, the protein pre-training model can be implemented by a combination of one or more of the following models: ESM-1b, UniRep, ProteinBert, TAPE, ProtGPT2, ProtTXL, ProtBert, ProtXLNet, ProtAlbert, ProtElectra, ProtT5-XL-BFD, ProtT5-XL-UniRef50 and ProtT5-XXL; more preferably, the protein pre-training model is ProtT5-XL-BFD or ProtT5-XL-UniRef50; more preferably, the protein pre-training model is ProtT5-XL-UniRef50.
[0012] In the method according to the first aspect of the present disclosure, preferably, the protein sequence can be used as input, and the ProtT5-XL-UniRef50 pre-trained model can be used to extract the embedding layer vector output by the encoder, wherein each amino acid corresponds to a 1024-dimensional vector, and each sequence corresponds to an L×1024-dimensional feature matrix, where L is the length of the amino acid sequence of the protein; the above feature matrix is averaged by column to obtain a 1024-dimensional feature vector.
[0013] Preferably, the Sentence-T5 Encoder-only mean method can be used to average the values of each dimension of the embedding layer feature values of all amino acids output by the encoder by column to obtain a 1024-dimensional feature vector.
[0014] In the method according to the first aspect of the present disclosure, preferably, the protein sequence can be used as input to extract the primary structural features of the protein using the iFeature tool.
[0015] Preferably, the primary structural features of the protein can be selected from AAC (Amino Acid Composition), DPC (Di-Peptide Composition), DDE (Dipeptide Deviation from Expected Mean), TPC (Tri-Peptide Composition), CKSAAP (Composition of k-spaced Amino Acid Pairs), EAAC (Enhanced Amino Acid Composition), GAAC (Grouped Amino Acid Composition), CKSAAGP (Composition of k-Spaced Amino Acid Group Pairs), GDPC (Grouped Di-Peptide Composition), GTPC (Grouped Tri-Peptide Composition), Moran (Moran correlation), Geary (Geary correlation), NMBroto (Normalized Moreau-BrotoAutocorrelation, Normalized Moreau-Broto Autocorrelation), CTDC (Composition, Transition and Distribution: Composition, Composition, Transition and Distribution: Composition), CTDT (Composition, Transition and Distribution: Transition, Composition, Transition and Distribution: Transition), CTDD (Composition,One or more of: Transition and Distribution: Distribution, Composition, Transition and Distribution: Distribution), CTriad (Conjoint Triad), KSCTriad (k-Spaced Conjoint Triad), SOCNumber (Sequence-Order-Coupling Number), QSOrder (Quasi-sequence-order), PAAC (Pseudo-Amino Acid Composition), APAAC (Amphiphilic Pseudo-Amino Acid Composition), KNNprotein (K-Nearest Neighbor for proteins), PSSM (PSSM profile), AAINDEX (AAindex), and BLOSUM62 (blocks substitution matrix, an amino acid substitution scoring matrix used in bioinformatics for sequence alignment).
[0016] More preferably, the primary structural features of the protein extracted using the iFeature tool include 11 major categories, namely AAC, GAAC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, PAAC and APAAC.
[0017] In the method according to the first aspect of the present disclosure, preferably, the protein solubility prediction model can be constructed by the following steps: obtaining a data set, the data set including multiple training samples, each training sample including a sample protein sequence and its solubility data; extracting features based on the sample protein sequence using a protein pre-training model; extracting the primary structural features of the protein based on the sample protein sequence; splicing the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the sample protein; using the feature vector of the sample protein as input data and the solubility data of the sample protein as output data, and obtaining a final protein solubility prediction model by training a machine learning model.
[0018] In the method according to the first aspect of the present disclosure, preferably, the protein solubility prediction model can be constructed and trained based on an automatic machine learning framework using one or more machine learning models as candidate models. In some embodiments, the machine learning model can be a deep learning model. In some embodiments, the automatic machine learning framework can be selected from AutoGluon (https: / / auto.gluon.ai / stable / index.html), Auto-Sklearn (https: / / automl.github.io / auto-sklearn / master / ), TPOT (http: / / epistasislab.github.io / tpot / ), Hyperopt Sklearn (https: / / github.com / hyperopt / hyperopt-sklearn), Auto-Keras (https: / / autokeras.com / ), H2O AutoML (https: / / docs.h2o.ai / h2o / latest-stable / h2o-docs / automl.html), TransmogrifAI (https: / / transmogrif.ai / ). In some specific embodiments, the automatic learning framework can be AutoGluon.
[0019] In some embodiments, the machine learning model is a deep learning model, which can be selected from NN_Torch, FASTAI, LGBModel, RFModel, KNNModel, etc. In some embodiments, the deep learning model can be one or more. In a preferred embodiment, the deep learning model can be NN_Torch and / or FASTAI. In a specific embodiment, the deep learning model can be NN_Torch and FASTAI.
[0020] In the method according to the first aspect of the present disclosure, the machine learning model can be automatically selected by an automated learning framework, or a list of machine learning models and a screening range can be manually set and then selected by the automated learning framework. Preferably, AutoGluon can be used as the automated machine learning framework, and the deep learning models NN_Torch and FASTAI can be selected as candidate models to construct and train the protein solubility prediction model.
[0021] In the method according to the first aspect of the present disclosure, preferably, other features of the protein can also be extracted based on the protein sequence; the features extracted by the protein pre-training model are spliced with the primary structure features of the protein and other features of the protein to obtain a feature vector of the spliced protein.
[0022] Preferably, the other characteristics of the protein are selected from molecular weight, aromaticity, instability index, flexibility, isoelectric point, molar absorption coefficient, grand average of hydropathicity, the proportion of helix / turn / sheet in the secondary structure of the protein, SSEC (Secondary Structure Elements Content), SSEB (Secondary Structure Elements Binary), Disorder, DisorderC (Disorder Content), DisorderB (Disorder Binary), ASA (Accessible Solvent Accessibility) and TA (Torsion Angle), sequence length, ZSCALE (Z-scale) and 48 PseKRAAC (48 pseudo K-tuple reduced amino acids composition, one or more of 48 pseudo K-tuple reduced amino acid composition); more preferably, other characteristics of the protein include molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorptivity, overall average hydrophilicity and the proportion of helices / turns / folds in the secondary structure of the protein.
[0023] According to a second aspect of the present disclosure, a method for constructing a protein solubility prediction model is provided. The method may include: obtaining a data set, the data set including multiple training samples, each training sample including a sample protein sequence and its solubility data; extracting features based on the sample protein sequence using a protein pre-training model; extracting primary structural features of the protein based on the sample protein sequence; concatenating the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the sample protein; using the feature vector of the sample protein as input data and the solubility data of the sample protein as output data, to train a machine learning model to obtain a final protein solubility prediction model.
[0024] The method for constructing a protein solubility prediction model in the first aspect of the present disclosure is suitable for the second aspect of the present disclosure.
[0025] According to a third aspect of the present disclosure, a system for predicting protein solubility is provided. The system may include: an acquisition module for acquiring a sample to be predicted, wherein the sample to be predicted includes a protein sequence of the sample to be predicted; a processing module for obtaining a prediction result of the protein solubility of the sample to be predicted by inputting the protein sequence of the sample to be predicted, wherein the processing module further includes the following submodules: a first feature extraction submodule for extracting features based on the protein sequence of the sample to be predicted using a protein pre-training model; a second feature extraction submodule for extracting primary structural features of the protein based on the protein sequence of the sample to be predicted; a feature splicing submodule for splicing the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the protein of the sample to be predicted; and a model prediction submodule for inputting the feature vector of the protein of the sample to be predicted into a protein solubility prediction model to obtain a prediction result of the protein solubility of the sample to be predicted.
[0026] The system for predicting protein solubility according to the third aspect of the present disclosure can be implemented by a computer. Preferably, the following operations can be implemented by executing a computer program: obtaining a sample to be predicted, wherein the sample to be predicted includes a protein sequence of the sample to be predicted; obtaining a prediction result of the protein solubility of the sample to be predicted by inputting the protein sequence of the sample to be predicted, wherein the operation further includes the following sub-operations: extracting features using a protein pre-training model based on the protein sequence of the sample to be predicted; extracting primary structural features of the protein based on the protein sequence of the sample to be predicted; splicing the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the protein of the sample to be predicted; and inputting the feature vector of the protein of the sample to be predicted into the protein solubility prediction model to obtain a prediction result of the protein solubility of the sample to be predicted.
[0027] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is provided for storing a computer program. The computer program includes instructions. When executed by a processor of an electronic device, the instructions cause the electronic device to implement the protein solubility prediction method according to the first aspect of the present disclosure or the method for constructing a protein solubility prediction model according to the second aspect of the present disclosure.
[0028] According to a fifth aspect of the present disclosure, a computer system is provided. The computer system includes a processor, a memory, and a computer program. The computer program is stored in the memory and is configured to be executed by the processor. The computer program includes instructions for implementing the protein solubility prediction method according to the first aspect of the present disclosure or the method for constructing a protein solubility prediction model according to the second aspect of the present disclosure.
[0029] According to a sixth aspect of the present disclosure, a protein solubility prediction model constructed by the method of the second aspect of the present disclosure is provided. Further provided is the application of the protein solubility prediction model in predicting protein solubility. For example, the protein solubility prediction model can be applied to predict the solubility of any protein sample of a known sequence (e.g., a mutant of a known protein), and the predicted result can be a probability of solubility or a yes or no result of solubility.
[0030] The comprehensive performance of the method disclosed in the present invention in predicting protein solubility surpasses the NetSolP method, which is currently ranked first in classification performance.
[0031] In addition, the method disclosed in the present invention has a high accuracy rate for predicting protein solubility. It is applicable to both the classification prediction problem of predicting whether a protein is soluble and the regression problem of predicting the effect of amino acid mutations on solubility. It can be used to predict the solubility of recombinant protein expression and to screen enzyme engineering mutants based on solubility potential. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] The present disclosure will be more fully understood from the following detailed description taken in conjunction with the accompanying drawings, in which like elements are numbered in a similar manner, and in which:
[0033] FIG1 is a flowchart of a method for predicting protein solubility according to an embodiment of the present disclosure.
[0034] FIG2 is a schematic diagram illustrating the association between a protein solubility prediction method and a prediction model construction method according to an embodiment of the present disclosure.
[0035] FIG3 is a more detailed flowchart of a protein sequence-based feature extraction method according to an embodiment of the present disclosure.
[0036] FIG4 is a more detailed flowchart of the method for constructing a protein solubility prediction model used in the present disclosure.
[0037] FIG5 shows a schematic block diagram of a system for predicting protein solubility according to the present disclosure. DETAILED DESCRIPTION
[0038] Unless otherwise defined, technical and scientific terms used in the present disclosure have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs.
[0039] The technical solutions of the present disclosure are further described in detail below through examples and in conjunction with the accompanying drawings. Unless otherwise stated, the methods and materials of the embodiments described below are conventional products that can be purchased on the market. Those skilled in the art will understand that the methods and materials described below are merely exemplary and should not be construed as limiting the scope of the present disclosure.
[0040] Predicting protein solubility can be divided into two categories. One is a binary or multi-classification problem, which involves predicting whether a protein is soluble or insoluble, or whether it is easy or difficult to solubilize. The other is a regression problem, which involves predicting the probability of protein solubility. Both types of predictions rely on the collection of solubility data from expression experiments. Commonly used protein solubility datasets are listed below (see Table 2).
[0041] Table 2: List of protein solubility experimental datasets
[0042] Because classification cannot be simply predicted based on regression, existing protein solubility classification prediction models struggle to achieve high accuracy for both classification and regression. Consequently, existing protein solubility classification prediction methods have limited compatibility with regression prediction. For example, the NetSolP method achieved an accuracy of 72.8% for classification on the independent test set NESG, but only 66.1% for regression on the independent test set CamSol.
[0043] The present disclosure provides a method for predicting protein solubility. By constructing a prediction model that is compatible with classification and regression problems, it can not only predict whether a protein is soluble, but also predict the probability of different proteins being soluble, which helps to screen high-potential variants in advance.
[0044] From the overall perspective of the prediction method, FIG1 shows a flowchart of the protein solubility prediction method according to an embodiment of the present disclosure.
[0045] As shown in FIG1 , the protein solubility prediction method 100 according to an embodiment of the present disclosure begins at step S110 , where features are extracted based on the protein sequence using a protein pre-training model.
[0046] Next, in step S120 , still based on the protein sequence, the primary structural features of the protein are extracted.
[0047] In step S130 , the features extracted by the protein pre-training model in step S110 are concatenated with the primary structure features of the protein extracted in step S120 to obtain a feature vector of the protein.
[0048] Finally, in step S140, the feature vector of the protein obtained in step S130 is input into a protein solubility prediction model to obtain a prediction result of protein solubility.
[0049] In a preferred embodiment of the present disclosure, the prediction result of protein solubility is: the protein is classified as soluble or insoluble; in another preferred embodiment of the present disclosure, the prediction result of protein solubility is the probability that the protein is soluble.
[0050] As mentioned above, the method disclosed in the present invention combines the features extracted based on the protein pre-training model with the primary structure features of the protein, and uses an automatic learning framework to construct a prediction model that is compatible with classification problems (predicting whether a protein is soluble) and regression problems (predicting the probability of different proteins being soluble).
[0051] Figure 2 shows a schematic diagram 200 of the association between the protein solubility prediction method and the prediction model construction method according to an embodiment of the present disclosure. In the schematic diagram 200 shown in Figure 2, the left side of the dotted line shows the model construction method, and the right side of the dotted line shows the method of using the model for prediction.
[0052] The process on the right side of the dotted line in Figure 2 is actually the protein solubility prediction method according to the embodiment of the present disclosure shown in Figure 1. The process on the left side of the dotted line in Figure 2 will be described in detail below (for example, Figure 4 and its corresponding text description). As can be seen from Figure 2, both model construction and actual prediction process require the use of a protein feature extraction method, which will be described in detail below (for example, Figure 3 and its corresponding text description). For training samples, both the protein sequence of the training sample and the protein solubility data of the training sample are included. As described later, the protein solubility data may include whether the protein is soluble or insoluble. Therefore, in the model construction process, the protein sequence in the training sample is used to extract features. The model based on the automatic machine learning framework (for example, using the AutoGluon framework) is fully trained using the extracted protein features as input data and the protein solubility data in the training sample as output data, thereby obtaining the final protein solubility prediction model based on the automatic machine learning framework. For the constructed and trained prediction model, it can be put into the actual work of predicting the protein solubility of the predicted sample, that is, the process on the right side of the dotted line in Figure 2.
[0053] The following describes in detail the protein feature extraction method according to the embodiments of the present disclosure and the more detailed processes of the method for constructing a protein solubility prediction model based on an automatic machine learning framework.
[0054] FIG3 is a more detailed flowchart of a protein feature extraction method 300 according to an embodiment of the present disclosure.
[0055] As shown in step S110 of the prediction method of FIG1 , in the protein feature extraction step, first, based on the protein sequence, a protein pre-training model is used to extract features.
[0056] The protein pre-training model described here can be implemented by a combination of one or more of the following models: ESM-1b, UniReP, ProteinBert, TAPE, ProtGPT2, ProtTXL, ProtBert, ProtXLNet, ProtAlbert, ProtElectra, ProtT5-XL-BFD, ProtT5-XL-UniRef50 and ProtT5-XXL.
[0057] Preferably, ProtT5-XL-BFD or ProtT5-XL-UniRef50 can be used as the protein pre-training model.
[0058] In a preferred embodiment of the present disclosure, the protein sequence pre-training model can be a ProtT5-XL-UniRef50 pre-training model. ProtT5-XL-UniRef50 is a protein sequence pre-training model using a masked language modeling (MLM) objective. Reference may be made to A.Elnaggar et al., “ProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning”, in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.44, no.10, pp.7112-7127, 1Oct.2022, doi:10.1109 / TPAMI.2021.3095381. By reference herein, the entire contents of the above-mentioned document are incorporated into the present disclosure to make it a part of the contents of the present disclosure. In addition, the source code of the ProtT5-XL-UniRef50 model and its more detailed application can be found at https: / / github.com / agemagician / ProtTrans, where the model was also first published. ProtT5-XL-UniRef50 is based on the t5-3b model and is pre-trained on a large number of protein sequences in a self-supervised manner. This model can be used for protein feature extraction and preferably uses features extracted from the encoder.
[0059] As shown in Figure 3, the protein feature extraction method 300 according to an embodiment of the present disclosure can start at step S310. That is, as described above, according to a preferred embodiment of the present disclosure, in step S310, the protein sequence of the sample to be predicted is used as input, and the ProtT5-XL-UniRef50 pre-training model is used to extract the embedding layer vector output by the encoder, wherein each amino acid corresponds to a 1024-dimensional vector, and each sequence corresponds to an L×1024-dimensional feature matrix, where L is the length of the amino acid sequence of the protein.
[0060] Although the ProtT5-XL-UniRef50 pre-training model is used to extract the feature matrix of protein sequences in the preferred embodiment of the present disclosure, those skilled in the art are fully motivated to adopt another or newly developed protein pre-training model to replace the ProtT5-XL-UniRef50 pre-training model used in the present disclosure when the input and output are essentially unchanged and the functions are similar and the effects are similar or better. Those skilled in the art should understand that even after such a pre-training model is replaced, the replaced solution still falls within the scope of protection claimed in the present disclosure.
[0061] Next, the obtained feature matrix can be compressed. Specifically, the feature matrix can be averaged by column to obtain a 1024-dimensional feature vector. More specifically, as shown in Figure 3, in step S320, the feature matrix obtained in step 310 is compressed. That is, using the Sentence-T5 Encoder-only mean processing method, the embedding layer feature values of all amino acids output by the encoder are averaged by column to obtain a 1024-dimensional feature vector. Here, the embedding layer feature values of the padding code are not included in the calculation, thereby converting the embedding layer feature values of all amino acids constituting a single protein into 1024-dimensional feature values of that protein. The Sentence-T5 encoder-only mean is one way to process the features output by the encoder in Sentence-T5. For more information about the Sentence-T5 encoder-only mean, see Jianmo Ni, Gustavo Hernandez Abrego, Noah Constant, Ji Ma, Keith Hall, Daniel Cer, and Yinfei Yang, “Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Models”, in Findings of the Association for Computational Linguistics: ACL 2022, pages 1864–1874, Dublin, Ireland; Association for Computational Linguistics; doi:10.18653 / v1 / 2022.findings-acl.146. The entire contents of the above document are hereby incorporated into the present disclosure by reference, making it a part of the content of the present disclosure.
[0062] Those skilled in the art should note that although Sentence-T5 Encoder-only mean is a disclosed technology, its application in the method of the present disclosure is novel because the application field involved in the present disclosure is completely different from the application field involved in the original disclosure of Sentence-T5 Encoder-only mean. The present disclosure uses the Sentence-T5 Encoder-only mean method to compress the extracted feature matrix, so that the extracted feature vector integrates the protein features at a deeper level, which is helpful for the subsequent construction of a prediction model with better comprehensive performance based on this feature, and the final protein solubility prediction result is more accurate.
[0063] According to step S120 of FIG. 1 , next, the primary structural features of the protein are extracted based on the protein sequence.
[0064] In a preferred embodiment of the present disclosure, more specifically, the iFeature tool is used to extract the primary structural features of the protein using the protein sequence as input. As shown in Figure 3, in step S330, the iFeature tool is used to extract the primary structural features of the protein using the protein sequence as input.
[0065] For the iFeature tool, see Zhen Chen, Pei Zhao, Fuyi Li, André Leier, Tatiana T Marquez-Lago, Yanan Wang, Geoffrey I Webb, A Ian Smith, Roger J Daly, Kuo-Chen Chou, Jiangning Song, “iFeature: a Python package and web server for features extraction and selection from protein and peptide sequences”, Bioinformatics, Volume 34, Issue 14, July 2018, Pages 2499–2502, https: / / doi.org / 10.1093 / bioinformatics / bty140. The entire contents of the above-mentioned documents are hereby incorporated into the present disclosure by reference, making them a part of the content of the present disclosure.
[0066] Although the iFeature tool is used to extract protein primary structural features in the preferred embodiments of this disclosure, those skilled in the art are fully motivated to replace the iFeature tool used in this disclosure with another or newly developed extraction tool if the input and output data remain essentially unchanged and the method functions similarly and achieves comparable or better results. Those skilled in the art should understand that even with such an extraction tool replacement, the replacement solution would still fall within the scope of protection claimed in this disclosure.
[0067] In some embodiments of the present disclosure, the primary structural features of proteins extracted using the iFeature tool may include AAC (Amino Acid Composition), DPC (Di-Peptide Composition), DDE (Dipeptide Deviation from Expected Mean), TPC (Tri-Peptide Composition), CKSAAP (Composition of k-spaced Amino Acid Pairs), EAAC (Enhanced Amino Acid Composition), GAAC (Grouped Amino Acid Composition), CKSAAGP (Composition of k-Spaced Amino Acid Group Pairs), GDPC (Grouped Di-Peptide Composition), GTPC (Grouped Tri-Peptide Composition), Moran (Moran correlation), Geary (Geary correlation), NMBroto (Normalized Moreau-BrotoAutocorrelation, Normalized Moreau-Broto Autocorrelation), CTDC (Composition, Transition and Distribution: Composition, Composition, Transition and Distribution: Composition), CTDT (Composition, Transition and Distribution: Transition, Composition, Transition and Distribution: Transition), CTDD (Composition,One or more of: Transition and Distribution: Distribution, Composition, Transition and Distribution: Distribution), CTriad (Conjoint Triad), KSCTriad (k-Spaced Conjoint Triad), SOCNumber (Sequence-Order-Coupling Number), QSOrder (Quasi-sequence-order), PAAC (Pseudo-Amino Acid Composition), APAAC (Amphiphilic Pseudo-Amino Acid Composition), KNNprotein (K-Nearest Neighbor for proteins), PSSM (PSSM profile), AAINDEX (AAindex), and BLOSUM62 (BLOSUM62).
[0068] In a preferred embodiment of the present disclosure, the primary structural features of proteins extracted using the iFeature tool include 11 major categories: AAC, GAAC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, PAAC and APAAC.
[0069] In addition, these features can be normalized. The parameters of the standard normalization will be saved and used to process similar features of the input sequence during the prediction process, that is, the standard normalization parameters of the training set are saved for processing the same features of the independent test set.
[0070] According to step S130 of FIG. 1 , the features extracted by the protein pre-training model are then concatenated with the primary structural features of the protein to obtain a feature vector of the protein.
[0071] As shown in FIG3 , in step S340 , the feature vector obtained in step S320 is concatenated with the primary structural feature of the protein obtained in step S330 to obtain a feature vector of the protein.
[0072] Those skilled in the art should note that the method disclosed herein creatively combines features extracted by a pre-trained protein model with primary protein structural features extracted, for example, using the iFeature tool, to form protein features for use as input in a prediction model. This approach is not only new but, as described below, has also achieved promising results in predicting protein solubility.
[0073] In addition, in an alternative embodiment, in addition to the primary structural features of the protein, other features of the protein can be extracted based on the protein sequence, and the other features include molecular features, secondary structure and / or higher-order structural features of the protein sequence, thereby further optimizing the model.
[0074] In some specific embodiments, the iFeature tool can be used to extract other features of the protein, for example, the molecular features of the sequence can be selected from one or more of the following: sequence length, ZSCALE (Z-scale) and 48PseKRAAC (48pseudo K-tuple reduced amino acids composition); the secondary structure features can be selected from one or more of the following: SSEC (Secondary Structure Elements Content), SSEB (Secondary Structure Elements Binary); the high-level structure features can be selected from one or more of the following: Disorder, DisorderC (DisorderC), DisorderB (Disorder Binary), ASA (Accessible Solvent accessibility) and TA (Torsion angle).
[0075] In other specific embodiments, other features can also be extracted using the Bio.SeqUtils.ProtParam module of Biopython, for example, other features of the protein sequence are selected from one or more of the following: molecular weight, aromaticity, instability index, flexibility, isoelectric point, molar absorption coefficient and grand average of hydrophilicity, and the secondary structure features include the proportion of helix / turn / sheet in the secondary structure of the protein.
[0076] In some preferred embodiments, other features of the protein include molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorptivity, overall average hydrophilicity and the proportion of helices / turns / folds in the protein secondary structure extracted using the Bio.SeqUtils.ProtParam module of Biopython.
[0077] For Biopython, see Peter JACock, Tiago Antao, Jeffrey T. Chang, Brad A. Chapman, Cymon J. Cox, Andrew Dalke, Iddo Friedberg, Thomas Hamelryck, Frank Kauff, Bartek Wilczynski, Michiel JL de Hoon, Biopython: freely available Python tools for computational molecular biology and bioinformatics, Bioinformatics, Volume 25, Issue 11, June 2009, Pages 1422–1423, https: / / doi.org / 10.1093 / bioinformatics / btp163. The entire contents of the above-mentioned documents are hereby incorporated into the present disclosure by reference, making them a part of the contents of the present disclosure.
[0078] Although the Bio.SeqUtils.ProtParam module of Biopython is used to extract other protein features in the preferred embodiment of the present disclosure, if the input and output data do not change in essence and the method functions similarly and the effect is similar or better, those skilled in the art are fully motivated to use other or newly developed extraction tools to replace the Biopython tools used in the present disclosure. Those skilled in the art should understand that even after such replacement of the extraction tool, the replaced solution still falls within the scope of protection claimed in the present disclosure.
[0079] In a specific embodiment of the present disclosure, the iFeature tool is used to extract the primary structural features of the protein, including AAC, GAAC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, PAAC and APAAC; at the same time, the Bio.SeqUtils.ProtParam module of Biopython is used to extract other features of the protein, including molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorptivity, total average hydrophilicity and the proportion of helices / turns / folds in the secondary structure of the protein, a total of 1502 features.
[0080] In this way, the features extracted by the protein pre-training model are concatenated with the primary structural features and other protein features to obtain a concatenated protein feature vector. Specifically, the 1024-dimensional feature vector extracted by the protein pre-training model is concatenated with the primary structural features and other features of the 1502 proteins extracted to obtain a 2526-dimensional feature vector for each protein.
[0081] Through the above series of steps S310-S340, according to the method disclosed in the present invention, based on the protein sequence of the sample to be predicted, (1) protein feature vectors can be extracted through the protein pre-training model, (2) primary structural features of the protein can be extracted through the iFeature tool, and (3) other protein features can be extracted through the Bio.SeqUtils.ProtParam module of Biopython, and then the three types of features can be spliced. Thus, the spliced features can be used as input to predict the protein solubility of the sample to be predicted through the protein solubility prediction model, and the predicted protein solubility result can be output.
[0082] It is understandable that this embodiment does not impose any restrictions on the order of splicing the three types of features, namely, features extracted by the protein pre-training model, primary structural features of the protein, and other features of the protein. It is only necessary to ensure that the splicing order of the three types of features is consistent during model construction and actual prediction. For example, during the model construction process, the features extracted by the protein pre-training model, primary structural features of the protein, and other features of the protein are spliced in this order to obtain the feature vector of the protein; then, during the actual prediction process, the features extracted by the protein pre-training model, primary structural features of the protein, and other features of the protein are spliced in this order to obtain the feature vector of the protein of the sample to be predicted.
[0083] In some embodiments, the method may further include performing dimensionality reduction on the feature vectors of the spliced proteins. This dimensionality reduction process can further extract features from the protein mutation vectors, maximizing the extraction of relevant and effective features, simplifying the prediction model, and reducing computational complexity and time, thereby increasing efficiency in subsequent steps. For example, a principal component analysis (PCA) model can be used to reduce the dimensionality of the spliced protein feature vectors.
[0084] On the other hand, as shown in FIG2 , the protein features of the training samples can also be extracted through the above series of steps, so that the extracted protein features are used as input data and input into the prediction model constructed by the automatic machine learning framework (such as the AutoGluon framework), and the protein solubility data in the training samples are used as output data to train the protein solubility prediction model based on the automatic machine learning framework, so as to finally determine the model parameters, thereby constructing a protein solubility prediction model with relatively satisfactory accuracy.
[0085] FIG4 is a more detailed flowchart of a method 400 for constructing a protein solubility prediction model used in the present disclosure.
[0086] Those skilled in the art will appreciate that, as shown in Figure 4, in step S410, a data set comprising multiple training samples should first be acquired. Specifically, each of the multiple training samples in the data set comprises a sample protein sequence and its solubility data.
[0087] In a preferred embodiment of the present disclosure, the PSI:Biology dataset was selected as the training set to construct a binary classification model. The dataset source is listed in Table 2. Protein sequences and solubility data were selected from the dataset, with the solubility data divided into soluble and insoluble categories. This served as the training dataset. The probability values output by the classification model (e.g., the probability of a protein being classified as soluble) were used for the regression problem.
[0088] Then, as previously mentioned, protein feature extraction is required based on the sample protein sequences in the dataset. As shown in step S420 in Figure 4 , feature extraction is performed on all training sample proteins according to the method in Figure 3 . Specifically, features are first extracted based on the sample protein sequences using a pre-trained protein model. Then, based on the sample protein sequences, the primary structural features of the proteins are extracted. Finally, the features extracted using the pre-trained protein model are combined with the primary structural features of the proteins to obtain a feature vector for the sample proteins.
[0089] Furthermore, in an alternative embodiment, in addition to the primary structural features of the sample protein, other features of the sample protein can also be extracted based on the sample protein sequence. Thus, the features extracted by the protein pre-training model are concatenated with the extracted primary structural features and other extracted protein features to obtain a concatenated feature vector for the sample protein.
[0090] Next, the feature vector of the spliced sample protein is used as input data, and the solubility data of the sample protein is used as output data. The machine learning model is trained to obtain the final protein solubility prediction model.
[0091] Specifically, the protein solubility prediction model can be constructed and trained based on an automatic machine learning (eg, AutoML) framework, using one or more machine learning (eg, deep learning models) as candidate models.
[0092] Those skilled in the art will appreciate that implementing a machine learning model involves significant complexity and learning. It also requires extensive domain expertise to generate and compare multiple models before selecting the optimal model to achieve our ultimate goal. Automated machine learning can streamline this process, automating the entire machine learning pipeline with minimal human intervention. Furthermore, automated machine learning significantly accelerates the training of small and medium-sized datasets compared to larger datasets.
[0093] AutoGluon is a popular automatic machine learning framework. This open-source AutoML toolkit, developed by AWS, helps achieve strong predictive performance across a variety of machine learning and deep learning models.
[0094] AutoGluon (see doi:10.48550 / arXiv.2003.06505, the entire contents of which are incorporated herein by reference as a part of the present disclosure) has achieved excellent performance in the evaluation of different automatic machine (deep) learning frameworks (see doi:10.48550 / arXiv.2111.02705) due to its powerful feature engineering processing capabilities, automatic model selection and automatic combination and layer stacking technology, and automatic hyperparameter search technology, and can effectively avoid overfitting problems. It has been selected as the model automatic learning framework in the embodiments of the present disclosure.
[0095] As shown in FIG4 , in step S430 , AutoGluon is used as an automatic machine learning framework, and the protein solubility prediction model is constructed and trained based on the deep learning models NN_Torch and FASTAI as candidate models.
[0096] More specifically, AutoGluon selects the deep learning models NN_Torch and FASTAI as candidate models, turns on the Bayesian automatic hyperparameter selection strategy, and searches ten times. The maximum number of model layers stacked is three, that is, the actual model parameters can be set as follows:
[0097] problem_type="binary"
[0098] eval_metric = 'mcc'
[0099] presets = 'best_quality'
[0100] num_bag_folds=5
[0101] num_bag_sets=5
[0102] num_stack_levels=3
[0103] excluded_model_types=["GBM","CAT","XGB","RF","XT","KNN","LR"]
[0104] hyperparameter_tune_kwargs='bayes'
[0105] hyperparameters={"FASTAI":{},"NN_TORCH":{}}.
[0106] In the process of building the model, five-fold cross-validation was used (i.e., 80% of the data in the PSI:Biology dataset was used as the training set and 20% of the data was used as the validation set), whether the training set protein was soluble was used as the classification value, and MCC was used as the optimization indicator for evaluating the performance of the prediction model.
[0107] MCC (Matthews Correlation Coefficient) is an indicator commonly used to evaluate the predictive performance of classification models. It is recommended by the literature as an optimization indicator for binary classification models, especially when the number of sample categories is unbalanced. For example, see Chicco, D., Jurman, G. "The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation", BMC Genomics 21, 6 (2020); https: / / doi.org / 10.1186 / s12864-019-6413-7. Therefore, in a preferred embodiment of the present disclosure, MCC is selected as the optimization indicator (as a performance indicator for evaluating the classification model) to improve the generalization ability of the model.
[0108] As previously described, during the model construction process, the feature vectors of the spliced sample proteins are used as input data, and the solubility data of the sample proteins (i.e., soluble or insoluble) is used as the output data of the prediction model to train the model. As the model parameters are continuously optimized, the output data of the prediction model will be consistent with the solubility data of the sample proteins.
[0109] Those skilled in the art will appreciate that the prediction model constructed by, for example, the method of FIG. 4 can be used in the prediction method shown in FIG. As previously mentioned, although the protein solubility data of the training samples includes whether the protein is soluble during model construction and training, in the actual prediction process using the prediction model, the prediction result of protein solubility can be the probability of protein solubility.
[0110] There are three independent test sets disclosed in this paper, among which the NESG dataset is used as an independent test set for classification problems, and the CamSol dataset and eSol dataset are used as independent test sets for regression problems.
[0111] Those skilled in the art should understand that in the embodiments of the present disclosure, the same steps as shown in Figure 3 are used to extract the features of proteins in each independent test set, and the features are used as input to the model to output the prediction results of protein solubility.
[0112] The results of the training set and independent test set are shown in Tables 3 to 6 below. The comparison data in Tables 3 to 5 are all from: Vineet Thumuluri, Hannah-Marie Martiny, Jose J Almagro Armenteros, Jesper Salomon, Henrik Nielsen, Alexander Rosenberg Johansen, "NetSolP: predicting protein solubility in Escherichia coli using language models", Bioinformatics, Volume 38, Issue 4, February 2022, Pages 941–946, https: / / doi.org / 10.1093 / bioinformatics / btab801.
[0113] Table 3: Test results of the training set PSI:Bio VCXlogy dataset
[0114] Table 4: Test results of the independent test set NESG dataset
[0115] Table 5: Test results of the independent test set Camsol dataset
[0116] Regarding Table 5, it should be noted that the independent test set Camsol dataset contains 56 mutant sequences, of which the literature Trevino contains 22 protein variants, Miklos contains 3 protein variants, Tan contains 1 protein variant, and Dudgeon contains 30 protein variants. Different prediction models and the disclosed model predict the accuracy of changes in protein solubility after point mutations.
[0117] Table 6: Test results of the independent test set eSol dataset
[0118] Those skilled in the art will understand that the prediction of a regression problem is generally expressed by the Pearson correlation coefficient, where a larger value indicates a higher positive correlation between the predicted value and the true value.
[0119] The indicators in Tables 3 to 5 are explained as follows:
[0120] Area Under Curve (AUC) refers to the area under the receiver operating characteristic curve (ROC);
[0121] Accuracy (ACC) refers to the ratio of correctly classified samples to the total number of samples. The accuracy calculation formula is: ACC = (TP + TN) / (P + N). Generally, the higher the accuracy, the better the classifier.
[0122] Precision refers to the proportion of correctly predicted results among all the results predicted as positive samples. The precision calculation formula is: Precision = TP / (TP + FP). The closer the value is to 1, the better the performance.
[0123] MCC (Matthews correlation coefficient) is an indicator often used to evaluate the predictive performance of classification models. It describes the correlation coefficient between the actual classification and the predicted classification. The calculation formula of MCC is: MCC = (TP*TN-FP*FN) / (sqrt((TP+FP)*(TP+FN)*(TN+FP)*(TN+FN))). The value range of MCC is between -1 and +1, where: +1 indicates perfect prediction, 0 indicates random prediction, and -1 indicates that the prediction is completely inconsistent with the actual observation.
[0124] Among them, TP (True Positive) and TN (True Negative) represent the number of correctly classified positive samples and negative samples, FP (False Positive) and FN (False Negative) represent the number of incorrectly classified positive samples and negative samples, and P and N represent the number of positive samples and negative samples.
[0125] From the comprehensive test results of the training set and three independent test sets, it can be seen that the prediction model constructed by the present invention not only has better classification performance, but also takes into account the application in regression problems (such as predicting the impact of amino acid mutations on protein solubility). Due to the different sources of different data sets, the experimental conditions (including conditions such as culture medium, cell line, protein expression and purification) are also different, resulting in a decline in the performance of the model obtained by training when generalized use. However, the method flow of the present invention is simple and easy to transplant. By retraining the experimental data generated for specific experimental conditions as a training set, the performance of the model prediction solubility can be further improved, and even the process can be used to solve problems such as protein expression prediction, protein folding type and folding rate constant prediction.
[0126] Those skilled in the art will appreciate that, based on the protein solubility prediction method disclosed herein, a system for predicting protein solubility can be developed. FIG5 shows a schematic block diagram of a system for predicting protein solubility according to the present disclosure. Specifically, the system 500 for predicting protein solubility includes an acquisition module 510 and a processing module 520.
[0127] The acquisition module 510 is used to acquire a sample to be predicted, where the sample to be predicted includes a protein sequence of the sample to be predicted.
[0128] The processing module 520 is used to obtain a prediction result of the protein solubility of the sample to be predicted by inputting the protein sequence of the sample to be predicted. As shown in Figure 5, the processing module 520 can further include: a first feature extraction submodule 521, which is used to extract features based on the protein sequence of the sample to be predicted using a protein pre-training model; a second feature extraction submodule 522, which is used to extract the primary structural features of the protein based on the protein sequence of the sample to be predicted; a feature splicing submodule 523, which is used to splice the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the protein of the sample to be predicted; and a model prediction submodule 524, which is used to input the feature vector of the protein of the sample to be predicted into a protein solubility prediction model based on an automatic machine learning framework to obtain a prediction result of the protein solubility.
[0129] Those skilled in the art will appreciate that the system for predicting protein solubility described above can be implemented by a computer. For example, the following operations can be implemented by executing a computer program: obtaining a sample to be predicted, the sample to be predicted including a protein sequence of the sample to be predicted; obtaining a prediction result of the protein solubility of the sample to be predicted by inputting the protein sequence of the sample to be predicted, wherein the operation further includes the following sub-operations: extracting features using a protein pre-training model based on the protein sequence of the sample to be predicted; extracting the primary structural features of the protein based on the protein sequence of the sample to be predicted; concatenating the features extracted by the protein pre-training model with the primary structural features of the protein to obtain a feature vector of the protein of the sample to be predicted; inputting the feature vector of the protein of the sample to be predicted into the protein solubility prediction model to obtain a prediction result of the protein solubility of the sample to be predicted. That is, although in the system for predicting protein solubility according to the present disclosure, the various operational functions corresponding to the prediction method are described as modules or sub-modules, those skilled in the art will appreciate that such modules or sub-modules may not be composed of circuit components or other physical components, but rather functional modules constructed by computer programs.
[0130] In addition, those skilled in the art will recognize that the methods disclosed herein can be implemented as computer programs. As described above in conjunction with the accompanying drawings, the methods of the above embodiments are executed by one or more programs, and the instructions in the programs cause a computer or processor to execute the algorithms described in conjunction with the accompanying drawings. These programs can be stored and provided to a computer or processor using various types of non-transitory computer-readable media. Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (such as floppy disks, magnetic tapes, and hard disk drives), magneto-optical recording media (such as magneto-optical disks), CD-ROMs (compact disk read-only memories), CD-Rs, CD-R / Ws, and semiconductor memories (such as ROMs, PROMs (programmable ROMs), EPROMs (erasable PROMs), flash ROMs, and RAMs (random access memories)). Furthermore, these programs can be provided to a computer using various types of transient computer-readable media. Examples of transient computer-readable media include electrical signals, optical signals, and electromagnetic waves. Transient computer-readable media can be used to provide programs to a computer via wired communication paths such as wires and optical fibers or wireless communication paths.
[0131] For example, according to one embodiment of the present disclosure, a non-transitory computer-readable storage medium may be provided for storing a computer program, wherein the computer program includes instructions that, when executed by a processor of an electronic device, enable the electronic device to implement the protein solubility prediction method described above.
[0132] In addition, according to the present disclosure, a computer system may be provided, comprising: a processor; a memory; and a computer program. The computer program is stored in the memory and configured to be executed by the processor. The computer program includes instructions for implementing the above-described protein solubility prediction method.
[0133] On the other hand, according to one embodiment of the present disclosure, a non-transitory computer-readable storage medium may be provided for storing a computer program, wherein the computer program includes instructions that, when executed by a processor of an electronic device, enable the electronic device to implement the method for constructing a protein solubility prediction model as described above.
[0134] In addition, according to the present disclosure, a computer system may be provided, comprising: a processor; a memory; and a computer program. The computer program is stored in the memory and configured to be executed by the processor. The computer program includes instructions for implementing the method for constructing the protein solubility prediction model described above.
[0135] The implementation methods of the present disclosure are not limited to those described in the above embodiments. Without departing from the spirit and scope of the present disclosure, ordinary technicians in this field can make various changes and improvements to the present disclosure in form and details, and these are all considered to fall within the scope of protection of the present disclosure.
Claims
1. A method for predicting protein solubility, characterized in that: The method comprises: Based on the protein sequence, features are extracted using a protein pre-trained model; Extract the primary structural features of proteins based on protein sequences; The features extracted by the protein pre-training model are concatenated with the primary structure features of the protein to obtain a feature vector of the protein; The characteristic vector of the protein is input into a protein solubility prediction model to obtain a prediction result of the protein solubility.
2. The method according to claim 1, characterized in that The protein pre-training model is implemented by a combination of one or more of the following models: ESM-1b, UniRep, ProteinBert, TAPE, ProtGPT2, ProtTXL, ProtBert, ProtXLNet, ProtAlbert, ProtElectra, ProtT5-XL-BFD, ProtT5-XL-UniRef50 and ProtT5-XXL; preferably, the protein pre-training model is ProtT5-XL-BFD or ProtT5-XL-UniRef50; more preferably, the protein pre-training model is ProtT5-XL-UniRef50.
3. The method according to claim 1, characterized in that The method of extracting features based on protein sequence and using protein pre-training model includes: Taking the protein sequence as input, the ProtT5-XL-UniRef50 pre-trained model is used to extract the embedding layer vector output by the encoder, where each amino acid corresponds to a 1024-dimensional vector and each sequence corresponds to an L×1024-dimensional feature matrix, where L is the length of the amino acid sequence of the protein; Take the average value of the above feature matrix by column to obtain a 1024-dimensional feature vector.
4. The method according to claim 3, characterized in that The step of taking the average value of the feature matrix by columns to obtain a 1024-dimensional feature vector further includes: The Sentence-T5Encoder-onlymean method is used to average the values of each dimension of the embedding layer feature values of all amino acids output by the encoder by column to obtain a 1024-dimensional feature vector.
5. The method according to claim 1, characterized in that The method of extracting the primary structural features of a protein based on a protein sequence includes: Taking protein sequence as input, the iFeature tool is used to extract the primary structural features of the protein.
6. The method according to claim 1, characterized in that The primary structural features of the protein are selected from one or more of AAC, DPC, DDE, TPC, CKSAAP, EAAC, GAAC, CKSAAGP, GDPC, GTPC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, KSCTriad, SOCNumber, QSOrder, PAAC, APAAC, KNNprotein, PSSM, AAINDEX and BLOSUM62; preferably, the primary structural features of the protein include AAC, GAAC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, PAAC and APAAC.
7. The method according to claim 1, characterized in that The protein solubility prediction model is constructed through the following steps: Acquire a data set, wherein the data set includes a plurality of training samples, each of the training samples includes a sample protein sequence and solubility data thereof; Based on the sample protein sequence, features are extracted using a protein pre-trained model; Extract the primary structural features of the protein based on the sample protein sequence; The features extracted by the protein pre-training model are combined with the primary structure features of the protein to obtain a feature vector of the sample protein; The characteristic vector of the sample protein is used as input data, the solubility data of the sample protein is used as output data, and the machine learning model is trained to obtain a final protein solubility prediction model.
8. The method according to claim 7, characterized in that The training of the machine learning model includes: Based on the automatic machine learning framework, one or more machine learning models are selected as candidate models, and the protein solubility prediction model is constructed and trained.
9. The method according to claim 8, characterized in that The automatic machine learning framework is AutoGluon; the candidate models are deep learning models NN_Torch and FASTAI.
10. The method according to claim 1 or 7, characterized in that: The method further comprises: Extract other features of proteins based on protein sequences; The features extracted by the protein pre-training model are spliced with the primary structure features of the protein and other features of the protein to obtain a spliced feature vector of the protein.
11. The method according to claim 10, characterized in that The other characteristics of the protein are selected from one or more of molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorbance, total average hydrophilicity, the proportion of helix / turn / fold in the secondary structure of the protein, secondary structure element content (SSEC), secondary structure element binary (SSEB), disorder, disorder content (DisorderC), disorder binary (DisorderB), barrier-free solvent accessibility (ASA), torsion angle (TA), Z-scale (ZSCALE) and 48 pseudo K-tuple reduced amino acid composition (48PseKRAAC); preferably, the other characteristics include molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorbance, total average hydrophilicity and the proportion of helix / turn / fold in the secondary structure of the protein.
12. A method for constructing a protein solubility prediction model, characterized in that: The construction method comprises: Acquire a data set, wherein the data set includes a plurality of training samples, each of the training samples includes a sample protein sequence and solubility data thereof; Based on the sample protein sequence, features are extracted using a protein pre-trained model; Extract the primary structural features of the protein based on the sample protein sequence; The features extracted by the protein pre-training model are spliced with the primary structure features of the protein to obtain a feature vector of the sample protein; The characteristic vector of the sample protein is used as input data, the solubility data of the sample protein is used as output data, and the machine learning model is trained to obtain a final protein solubility prediction model.
13. The method according to claim 12, characterized in that The protein pre-training model is implemented by a combination of one or more of the following models: ESM-1b, UniRep, ProteinBert, TAPE, ProtGPT2, ProtTXL, ProtBert, ProtXLNet, ProtAlbert, ProtElectra, ProtT5-XL-BFD, ProtT5-XL-UniRef50 and ProtT5-XXL; preferably, the protein pre-training model is ProtT5-XL-BFD or ProtT5-XL-UniRef50; more preferably, the protein pre-training model is ProtT5-XL-UniRef50.
14. The method according to claim 12, characterized in that The method of extracting features based on the sample protein sequence using the protein pre-training model includes: The sample protein sequence is used as input, and the ProtT5-XL-UniRef50 pre-trained model is used to extract the embedding layer vector output by the encoder, where each amino acid corresponds to a 1024-dimensional vector and each sequence corresponds to an L×1024-dimensional feature matrix, where L is the length of the amino acid sequence of the sample protein; Take the average value of the above feature matrix by column to obtain a 1024-dimensional feature vector.
15. The method according to claim 14, characterized in that The step of taking the average value of the feature matrix by columns to obtain a 1024-dimensional feature vector further includes: The Sentence-T5Encoder-onlymean method is used to average the values of each dimension of the embedding layer feature values of all amino acids output by the encoder by column to obtain a 1024-dimensional feature vector.
16. The method according to claim 12, characterized in that The method of extracting the primary structural features of the protein based on the sample protein sequence includes: Using the sample protein sequence as input, the iFeature tool is used to extract typical primary structural features of the protein.
17. The method according to claim 12, characterized in that The primary structural features of the protein are selected from one or more of AAC, DPC, DDE, TPC, CKSAAP, EAAC, GAAC, CKSAAGP, GDPC, GTPC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, KSCTriad, SOCNumber, QSOrder, PAAC, APAAC, KNNprotein, PSSM, AAINDEX and BLOSUM62; preferably, the primary structural features of the protein include AAC, GAAC, Moran, Geary, NMBroto, CTDC, CTDT, CTDD, CTriad, PAAC and APAAC.
18. The method according to claim 12, characterized in that The machine learning model is a deep learning model.
19. The method according to claim 12, characterized in that The training of the machine learning model includes: Based on the automatic machine learning framework, one or more machine learning models are selected as candidate models, and the protein solubility prediction model is constructed and trained.
20. The method according to claim 19, characterized in that The automatic machine learning framework is AutoGluon; the candidate models are deep learning models NN_Torch and FASTAI.
21. The method according to claim 12, characterized in that The method further comprises: Extract other features of proteins based on sample protein sequences; The features extracted by the protein pre-training model are spliced with the primary structure features of the protein and other features of the protein to obtain a spliced feature vector of the sample protein.
22. The method according to claim 21, characterized in that The other characteristics of the protein are selected from one or more of molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorbance, total average hydrophilicity, the proportion of helix / turn / fold in the secondary structure of the protein, SSEC, SSEB, Disorder, DisorderC, DisorderB, ASA, TA, ZSCALE and 48PseKRAAC; preferably, the other characteristics include molecular weight, aromaticity, instability coefficient, elasticity, isoelectric point, molar absorbance, total average hydrophilicity and the proportion of helix / turn / fold in the secondary structure of the protein.
23. A protein solubility prediction model constructed according to the method described in any one of claims 12 to 22.
24. A computer-implemented system for predicting protein solubility, wherein the computer-implemented system implements the following operations by executing a computer program: Acquire a sample to be predicted, wherein the sample to be predicted includes a protein sequence of the sample to be predicted; By inputting the protein sequence of the sample to be predicted, a prediction result of the protein solubility of the sample to be predicted is obtained, and the operation further includes the following sub-operations: Based on the protein sequence of the sample to be predicted, a protein pre-training model is used to extract features; Based on the protein sequence of the sample to be predicted, extract the primary structural features of the protein; The features extracted by the protein pre-training model are spliced with the primary structure features of the protein to obtain a feature vector of the protein of the sample to be predicted; The characteristic vector of the protein of the sample to be predicted is input into the protein solubility prediction model to obtain the prediction result of the protein solubility of the sample to be predicted.
25. A non-transitory computer-readable storage medium for storing a computer program, wherein the computer program includes instructions, which, when executed by a processor of an electronic device, causes the electronic device to implement the protein solubility prediction method as described in any one of claims 1 to 11 or the method for constructing a protein solubility prediction model as described in any one of claims 12 to 22.
26. A computer system, comprising: processor; Memory; and A computer program, wherein the computer program is stored in the memory and is configured to be executed by the processor, and the computer program includes instructions for implementing the protein solubility prediction method according to any one of claims 1 to 11 or the method for constructing a protein solubility prediction model according to any one of claims 12 to 22.
Citation Information
Patent Citations
Protein-protein interaction prediction method based on deep forest
CN111210871A
Plant resistance protein recognition method, device and equipment and storage medium
CN112906755A
Protein solubility prediction method based on multi-dimensional sequence embedding
CN113223620A
Protein folding identification method and system based on embedded features and unbalanced classification loss
CN114550824A
Protein solubility prediction method based on combined machine learning model
CN114582423A
