Neural network-based acetylcholin esterase inhibitor screening method

Abstract

The invention provides an acetylcholin esterase inhibitor screening method based on a neural network. The method comprises the steps of 1, constructing a multi-species data set; step 2, data set preprocessing and sample labeling; step 3, constructing graph structure data representation; step 4, multi-scale feature extraction; 5, feature fusion and binding affinity prediction; step 6, model training and verification; and 7, screening and verifying the acetylcholin esterase inhibitor. According to the method, the problems of single species limitation and low precision in the existing method can be solved, the screening efficiency is improved by several times compared with that of the traditional method, the high-affinity AChE inhibitor can be rapidly identified, and a key tool is provided for research and development of Alzheimer disease treatment drugs through a differential feature extraction and alternating fusion mechanism.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, bioinformatics and drug research and development, and particularly relates to a screening method for acetylcholinesterase inhibitors based on a neural network. BACKGROUND

[0002] Alzheimer's Disease (AD) is a common neurodegenerative disease, and its pathological mechanism is closely related to the degeneration of basal forebrain cholinergic neurons, resulting in a significant decrease in acetylcholine (ACh) levels. Acetylcholinesterase (AChE) is a key enzyme for degrading ACh, and its increased activity will exacerbate cholinergic deficiency. Therefore, acetylcholinesterase inhibitors (AChEIs) have become the main drugs for the current symptomatic treatment of AD. Although existing marketed AChEIs (such as donepezil, rivastigmine, etc.) can alleviate symptoms, they cannot delay disease progression and have certain side effects. Traditional high-throughput screening methods are time-consuming and costly, and are difficult to effectively capture the complex interaction relationship between proteins and ligands.

[0003] In recent years, artificial intelligence technology has been gradually applied to the screening of AChEIs. For example, some studies use graph neural networks (GNN) to encode molecular graphs to predict inhibitory activity; some methods use deep convolutional neural networks (DCNN) to predict the binding affinity of AChE and molecules; in addition, machine learning methods such as random forests (RF) are also used to construct quantitative structure-activity relationship (QSAR) models. However, the existing technology still has the following obvious limitations:

[0004] Species limitation: Most models are only trained based on human AChE data, ignoring the conservation and differences of AChE between different species, resulting in insufficient generalization ability of the model and difficulty in applying to cross-species drug development scenarios; insufficient feature extraction: traditional graph neural networks can usually only capture local neighborhood information of molecules, and are difficult to effectively model long-range interactions between molecular skeletons and protein active pockets; weak interaction modeling capability: existing methods usually use simple vector splicing to fuse protein and ligand features, without fully considering the bidirectional dependency relationship between them, resulting in low prediction accuracy of binding affinity, and the AUC of existing models is generally lower than 95%.

[0005] Therefore, there is an urgent need in the art for a new AChEI screening method that can integrate multi-species information, implement multi-scale feature extraction, and perform deep interaction modeling. SUMMARY

[0006] Therefore, the embodiments of the present application provide a screening method for acetylcholinesterase inhibitors based on a neural network to solve the problems of insufficient generalization, feature extraction and interaction modeling of the AChEI screening method in the prior art.

[0007] In a first aspect, the application provides a method for screening acetylcholinesterase inhibitors based on a neural network, comprising:

[0008] Step one, constructing a multi-species dataset, including acetylcholinesterase data of six species and corresponding inhibitor molecule data; the species include Homo sapiens, Electrophorus electricus, Mus musculus, Bos taurus, Torpedo californica, and Rattus norvegicus; the inhibitor molecule data is from the BindingDB database, and the acetylcholinesterase data is from the AlphaFold2 database;

[0009] Step two, data set preprocessing and sample labeling, including:

[0010] 1) Inhibitor molecule preprocessing: removing molecules with a molecular weight > 500 Da or containing more than 3 ring structures, and retaining at least one active pharmacophore; using half-inhibition concentration as the labeling standard, defining IC50≤1 μM as a positive sample and IC50>1 μM as a negative sample, and randomly selecting positive and negative samples in a 1:1 ratio to construct a balanced dataset;

[0011] 2) Acetylcholinesterase protein preprocessing: for acetylcholinesterases with an amino acid sequence > 300 residues, extracting the key region around the active pocket; based on the 3D coordinates provided by AlphaFold2, calculating the Euclidean distance between residues, and adding edges when the distance < 8.0 Å to construct a residue contact map;

[0012] 3) External validation set construction: extracting acetylcholinesterase inhibitor data for the six species from the ChEMBL database, and constructing an external validation set using the labeling standard in the inhibitor molecule preprocessing;

[0013] Step three, constructing a graph structure data representation, including:

[0014] 1) Inhibitor molecule graph representation: taking atoms in the molecule as nodes and chemical bonds as edges; each atom node is encoded as a 105-dimensional feature vector , the formula is: wherein, is a 100-dimensional one-hot encoding of the atom type, is the atomic partial charge, is the charge mean and standard deviation, respectively, is the van der Waals radius, is the maximum radius, is the minimum radius, is the aromaticity binary indicator, is the chiral descriptor;

[0015] 2) Acetylcholinesterase protein graph representation: Acetylcholinesterase residues as nodes, residue contact relationship as edge; Each residue node is encoded as a 24-dimensional feature vector , the formula is: , wherein, is a 20-dimensional one-hot encoding of the amino acid type, is the secondary structure probability from the DSSP algorithm, is a binary indicator of solvent accessibility, which is 1 when the solvent accessibility ASA is greater than or equal to 25%;

[0016] Step four, multi-scale feature extraction, including:

[0017] 1) Inhibitor molecule feature extraction: Three-level dilated graph convolution based on SAGE network, dilation factor is 2 k , k is the number of layers, and each layer extracts features through mean pooling and nonlinear activation, the formula is: , wherein, is a weight matrix, is the neighborhood when dilation = 2 k ; Then through 4-head attention pooling, the atomic features are spliced and projected into 512-dimensional global molecular features ;

[0018] 2) Acetylcholinesterase protein feature extraction: Two layers of graph attention network, each layer contains 4 independent attention heads, calculate attention score and feature aggregation, the 4 attention heads output is spliced into 128-dimensional features, after GraphNorm and residual connection, 512-dimensional global protein features are obtained through global average pooling ;

[0019] Step five, feature fusion and binding affinity prediction, including:

[0020] 1) Bidirectional cross-attention fusion: Project the global molecular features and the global protein features into query (Q), key (K), and value (V) vectors, respectively, and calculate the attention weights and fuse the features in the "protein→ligand" and "ligand→protein" directions, the formulas are respectively:

[0021] ,

[0022] , wherein, are protein query, ligand key, and value vectors, respectively, is a projection matrix; After fusion, a 1024-dimensional joint feature is obtained ;

[0023] 2) Affinity prediction: the joint features are processed by a multi-layer perceptron , first compressed to 512 dimensions, then projected to 256 dimensions, and finally output the prediction probability by Softmax ;

[0024] The training uses binary cross-entropy loss, the formula is: Where, is the true label, is the predicted probability, and N is the number of samples;

[0025] Step six, model training and verification, including:

[0026] The balanced dataset is divided into training set, validation set and test set according to the ratio of 8:1:1; Set the training parameters, and determine that the model training is completed when the area under the curve is greater than or equal to 95%;

[0027] Step seven, acetylcholinesterase inhibitor screening and verification, including:

[0028] 1) Candidate molecule screening: extract small molecules from ZINC database, input the trained model after preprocessing the inhibitor molecules, and select molecules with a prediction probability greater than or equal to 0.97 as candidates;

[0029] 2) Molecular docking verification: using CB-Dock2 platform, taking the active pocket of human acetylcholinesterase as the docking center, calculating the Vina score, and selecting molecules with a Vina score less than or equal to -9 kcal / mol as high-affinity acetylcholinesterase candidates.

[0030] Preferably, the acetylcholinesterase data of the 6 species is specifically: 7274 for Homo sapiens, 5289 for Electrophorus electricus, 530 for Mus musculus, 303 for Bos taurus, 249 for Torpedo californica, and 2033 for Rattus norvegicus. Inhibitor molecules; There are a total of 19845 samples in the balanced dataset, of which 9974 are positive and 9871 are negative; There are a total of 6710 samples in the external verification set, of which 3535 are positive and 3175 are negative.

[0031] Preferably, the neighborhood range of the three-level expanded graph convolution in step four is respectively: the first layer (k=0) neighborhood is the directly bonded atom, the second layer (k=1) neighborhood is 2 atoms away from the atom, and the third layer (k=2) neighborhood is 4 atoms away from the atom.

[0032] Preferably, the number of attention heads in the bidirectional cross-attention in step five is 4, and the feature dimension of each attention head is 128; The dropout rate of the multi-layer perceptron is 0.2.

[0033] Preferably, the indicators used in the model verification in step six include accuracy, precision, recall, F1 score, Matthews correlation coefficient and area under the ROC curve.

[0034] Preferably, the docking center in step seven is consistent with the active pocket region in step two.

[0035] Preferably, after the attention head output of the graph attention network in step four is spliced, it is converted into a 128-dimensional feature through a linear projection matrix, and then subjected to GraphNorm and residual connection.

[0036] Preferably, the joint features in step five are The molecular-level representation is obtained by global average pooling.

[0037] Preferably, when constructing the residue contact map in step two, the Euclidean distance calculation is based on the Cα atomic coordinates provided by AlphaFold2.

[0038] Preferably, the solvent accessibility ASA in step three is determined by a relative accessibility surface area threshold of 25%, calculated by the ASAcalc tool.

[0039] The embodiments of the present application have the beneficial effects that: the embodiments of the present application have strong generalization ability: by integrating AChE data of Homo sapiens, Electrophorus electricus, Mus musculus and other 6 species at different evolutionary levels, the model can learn the conserved active site features across species. The AUC on the independent external validation set (Dataset B) reaches 89.1%, which is significantly better than the model trained based on only single species data, effectively solving the problem of insufficient generalization of existing methods due to single species.

[0040] High prediction accuracy: the present application innovatively adopts the technical path of "multi-scale feature extraction + bidirectional cross-attention fusion". On the one hand, the long-range dependence relationship of the inhibitor molecule and the context information of the protein residue are captured by the dilated graph convolution and GAT network respectively; on the other hand, the bidirectional cross-attention mechanism is used to model the complex interaction between protein and ligand, and finally an AUC of 97.3% is achieved on the test set, which is about 4.1 percentage points higher than the existing optimal model (such as MolTrans).

[0041] Comprehensive feature recognition: the model can capture both local and global molecular and protein features. For molecules, it can extend from atomic-level chemical bonds to scaffold-level long-range structures; for proteins, it can effectively integrate the conformation information of the active pocket and its surrounding, thereby more comprehensively reflecting the binding mechanism.

[0042] Screening efficiency and cost advantage are remarkable: the application realizes efficient computer virtual screening, and the screening speed can reach about 1000 molecules / min, which is several orders of magnitude higher than the traditional high-throughput screening (usually 100 molecules / day), and can greatly shorten the research and development period and reduce the economic cost.

[0043] Strong practicability and verifiable results: the final screened candidate molecules not only have high model prediction confidence (≥0.97), but also pass the molecular docking verification (Vina Score ≤ -9 kcal / mol), and 3 candidate molecules with high binding affinity are obtained, and the docking score of the optimal molecule is as low as -12.8 kcal / mol, which provides a direct and reliable lead compound target for subsequent Alzheimer's disease drug research and development. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0045] Figure 1 is a flowchart of a neural network-based acetylcholinesterase inhibitor screening method provided by an embodiment of the application; DETAILED DESCRIPTION

[0046] In the following description, specific details are set forth in order to provide a thorough understanding of the embodiments of the application. However, it will be apparent to those skilled in the art that the application can be practiced without these specific details. In other instances, well-known systems, devices, circuits, and methods have been omitted in order to not obscure the description of the application with unnecessary detail.

[0047] A neural network-based acetylcholinesterase inhibitor screening method according to an embodiment of the application will be described in detail below with reference to the accompanying drawings.

[0048] Figure 1 is a flowchart of a neural network-based acetylcholinesterase inhibitor screening method provided by an embodiment of the application. As shown in Figure 1 , the neural network-based acetylcholinesterase inhibitor screening method comprises:

[0049] S101, constructing a multi-species dataset;

[0050] Specifically, this multi-species dataset includes acetylcholinesterase data and corresponding inhibitor molecule data for six species: Homo sapiens, electric eel, house mouse, bovine, California electric ray, and brown rat. Inhibitor molecule data can be downloaded from the BindingDB database, and predicted acetylcholinesterase structures for the six species can be obtained from the AlphaFold2 database. These six species—Homo sapiens (UniProt ID: P22303), electric eel (Electrophorus electricus), house mouse (Mus musculus), bovine (Bos taurus), California electric ray (Torpedo californica), and brown rat (Rattus norvegicus)—were chosen because their AChE sequence homology covers an evolutionary lineage from fish to mammals, capturing conserved active site characteristics. AChE inhibitor data for these six species were downloaded from the ChEMBL database to validate the model's generalization ability.

[0051] The multi-species dataset is divided into two categories: one is inhibitor molecule features, which describe the molecular chemical structure based on SMILES strings (simplified linear input canonicalization); the other is AChE protein features, which describe the protein 3D structure based on residue contact maps. The number of inhibitor molecules for the six species are as follows: Homo sapiens 7274, electric eel 5289, house mouse 530, bovine 303, California electric ray 249, and brown rat 2033.

[0052] S102, Dataset Preprocessing and Sample Labeling;

[0053] This step involves data cleaning and labeling to ensure sample quality and balance, providing reliable input for model training. Specifically, it includes: inhibitor molecule preprocessing, removing molecules with a molecular weight >500 Da (not conforming to Lipinski rules) or containing more than 3 ring structures (high synthesis difficulty), and retaining at least one active pharmacophore (such as amino or ester groups—key pharmacophores of known AChE inhibitors). Sample labeling uses IC50 (half-maximal inhibitory concentration, the gold standard for measuring inhibitor potency) as an indicator, defining IC50 ≤ 1 μM as positive samples (effective inhibitors, significantly inhibiting AChE activity), and IC50 > 1 μM as negative samples (ineffective inhibitors); to avoid class imbalance, samples are randomly selected from the negative samples at a 1:1 ratio, finally constructing a balanced dataset (Dataset A), which contains 19845 samples: 9974 positive and 9871 negative. AChE protein pretreatment and key region extraction: For AChE with an amino acid sequence >300 residues (e.g., human AChE contains 537 residues), residues within a 20 Å radius around the active pocket are extracted. The active pocket is the core region for AChE binding to inhibitors. This pretreatment reduces redundant information and focuses on function-related structures. Residue contact map construction: Based on the Cα atom 3D coordinates provided by AlphaFold2, the Euclidean distance of all residue pairs is calculated. When the distance is <8.0 Å, the two residues are considered to be in contact, and an edge connection is added. This operation effectively preserves key interaction information of the protein structure. External validation set construction: Using the same pretreatment and annotation standards as Dataset A, data is extracted from the ChEMBL database to construct an external validation set (Dataset B). The external validation set includes 6710 samples: 3535 positive and 3175 negative, used to verify the generalization ability of the model under non-training data distributions.

[0054] S103, Construct a graph-structured data representation;

[0055] This step converts the inhibitor molecules and AChE protein into graph structure data, adapting to the feature extraction requirements of graph neural networks. The core is constructing node feature vectors rich in chemical / structural information. Specifically, this includes:

[0056] Inhibitor molecule graph representation: Atoms in the molecule are treated as nodes, and chemical bonds as edges; each atomic node is encoded as a 105-dimensional feature vector. The formula is: Among them, among them, 100-dimensional one-heat encoding for atomic types, The charge of the atom. These are the mean and standard deviation of the charge, respectively. For van der Waals radius, Maximum radius For the minimum radius, It is an aromatic binary indicator. This is a chiral descriptor; specifically, the inhibitor's molecular diagram structure is defined as follows: atoms are nodes, and chemical bonds (single, double, triple, and aromatic bonds) are edges; atomic node features (105 dimensions): composed of 5 parts: 100-dimensional one-heat encoding: covering common atoms (C, H, O, N, S, etc.) and rare atoms (such as P, F), capturing differences in atom types; 1-dimensional normalized formal charge: eliminating differences in charge scale between different molecules, as shown in the formula. The mean and standard deviation of the charges of all atoms in Dataset A; 1-dimensional normalized van der Waals radius: reflecting the spatial size of atoms. , These represent the minimum (1.2 Å, H atom) and maximum (2.0 Å, S atom) van der Waals radii of atoms in Dataset A, respectively; 1-dimensional aromaticity indicator: 1 represents an aromatic atom (such as C on a benzene ring), and 0 represents a non-aromatic atom. Aromaticity is key to the hydrophobic interaction between AChE inhibitors and the active pocket; 2-dimensional chirality descriptor: distinguishes between R-type and S-type chirality. Chirality differences significantly affect the binding conformation of the inhibitor and AChE.

[0057] Acetylcholinesterase protein diagram representation: Acetylcholinesterase residues are used as nodes, and residue contact relationships are used as edges; each residue node is encoded as a 24-dimensional feature vector. The formula is: ,in, It is a 20-dimensional unique heat encoding of amino acid types. The probability of secondary structure comes from the DSSP algorithm. This is a binary indicator for solvent accessibility, set to 1 when ASA accessibility is ≥25%. Specifically, the AChE protein graph structure is defined as follows: residues are nodes, and residue contact relationships are edges. Residue node features (24 dimensions) consist of four parts: 20-dimensional one-hot encoding covering 20 natural amino acids, distinguishing the chemical properties of different residues (e.g., acidic Asp, basic Lys); and 3-dimensional secondary structure probability calculated by the DSSP algorithm, outputting... -spiral( ), -fold( ), the probability of random coiling, secondary structure determines the local conformational stability of protein; 1-dimensional solvent accessibility indicator: relative accessible surface area (ASA) calculated by the ASAcalc tool, ASA≥25% is 1 (surface residues, easy to interact with inhibitors), otherwise it is 0 (internal residues, mainly maintain protein structure).

[0058] S104, Multi-scale feature extraction;

[0059] This step employs a hybrid architecture of "molecular SAGE network + protein GAT network" to extract multi-scale features of inhibitors and AChE respectively, overcoming the limitation of traditional GNNs that only capture local information. Specifically, it includes:

[0060] 1) Inhibitor molecule feature extraction: A three-level dilated graph convolution based on the SAGE network was used, with a dilation factor of 2. k k is the number of layers. Each layer extracts features through mean pooling and non-linear activation, as shown in the formula: ,in, This is the weight matrix. dilation=2 k The time neighborhood is then used; subsequently, through 4-head attention pooling, the atomic features are concatenated and projected into 512-dimensional global molecular features. The third-level dilated graph convolution is achieved through a dilation factor of 2. k (k=0,1,2) Multi-scale neighborhood sampling is implemented: First layer (k=0): dilation=1, sampling directly bonded atoms to capture local chemical structures (such as functional groups); Second layer (k=1): dilation=2, sampling atoms with 2 bond distances to capture medium-range interactions (such as spatial relationships between functional groups); Third layer (k=2): dilation=4, sampling atoms with 4 bond distances to capture long-range dependencies (such as molecular scaffold structures); Each layer aggregates neighborhood features through mean pooling, and then weighted by a weight matrix. With Sigmoid activation ( The data is converted to 128-dimensional atomic features; 4-head attention pooling: to highlight pharmacophore atoms (such as atoms that bind to the AChE active site), each attention head independently calculates the atomic importance weight. The outputs of the four heads are spliced ​​together and projected through a matrix. 512-dimensional global molecular features were obtained. This design allows for focusing on critical atoms and reducing interference from redundant information.

[0061] 2) Acetylcholinesterase protein feature extraction: A two-layer graph attention network is used, with each layer containing 4 independent attention heads. The attention score is calculated using the following formula: The outputs of the four attention heads are concatenated into 128-dimensional features. After being connected to the residuals using GraphNorm, global average pooling is applied to obtain 512-dimensional global protein features. Specifically, the two-layer graph attention network (GAT network) contains four independent attention heads in each layer, enhancing the expression of residue features; attention score calculation is performed by querying... AND key The dot product of vectors, divided by (Attention head dimension) normalization to avoid gradient explosion.

[0062] Feature aggregation: Attention weights are normalized using Softmax and combined with value vectors. The features of the aggregated neighborhood residues are then activated by Sigmoid to obtain 32-dimensional features per attention head, as shown in the following formula: .

[0063] Feature post-processing: The outputs of the four attention heads are concatenated into a 128-dimensional temporary feature. Through linear projection matrix The features are converted to 128 dimensions, then processed by GraphNorm (for stable training) and residual connections (to alleviate gradient vanishing), and finally global average pooling is used to obtain 512-dimensional global protein features. Global pooling can integrate the structural information of the entire AChE, providing a foundation for subsequent interaction modeling.

[0064] S105, Feature fusion and binding affinity prediction;

[0065] 1) Bidirectional cross-attention fusion: This integrates the global molecular features... With the global protein features Projected as query (Q), key (K), and value (V) vectors respectively, attention weights are calculated and features are fused in both the "protein→ligand" and "ligand→protein" directions, as shown in the formulas:

[0066] ,

[0067] ,in, These are protein lookup, ligand bond, and value vector, respectively. The projection matrix is ​​used; after fusion, a 1024-dimensional joint feature is obtained. Specifically, feature projection: projecting 512-dimensional molecular features... Protein characteristics Linear transformations are used to obtain query (Q), key (K), and value (V) vectors, each with a dimension of 128 (4 attention heads × 128 dimensions / head); bidirectional interaction: protein → ligand direction: using protein features as the query. The ligand is characterized by a bond. AND value Calculate attention weights Weighted summation yields 512-dimensional ligand enhancement features; ligand→protein direction: symmetric processing, using ligand features as the query... Proteins are characterized by bonds AND value 512-dimensional protein enhancement features were obtained; joint feature construction: the enhanced ligand / protein features were integrated through global average pooling and spliced ​​to obtain 1024-dimensional joint features. This feature contains bidirectional protein-ligand interaction information and is a core input for predicting binding affinity.

[0068] 2) Affinity prediction: The joint features are processed using a multilayer perceptron. First, it is compressed to 512 dimensions, then projected to 256 dimensions, and finally the predicted probability is output through Softmax. Training uses binary cross-entropy loss, the formula of which is: in, For real labels, To predict probabilities, N is the number of samples. Specifically, affinity prediction uses a three-layer MLP structure, including: the first layer: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] The first layer compresses the features to 512 dimensions and uses LeakyReLU activation (to address the vanishing negative gradient problem of ReLU), adding batch normalization (to mitigate internal covariate bias) and dropout (with a dropout rate of 0.2 to reduce overfitting). The second layer projects the 512-dimensional features to 256 dimensions and uses GELU activation (smoother than LeakyReLU, adapting to complex feature distributions). The third layer maps the 256-dimensional features to a 2D output and uses Softmax to obtain the predicted probabilities of positive (effective inhibitors) and negative (ineffective inhibitors). Loss function: The binary cross-entropy (BCE) loss is used to measure the difference between the predicted probability and the true label, and to guide the updating of model parameters.

[0069] S106, Model Training and Validation;

[0070] The balanced dataset is divided into training, validation, and test sets in an 8:1:1 ratio. Training parameters are set, and the model is considered complete when the area under the curve (AUC) is ≥95%. This step ensures that the model performance meets the target through a reasonable training strategy and comprehensive metric validation, as detailed below:

[0071] Data partitioning: Dataset A is divided in an 8:1:1 ratio: training set (15876 samples) for model parameter learning, validation set (1985 samples) for hyperparameter tuning, and test set (1984 samples) for performance evaluation.

[0072] The hardware environment can be: Ubuntu 22.04 system, NVIDIA RTX 4090 GPU (24 GB VRAM), AMD EPYC 9654 CPU (96 cores, 2.40 GHz, ensuring training efficiency on large-scale data); the software environment can be: Python 3.10, PyTorch 2.1.0, RDKit (molecular feature calculation), DSSP (protein secondary structure analysis); training strategy: epochs=100 (avoid underfitting), batch size=32 (balancing memory and training stability); Adam optimizer (adaptive learning rate, fast convergence), initial learning rate 0.001, weight decay 1e-5 (suppress overfitting); linear warming of parameters in the first 10 epochs (increasing from 0.0001 to 0.001 to avoid training oscillation caused by an excessively high initial learning rate), followed by ReduceLROnPlateau scheduling (if there is no improvement in AUC on the validation set after 5 epochs, the learning rate is multiplied by 0.7 to ensure later convergence).

[0073] The model validation process is as follows:

[0074] Evaluation metrics: Six complementary metrics are used to comprehensively evaluate model performance.

[0075] Accuracy: Overall classification accuracy, reflecting the model's basic performance;

[0076] Precision: Predicts the proportion of true positives among positive samples, avoiding the waste of screening costs caused by false positives;

[0077] Recall: The proportion of predicted positive samples among actual positive samples, to avoid missing effective inhibitors;

[0078] F1 score: harmonic mean of precision and recall, balancing the trade-off between the two;

[0079] MCC: The correlation coefficient of the four types of prediction results, suitable for evaluating imbalanced data;

[0080] AUC: Area under the ROC curve, a threshold-independent metric that reflects the model's discriminative ability;

[0081] Acceptance criteria: When the AUC on the test set is ≥ 95%, the model training is considered complete; otherwise, return to step 105, adjust parameters such as the number of attention heads and the number of MLP layers, and retrain. The model of this invention achieves an AUC of 97.3% on Dataset A, which is significantly better than existing models (such as MolTrans's 93.2%).

[0082] S107, screening and validation of acetylcholinesterase inhibitors.

[0083] 1) Candidate molecule screening: Small molecules were extracted from the ZINC database, preprocessed as inhibitor molecules, and then input into the trained model. Molecules with a prediction probability ≥ 0.97 were selected as candidates. Specifically, the molecular library source was as follows: 51,220 small molecules were downloaded from the ZINC database, which contains tens of millions of drug-like small molecules and is a commonly used library for drug screening; Preprocessing: Molecules that did not meet the requirements were removed according to the criteria in step 102 (molecular weight ≤ 500 Da, ≤ 3 ring structures, and contain active pharmacophores), resulting in 38,760 candidate molecules; Model prediction: The preprocessed molecules were input into the trained model, and molecules with a prediction probability ≥ 0.97 (corresponding to a positive prediction accuracy of 99.2% on Dataset A test set) were selected, resulting in 405 high-confidence candidates.

[0084] 2) Molecular docking verification: Using the CB-Dock2 platform, with the human acetylcholinesterase active pocket as the docking center, the Vina fraction was calculated, and molecules with a Vina fraction ≤ -9 kcal / mol were screened as high-affinity acetylcholinesterase candidates. Specifically, the docking platform used was CB-Dock2, which supports automated protein-ligand docking with accuracy validated across multiple datasets. The docking setup was centered on the active pocket of human AChE (UniProt ID: P22303) (coordinates: x=10.2 Å, y=25.6 Å, z=18.8 Å), with a grid size of 20×20×20 Å, covering key residues in the active pocket (such as Ser203 and His447). Affinity assessment used the Vina score (a commonly used indicator of binding affinity in molecular docking; a smaller value indicates stronger affinity), screening for molecules with a Vina Score ≤ -9 kcal / mol. Three optimal candidates were ultimately selected as high-affinity acetylcholinesterase candidates, with Vina scores of -12.8 kcal / mol (optimal), -10.4 kcal / mol, and -10.2 kcal / mol, respectively. These candidates are compatible with AChE. The active pockets form hydrogen bonds (such as with Ser203), hydrophobic interactions (such as with Trp86), and π-π stacking (such as with Tyr337), providing reliable targets for subsequent experimental verification.

[0085] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for screening acetylcholinesterase inhibitors based on neural networks, characterized in that, The method includes: Step 1: Construct a multi-species dataset, including acetylcholinesterase data and corresponding inhibitor molecule data for 6 species; the species include Homo sapiens, electric eel, house mouse, cattle, California electric ray, and brown rat; the inhibitor molecule data comes from the BindingDB database, and the acetylcholinesterase data comes from the AlphaFold2 database; Step 2, Dataset Preprocessing and Sample Labeling, includes: 1) Inhibitor molecule pretreatment: Molecules with molecular weight > 500 Da or containing more than 3 ring structures were removed, and at least one active pharmacophore was retained; IC50 ≤ 1 μM was defined as a positive sample and IC50 > 1 μM as a negative sample, and positive and negative samples were randomly selected in a 1:1 ratio to construct a balanced dataset. 2) Acetylcholinesterase protein pretreatment: For acetylcholinesterase with amino acid sequences >300 residues, the key region around the active pocket was extracted; based on the 3D coordinates provided by AlphaFold2, the Euclidean distance between residue pairs was calculated, and edges were added when the distance was <8.0 Å to construct a residue contact map; 3) Construction of external validation set: Data on acetylcholinesterase inhibitors of the six species were extracted from the ChEMBL database, and an external validation set was constructed using the labeling criteria described in the inhibitor molecule preprocessing. Step 3: Construct a graph-structured data representation, including: 1) Inhibitor molecule graph representation: Atoms in the molecule are used as nodes, and chemical bonds are used as edges; each atomic node is encoded as a 105-dimensional feature vector. The formula is: ,in, 100-dimensional one-heat encoding for atomic types, The charge of the atom. These are the mean and standard deviation of the charge, respectively. For van der Waals radius, Maximum radius For the minimum radius, It is an aromatic binary indicator. For chiral descriptors; 2) Acetylcholinesterase protein diagram representation: Acetylcholinesterase residues are used as nodes, and residue contact relationships are used as edges; each residue node is encoded as a 24-dimensional feature vector. The formula is: ,in, It is a 20-dimensional unique heat encoding of amino acid types. The probability of secondary structure comes from the DSSP algorithm. It is a binary indicator of solvent accessibility, and is 1 when the solvent accessibility ASA is ≥25%; Step 4: Multi-scale feature extraction, including: 1) Inhibitor molecule feature extraction: A three-level dilated graph convolution based on the SAGE network was used, with a dilation factor of 2. k k is the number of layers. Each layer extracts features through mean pooling and non-linear activation, as shown in the formula: ,in, This is the weight matrix. dilation=2 k The time neighborhood is then used; the atomic features are then concatenated and projected into 512-dimensional global molecular features through 4-head attention pooling. ; 2) Acetylcholinesterase protein feature extraction: A two-layer graph attention network is used, with each layer containing 4 independent attention heads. Attention scores and feature aggregation are calculated. The outputs of the 4 attention heads are concatenated into 128-dimensional features. After being connected to the residuals by GraphNorm, global average pooling is used to obtain 512-dimensional global protein features. ; Step 5, Feature Fusion and Binding Affinity Prediction, includes: 1) Bidirectional cross-attention fusion: This integrates the global molecular features... With the global protein features Projected as query (Q), key (K), and value (V) vectors respectively, attention weights are calculated and features are fused in both the "protein→ligand" and "ligand→protein" directions, as shown in the formulas: ， ,in, These are protein lookup, ligand bond, and value vector, respectively. The projection matrix is ​​used; after fusion, a 1024-dimensional joint feature is obtained. ; 2) Affinity prediction: The joint features are processed using a multilayer perceptron. First, it is compressed to 512 dimensions, then projected to 256 dimensions, and finally the predicted probability is output through Softmax. ; Training uses binary cross-entropy loss, the formula of which is: in, For real labels, To predict the probability, N is the number of samples; Step 6, Model Training and Validation, including: The balanced dataset is divided into training, validation, and test sets in a ratio of 8:1:1; training parameters are set, and the model is considered to be trained when the area under the curve is ≥95%. Step 7: Screening and validation of acetylcholinesterase inhibitors, including: 1) Candidate molecule screening: Small molecules are extracted from the ZINC database, preprocessed with the inhibitor molecules, and then input into the trained model. Molecules with a prediction probability ≥ 0.97 are selected as candidates. 2) Molecular docking verification: Using the CB-Dock2 platform, with the human acetylcholinesterase active pocket as the docking center, the Vina fraction was calculated, and molecules with a Vina fraction ≤ -9 kcal / mol were screened as high-affinity acetylcholinesterase candidates.

2. The method according to claim 1, characterized in that, The acetylcholinesterase data for the six species are as follows: 7274 for Homo sapiens, 5289 for electric eel, 530 for house mouse, 303 for bovine, 249 for California electric ray, and 2033 for brown rat. The balanced dataset contains 19845 samples, of which 9974 are positive and 9871 are negative. The external validation set contains 6710 samples, of which 3535 are positive and 3175 are negative.

3. The method according to claim 1, characterized in that, The neighborhood ranges of the three-level expanded graph convolution in step four are as follows: the first layer (k=0) neighborhood is directly bonded atoms, the second layer (k=1) neighborhood is 2 bond-distance atoms, and the third layer (k=2) neighborhood is 4 bond-distance atoms.

4. The method according to claim 1, characterized in that, In step five, the number of attention heads in the bidirectional cross-attention is 4, and the feature dimension of each attention head is 128; the dropout rate of the multilayer perceptron is 0.

2.

5. The method according to claim 1, characterized in that, The metrics used for model validation in step six include accuracy, precision, recall, F1 score, Matthews correlation coefficient, and area under the ROC curve.

6. The method according to claim 1, characterized in that, The docking center mentioned in step seven is consistent with the active pocket region mentioned in step two.

7. The method according to claim 1, characterized in that, After the attention head outputs of the graph attention network described in step four are concatenated, they are converted into 128-dimensional features through a linear projection matrix, and then connected to the residuals via GraphNorm.

8. The method according to claim 1, characterized in that, The joint features described in step five Molecular-level representations are obtained through global average pooling.

9. The method according to claim 1, characterized in that, In step two, when constructing the residue contact diagram, the Euclidean distance calculation is based on the Cα atom coordinates provided by AlphaFold2.

10. The method according to claim 1, characterized in that, The solvent accessibility mentioned in step three is determined by the relative accessible surface area threshold of 25%, calculated by the ASAcalc tool.

Images