Molecular fingerprint vector similarity determination method and device, equipment and medium

By converting the molecular fingerprint vector into a matrix and calculating the matching statistics matrix, the problem of inefficient similarity calculation of molecular fingerprint vectors in the prior art is solved, and efficient similarity calculation and fast search and screening are realized.

CN119091999APending Publication Date: 2024-12-06ANALYTICAL BIOSCIENCES LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411271301.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, the molecular fingerprint vector similarity calculation efficiency is low, especially when the data volume of the molecular fingerprint vector and molecular fingerprint vector to be processed is large, resulting in low search and screening efficiency.

Method used

By converting the reference molecular fingerprint vector and the molecular fingerprint vector to a matrix, the matching statistics matrix is ​​calculated, and the similarity is determined, and the conversion cycle is calculated as a matrix operation to improve efficiency.

Benefits of technology

The efficiency of molecular fingerprint vector similarity calculation is greatly improved, the calculation time is reduced, and the efficiency of search and screening is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091999A_ABST
    Figure CN119091999A_ABST
Patent Text Reader

Abstract

The invention discloses a molecular fingerprint vector similarity determination method and device, equipment and a medium, and the method comprises the steps: determining a reference matrix matched with a preset reference molecular fingerprint vector, and determining a to-be-processed matrix matched with a to-be-processed molecular fingerprint vector; wherein the number of the reference molecular fingerprint vectors and / or the number of the to-be-processed molecular fingerprint vectors are / is at least two, each reference molecular fingerprint vector comprises at least two vector elements, and the number of the vector elements of the to-be-processed molecular fingerprint vectors is the same as the number of the vector elements of the reference molecular fingerprint vectors; determining a matching statistic matrix according to the reference matrix and the to-be-processed matrix; and determining the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic matrix. According to the invention, the efficiency of molecular fingerprint vector similarity calculation can be improved, and the calculation process of molecular fingerprint similarity is accelerated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, device, equipment and medium for determining the similarity of molecular fingerprint vectors. Background Art

[0002] Similarity is used to measure the degree of difference between two vectors. The calculation of similarity plays a vital role in the retrieval and screening of molecular fingerprint vectors. For the molecular fingerprint vector to be processed, the similarity between it and each molecular fingerprint vector in the molecular fingerprint vector database is calculated one by one, and each molecular fingerprint vector in the molecular fingerprint vector database is sorted according to the similarity, so as to find the molecular fingerprint vector in the molecular fingerprint vector database that is most similar to the molecular fingerprint vector to be processed, so as to realize the search and screening of the molecular fingerprint vector to be processed.

[0003] However, the molecular fingerprint vector similarity calculation method in the prior art requires cyclically performing similarity calculations on the molecular fingerprint vector to be processed and each molecular fingerprint vector in the molecular fingerprint vector database. When the data volume and number of the molecular fingerprint vectors to be processed are relatively large, the similarity calculation will be very slow, resulting in low efficiency in searching and screening molecular fingerprint vectors. Summary of the invention

[0004] The present invention provides a method, device, equipment and medium for determining the similarity of molecular fingerprint vectors, so as to improve the efficiency of molecular fingerprint vector similarity calculation and accelerate the calculation process of molecular fingerprint vector similarity.

[0005] In a first aspect, an embodiment of the present invention provides a method for determining the similarity of a molecular fingerprint vector, the method comprising:

[0006] Determine a reference matrix that matches a preset reference molecular fingerprint vector, and determine a matrix to be processed that matches a molecular fingerprint vector to be processed;

[0007] The number of the reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as the number of vector elements of the reference molecular fingerprint vector;

[0008] Determine a matching statistic matrix according to a reference matrix and a matrix to be processed;

[0009] According to the matching statistic matrix, the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector is determined.

[0010] In a second aspect, an embodiment of the present invention further provides a device for determining the similarity of a molecular fingerprint vector, the device comprising:

[0011] A matrix determination module, used to determine a reference matrix matching a preset reference molecular fingerprint vector, and to determine a matrix to be processed matching a molecular fingerprint vector to be processed;

[0012] The number of the reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as the number of vector elements of the reference molecular fingerprint vector;

[0013] A matching statistic matrix determination module is used to determine a matching statistic matrix according to a reference matrix and a matrix to be processed;

[0014] The similarity determination module is used to determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic matrix.

[0015] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for determining the similarity of molecular fingerprint vectors as described in any one of the embodiments of the present invention is implemented.

[0016] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to perform a method for determining the similarity of molecular fingerprint vectors as described in any one of the embodiments of the present invention.

[0017] The technical solution of the embodiment of the present invention is to determine the corresponding matrices for the reference molecular fingerprint vector and the molecular fingerprint vector to be processed respectively when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed, determine the matching statistic matrix according to the reference matrix and the matrix to be processed, and then determine the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistic matrix. The problem of slow and inefficient similarity calculation caused by the cyclic calculation of similarity in the prior art when the amount and number of data of the molecular fingerprint vector to be processed and / or the molecular fingerprint vector to be processed are relatively large is solved. The cyclic operation of similarity calculation is converted into vector and matrix operations, which can greatly improve the efficiency of similarity calculation and accelerate the similarity calculation process of molecular fingerprint vectors.

[0018] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 is a flow chart of a method for determining the similarity of a molecular fingerprint vector provided in the first embodiment of the present invention;

[0021] Figure 2 is a flow chart of another method for determining the similarity of molecular fingerprint vectors provided in the second embodiment of the present invention;

[0022] Figure 3 It is a structural schematic diagram of a device for determining the similarity of molecular fingerprint vectors provided in Embodiment 3 of the present invention;

[0023] Figure 4 It is a structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. In the embodiments of the present application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned, and they should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application, but it does not mean that the applicant has or must use the solution.

[0026] The acquisition, transmission, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0027] Embodiment 1

[0028] Figure 1 A flowchart of a method for determining the similarity of molecular fingerprint vectors is provided for the first embodiment of the present invention. This embodiment is applicable to the case where, when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed, the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector is calculated based on matrix operations. The method can be executed by a device for determining the similarity of molecular fingerprint vectors. The device for determining the similarity of molecular fingerprint vectors can be implemented in the form of hardware and / or software. The device for determining the similarity of molecular fingerprint vectors can be configured in an electronic device or a server.

[0029] like Figure 1 As shown, the method includes:

[0030] S110, determining a reference matrix that matches a preset reference molecular fingerprint vector, and determining a matrix to be processed that matches a molecular fingerprint vector to be processed.

[0031] The number of reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as that of the reference molecular fingerprint vector.

[0032] This embodiment provides a solution for accelerating the similarity calculation process by matrix operation when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed. It should be noted that the reference molecular fingerprint vector and the molecular fingerprint vector to be processed mentioned in this embodiment are both in the form of binary vectors. At the same time, the number of vector elements of the reference molecular fingerprint vector and the molecular fingerprint vector to be processed is the same, but this embodiment does not limit the specific value of the number of vector elements.

[0033] The molecular fingerprint vector represents the specific properties and structural characteristics of a molecule by converting the structural information of the molecule into a vector consisting of a series of binary codes (0 and 1). The molecular fingerprint vector has a wide range of applications in chemical informatics, drug design and other fields.

[0034] The reference molecular fingerprint vector can be either a molecular fingerprint vector in a pre-set molecular fingerprint vector database or a molecular fingerprint vector obtained by preliminary screening in a molecular fingerprint vector database according to a certain screening algorithm. This embodiment does not limit the source of the reference molecular fingerprint vector. The reference matrix is ​​a matrix constructed with the number of reference molecular fingerprint vectors as rows and the vector elements of the reference molecular fingerprint vectors as columns. The molecular fingerprint vector to be processed refers to the molecular fingerprint vector currently being processed and the similarity between which needs to be determined with the reference molecular fingerprint vector. The matrix to be processed is a matrix constructed with the number of molecular fingerprint vectors to be processed as rows and the vector elements of the molecular fingerprint vector to be processed as columns.

[0035] Specifically, if there are m reference molecular fingerprint vectors and p vector elements, the size of the reference matrix is ​​m×p. If there are n molecular fingerprint vectors to be processed and p vector elements, the size of the matrix to be processed is n×p. In a specific example, the reference molecular fingerprint vector is a vector in the form of 10-bit binary code. There are 2 reference molecular fingerprint vectors, namely x1=(0, 1, 1, 0, 1, 0, 0, 1) and x2=(1, 1, 0, 1, 0, 0, 1, 1), then the reference matrix is

[0036] In this embodiment, when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed, the reference molecular fingerprint vectors and the molecular fingerprint vectors to be processed are converted into matrix forms respectively. The advantage of such a setting is that the operation of cyclically calculating similarities can be converted into matrix operations, thereby accelerating the similarity calculation process and improving the similarity calculation efficiency.

[0037] Furthermore, before S110, the method further includes: determining a reference molecular structure in a preset reference molecular library, and converting the reference molecular structure into a reference molecular fingerprint vector, and determining a molecular structure to be processed, and converting the molecular structure to be processed into a molecular fingerprint vector to be processed.

[0038] In a specific application scenario, this embodiment can be applied to similarity calculation of molecular fingerprint vectors.

[0039] In a specific example, each reference molecule in the reference molecule library can be a molecule of known activity. By converting each reference molecule structure and the structure of the molecule to be processed into a reference molecule fingerprint vector and a molecule fingerprint vector to be processed, and calculating the similarity between the reference molecule fingerprint vector and the molecule fingerprint vector to be processed, molecules with potential activity in the molecules to be processed can be quickly screened out.

[0040] In another specific example, each reference molecule in the reference molecule library can be a molecule with specific structural features. Similarly, by converting each reference molecule structure and the structure of the molecule to be processed into a reference molecule fingerprint vector and a molecule fingerprint vector to be processed, and calculating the similarity between the reference molecule fingerprint vector and the molecule fingerprint vector to be processed, it is possible to quickly query and obtain reference molecules with similar structures to the molecules to be processed, thereby exploring the structure and properties of the molecules to be processed.

[0041] Specifically, converting the reference molecular structure into a reference molecular fingerprint vector, and converting the molecular structure to be processed into a molecular fingerprint vector to be processed can be achieved by the following steps: taking the molecular structure (this embodiment does not limit the data type of the molecular structure) as input, extracting characteristic fragments from the molecular structure, such as atom type, bond type, and ring structure, according to a pre-selected fingerprint type (such as topological fingerprint, etc., this embodiment does not limit the fingerprint type); converting the extracted characteristic fragments into binary codes to obtain a molecular fingerprint vector of fixed length. The process of converting the molecular structure into a molecular fingerprint vector can be achieved by existing conversion tools, and this embodiment does not limit the specific method of converting the molecular structure into a molecular fingerprint vector.

[0042] S120: Determine a matching statistic matrix according to a reference matrix and a matrix to be processed.

[0043] Among them, the matrix elements in the matching statistic matrix are used to represent the statistics of different matching situations between a certain reference molecular fingerprint vector constituting the reference matrix and a certain to-be-processed molecular fingerprint vector constituting the to-be-processed matrix, corresponding to the vector elements of the same digit that meet different matching situations. When the vector elements are in the form of binary code, the matching situations include the following four types: the vector element of the reference molecular fingerprint vector takes 1, and the vector element of the to-be-processed molecular fingerprint vector takes 0; the vector element of the reference molecular fingerprint vector takes 0, and the vector element of the to-be-processed molecular fingerprint vector takes 1; the vector element of the reference molecular fingerprint vector takes 1, and the vector element of the to-be-processed molecular fingerprint vector takes 1; the vector element of the reference molecular fingerprint vector takes 0, and the vector element of the to-be-processed molecular fingerprint vector takes 0. Therefore, when the vector elements are in the form of binary code, the matching statistic matrix can be four, corresponding to the statistics of the above four matching situations respectively. The statistic represents the number of vector elements in each vector element that meet the above matching situations on the same digit of the reference molecular fingerprint vector and the to-be-processed molecular fingerprint vector.

[0044] In a specific example, if the reference molecular fingerprint vector is (0, 1, 1, 0, 1, 0, 0, 1, 0, 1) and the molecular fingerprint vector to be processed is (1, 1, 0, 1, 0, 0, 0, 1, 1), the statistics of the above four matching situations are 3, 4, 2 and 1 respectively.

[0045] In this embodiment, the calculation of matching statistics is realized through matrix operations, thereby realizing the calculation of similarity. The advantage of such a setting is that the matrix operation can maximize the use of hardware resources and greatly improve the efficiency of similarity calculation by packaging multiple calculation operations into a single instruction and utilizing the parallel processing capability of a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), compared with performing cyclic similarity calculations between molecular fingerprint vectors one by one.

[0046] Further, taking the vector elements in binary code form as an example, S120 may further include:

[0047] A1. Performing an inversion operation on each matrix element in the reference matrix to obtain an inverted reference matrix, and performing an inversion operation on each matrix element in the matrix to be processed to obtain an inverted matrix to be processed;

[0048] A2, performing a matrix inner product operation on the reference matrix and the transposed matrix of the inverted matrix to be processed to obtain a first matching statistic matrix;

[0049] A3, performing a matrix inner product operation on the transposed matrix of the inverted reference matrix and the matrix to be processed to obtain a second matching statistic matrix;

[0050] A4, performing a matrix inner product operation on the reference matrix and the transposed matrix of the matrix to be processed to obtain a third matching statistic matrix;

[0051] A5. Perform a matrix inner product operation on the transposed matrix of the negated reference matrix and the negated matrix to be processed to obtain a fourth matching statistic matrix.

[0052] The inversion operation means that if the matrix element is 1, it is converted to 0, and if the matrix element is 0, it is converted to 1. Take it as an example, and perform the inversion operation on its matrix elements to obtain the inverted matrix

[0053] The transposed matrix of the matrix to be processed refers to a matrix obtained by changing the rows of the matrix to be processed into columns and the columns into rows. Similarly, the transposed matrix of the inverse matrix to be processed refers to a matrix obtained by changing the rows of the matrix to be processed into columns and the columns into rows.

[0054] Furthermore, the matching statistics matrix can be expressed by the following formula:

[0055] C=X·Y T ;

[0056] Where X represents an m×p reference matrix, m×p means that the reference matrix consists of m reference molecular fingerprint vectors, each of which includes p vector elements. represents the inverted reference matrix, Y represents the n×p matrix to be processed, n×p represents that the matrix to be processed is composed of n molecular fingerprint vectors to be processed, and each molecular fingerprint vector to be processed includes p vector elements, represents the inverse of the matrix to be processed, A represents the first matching statistic matrix, B represents the second matching statistic matrix, C represents the third matching statistic matrix, D represents the fourth matching statistic matrix, and the capacities of A, B, C and D are all m×n.

[0057] Taking the calculation process of the first matching statistic matrix A as an example, the reference matrix X is an m×p matrix, and the inverse of the matrix to be processed is is an n×p matrix, That is, the transposed matrix of the inverse matrix to be processed is a p×n matrix. According to the vector inner product operation formula, the matrix elements in the first matching statistic matrix A can be calculated by the following formula: Among them, x ik represents the k-th matrix element in the i-th row of the first matching statistic matrix A, Represents the transposed matrix of the inverse matrix to be processed In the jth element of the kth row, i takes the value of 1-m, j takes the value of 1-n, and the final first matching statistic matrix A is an m×n matrix. ij It represents the statistics of the vector elements that meet the first matching situation among the vector elements of the i-th reference molecular fingerprint vector in the reference matrix X and the j-th molecular fingerprint vector to be processed in the matrix to be processed. The first matching situation indicates that the vector elements of the reference molecular fingerprint vector are 1 and the vector elements of the molecular fingerprint vector to be processed are 0.

[0058] S130 , determining the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic matrix.

[0059] In this embodiment, a plurality of reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed are converted into a matrix form, and matrix operations are performed based on the reference matrix and the matrix to be processed to obtain a matching statistic matrix, and a similarity matrix is ​​calculated based on the matching statistic matrix. The advantage of such a setting is that not only can the operation of calculating similarity cyclically be converted into a matrix operation to accelerate the similarity calculation process, but also, the calculation of the similarity matrix based on the matching statistic matrix facilitates the similarity calculation of multiple indicators.

[0060] It is understandable that there are a variety of similarity calculation methods in the prior art. Taking the similarity calculation of molecular fingerprint vectors as an example, it can be realized by calculation methods such as Tanimoto similarity, cosine similarity and Euclidean similarity. Under different similarity calculation methods, there may be different similarity calculation results between the reference molecular fingerprint vector and the molecular fingerprint vector to be processed. Therefore, in order to improve the accuracy and precision of the similarity calculation results, a variety of different similarity calculation methods can be selected to perform the similarity calculation between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector respectively. If the operation of cyclic calculation of similarity in the prior art is still adopted, after selecting different similarity calculation methods, it is necessary to recalculate the similarity. The technical solution of the present application, when selecting a variety of similarity calculation methods for similarity calculation, does not need to be recalculated, but can directly use the intermediate result of the matching statistic matrix for calculation, which further improves the efficiency of similarity calculation.

[0061] Furthermore, S130 may further include: determining a similarity matrix between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic matrix and at least one pre-selected similarity calculation formula.

[0062] In a specific example, taking the similarity matrix calculation based on the Tanimoto similarity method as an example, the similarity calculation formula is: S represents the similarity matrix, the capacity of the similarity matrix is ​​m×n, s ij It represents the similarity between the i-th reference molecular fingerprint vector among the m reference molecular fingerprint vectors of the reference matrix X and the j-th to-be-processed molecular fingerprint vector among the n to-be-processed molecular fingerprint vectors of the to-be-processed matrix Y. ij represents the matrix element of the i-th row and j-th column of the first matching statistic matrix A, that is, the statistics of the vector elements that meet the first matching situation among the vector elements of the i-th reference molecular fingerprint vector of the reference matrix X and the j-th processed molecular fingerprint vector of the processed matrix Y; b ij represents the matrix element of the i-th row and j-th column of the second matching statistic matrix B, that is, the statistics of the vector elements that meet the second matching situation among the vector elements of the i-th reference molecular fingerprint vector of the reference matrix X and the j-th processed molecular fingerprint vector of the processed matrix Y; c ij Represents the matrix element in the i-th row and j-th column of the third matching statistic matrix C, that is, the statistics of the vector elements in the i-th reference molecular fingerprint vector of the reference matrix X and the j-th processed molecular fingerprint vector of the processed matrix Y that meet the third matching situation.

[0063] In another specific example, taking the similarity matrix calculation based on the cosine similarity method as an example, the similarity calculation formula is:

[0064] Table 1 provides a schematic table of calculation formulas under different similarity indicators when calculating a similarity matrix. As shown in Table 1, different common indicators that can be used to calculate similarity and their corresponding calculation formulas are shown.

[0065] Table 1

[0066]

[0067] This embodiment does not limit the similarity calculation method and its corresponding similarity calculation formula, which can be flexibly selected according to the actual needs of vector similarity calculation, including but not limited to the calculation formulas under different similarity indicators provided in Table 1 above, and one or more can be selected according to actual needs.

[0068] The technical solution of the embodiment of the present invention is to determine the corresponding matrices for the reference molecular fingerprint vector and the molecular fingerprint vector to be processed respectively when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed, determine the matching statistic matrix according to the reference matrix and the matrix to be processed, and then determine the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistic matrix. The problem of slow and inefficient similarity calculation caused by the cyclic calculation of similarity in the prior art when the amount and number of data of the molecular fingerprint vector to be processed and / or the molecular fingerprint vector to be processed are relatively large is solved, and the cyclic operation of similarity calculation is converted into vector and matrix operations, which can greatly improve the efficiency of similarity calculation.

[0069] Embodiment 2

[0070] Figure 2 A flowchart of another method for determining the similarity of molecular fingerprint vectors provided in Embodiment 2 of the present invention. This embodiment is applicable to the case where the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector is calculated based on vector operations when the reference molecular fingerprint vector and the molecular fingerprint vector to be processed are both the same. The method can be executed by a device for determining the similarity of molecular fingerprint vectors. The device for determining the similarity of molecular fingerprint vectors can be implemented in the form of hardware and / or software. The device for determining the similarity of molecular fingerprint vectors can be configured in an electronic device or a server.

[0071] like Figure 2 As shown, the method includes:

[0072] S210: If it is determined that the number of the reference molecular fingerprint vector and the number of the molecular fingerprint vector to be processed are both one, then a matching statistic is determined according to the reference molecular fingerprint vector and the molecular fingerprint vector to be processed.

[0073] This embodiment provides a solution for accelerating the similarity calculation process by vector operation when the reference molecular fingerprint vector and the molecular fingerprint vector to be processed are both the same. Similarly, the reference molecular fingerprint vector and the molecular fingerprint vector to be processed mentioned in this embodiment are both in the form of binary vectors, and at the same time, the reference molecular fingerprint vector and the molecular fingerprint vector to be processed have the same number of vector elements. The sources of the reference molecular fingerprint vector and the molecular fingerprint vector to be processed and the vector conversion process have been described in the above embodiment, and will not be repeated in this embodiment.

[0074] Specifically, before S210, the process may further include: determining a reference molecular structure in a preset reference molecular library, and converting the reference molecular structure into a reference molecular fingerprint vector, and determining a molecular structure to be processed, and converting the molecular structure to be processed into a molecular fingerprint vector to be processed.

[0075] The matching statistic refers to the number of vector elements corresponding to the same digit that meet different matching situations between a reference molecular fingerprint vector constituting the reference matrix and a processed molecular fingerprint vector constituting the processed matrix.

[0076] In this embodiment, when the reference molecular fingerprint vector and the to-be-processed molecular fingerprint vector are both the same, vector operations are performed to replace the operation of calculating similarities in a loop in the prior art, thereby speeding up the similarity calculation process and improving the similarity calculation efficiency.

[0077] Further, taking the vector elements in binary code form as an example, S210 may further include:

[0078] B1, performing a negation operation on each vector element in the reference molecular fingerprint vector to obtain a negated reference molecular fingerprint vector, and performing a negation operation on each vector element in the to-be-processed molecular fingerprint vector to obtain a negated to-be-processed molecular fingerprint vector;

[0079] B2, performing a vector inner product operation on the reference molecular fingerprint vector and the transposed vector of the inverted molecular fingerprint vector to be processed to obtain a first matching statistic;

[0080] B3, performing a vector inner product operation on the transposed vector of the inverted reference molecular fingerprint vector and the molecular fingerprint vector to be processed to obtain a second matching statistic;

[0081] B4, performing a vector inner product operation on the reference molecular fingerprint vector and the transposed vector of the molecular fingerprint vector to be processed to obtain a third matching statistic;

[0082] B5. Perform a vector inner product operation on the transposed vector of the negated reference molecular fingerprint vector and the negated molecular fingerprint vector to be processed to obtain a fourth matching statistic.

[0083] The negation operation has been described in the above embodiment, and will not be repeated in this embodiment. Vector transposition refers to converting a row vector into a column vector, and vector inner product operation refers to multiplying two vectors corresponding to the same bit number and then adding them.

[0084] The first matching statistic represents the statistics of vector elements that meet the first matching situation, that is, the statistics that the vector elements of the reference molecular fingerprint vector are 1 and the vector elements of the molecular fingerprint vector to be processed are 0; the second matching statistic represents the statistics of vector elements that meet the second matching situation, that is, the vector elements of the reference molecular fingerprint vector are 0 and the vector elements of the molecular fingerprint vector to be processed are 1; the third matching statistic represents the statistics of vector elements that meet the third matching situation, that is, the vector elements of the reference molecular fingerprint vector are 1 and the vector elements of the molecular fingerprint vector to be processed are 1; the fourth matching statistic represents the statistics of vector elements that meet the fourth matching situation, that is, the vector elements of the reference molecular fingerprint vector are 0 and the vector elements of the molecular fingerprint vector to be processed are 0.

[0085] Furthermore, the matching statistics can be expressed by the following formula:

[0086] c=x·y T ;

[0087] Where x represents the reference molecular fingerprint vector, represents the inverted reference molecular fingerprint vector, y represents the molecular fingerprint vector to be processed, represents the inversion of the molecular fingerprint vector to be processed, a represents the first matching statistic, b represents the second matching statistic, c represents the third matching statistic, and d represents the fourth matching statistic.

[0088] Taking the calculation process of the first matching statistic a as an example, according to the calculation formula of the vector inner product operation, accumulation is performed only when the vector element in the reference molecular fingerprint vector is 1 and the vector element of the same bit number of the inverse molecular fingerprint vector to be processed is also 1, and the vector element of the inverse molecular fingerprint vector to be processed is obtained by inverting the vector element of the molecular fingerprint vector to be processed. Therefore, the calculated a can be used to represent the statistic that the vector element of the reference molecular fingerprint vector is 1 and the vector element of the molecular fingerprint vector to be processed is 0.

[0089] S220: Determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistics.

[0090] Similarly, vector operations are performed based on the reference molecular fingerprint vector and the molecular fingerprint vector to be processed to determine the matching statistics, and the similarity is calculated based on the matching statistics. The advantage of this setting is that it can not only speed up the similarity calculation process, but also facilitate the similarity calculation of multiple indicators by calculating the similarity based on the matching statistics.

[0091] Furthermore, S220 may further include: determining the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic and at least one pre-selected similarity calculation formula.

[0092] The above embodiment has described the similarity calculation method and its corresponding calculation formula, which will not be described in detail in this embodiment.

[0093] In a specific example, taking the similarity calculation based on Tanimoto similarity as an example, the similarity calculation formula is: In another specific example, taking the similarity calculation based on the cosine similarity method as an example, the similarity calculation formula is:

[0094] Table 2 provides a schematic table of calculation formulas under different similarity indicators when calculating vector similarity. As shown in Table 2, different indicators that can be commonly used to calculate vector similarity and their corresponding calculation formulas are shown.

[0095] Table 2

[0096]

[0097]

[0098] This embodiment does not limit the similarity calculation method and its corresponding similarity calculation formula, which can be flexibly selected according to the actual needs of vector similarity calculation, including but not limited to the calculation formulas under different similarity indicators provided in Table 2 above, and one or more can be selected according to actual needs.

[0099] The technical solution of the embodiment of the present invention is to calculate the matching statistics according to the reference molecular fingerprint vector and the molecular fingerprint vector to be processed when the number of the reference molecular fingerprint vector and the molecular fingerprint vector to be processed is one, and then determine the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistics. Calculating the similarity based on the vector operation method can accelerate the calculation of the matching statistics, thereby improving the calculation efficiency of the similarity. At the same time, similarity calculation is performed based on the intermediate result of the matching statistics, which facilitates the realization of multi-index similarity calculation.

[0100] Embodiment 3

[0101] Figure 3 This is a schematic diagram of a structure of a device for determining the similarity of a molecular fingerprint vector provided in the third embodiment of the present invention. Figure 3 As shown, the device comprises:

[0102] The matrix determination module 310 is used to determine a reference matrix that matches a preset reference molecular fingerprint vector, and to determine a matrix to be processed that matches a molecular fingerprint vector to be processed;

[0103] The number of the reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as the number of vector elements of the reference molecular fingerprint vector;

[0104] A matching statistic matrix determination module 320 is used to determine a matching statistic matrix according to a reference matrix and a matrix to be processed;

[0105] The similarity determination module 330 is used to determine the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistic matrix.

[0106] The technical solution of the embodiment of the present invention is to determine the corresponding matrices for the reference molecular fingerprint vector and the molecular fingerprint vector to be processed respectively when there are multiple reference molecular fingerprint vectors and / or molecular fingerprint vectors to be processed, determine the matching statistic matrix according to the reference matrix and the matrix to be processed, and then determine the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistic matrix. The problem of slow and inefficient similarity calculation caused by the cyclic calculation of similarity in the prior art when the amount and number of data of the molecular fingerprint vector to be processed and / or the molecular fingerprint vector to be processed are relatively large is solved, and the cyclic operation of similarity calculation is converted into vector and matrix operations, which can greatly improve the efficiency of similarity calculation.

[0107] Based on the above embodiment, optionally, the vector elements are in binary code form, and the matching statistic matrix determination module 320 includes:

[0108] A matrix inversion unit is used to perform an inversion operation on each matrix element in a reference matrix to obtain an inverted reference matrix, and to perform an inversion operation on each matrix element in a to-be-processed matrix to obtain an inverted to-be-processed matrix;

[0109] A first matrix inner product operation unit is used to perform a matrix inner product operation on a reference matrix and a transposed matrix of the inverted matrix to be processed to obtain a first matching statistic matrix;

[0110] A second matrix inner product operation unit is used to perform a matrix inner product operation on the transposed matrix of the inverted reference matrix and the matrix to be processed to obtain a second matching statistic matrix;

[0111] A third matrix inner product operation unit is used to perform a matrix inner product operation on the reference matrix and the transposed matrix of the matrix to be processed to obtain a third matching statistic matrix;

[0112] The fourth matrix inner product operation unit is used to perform a matrix inner product operation on the transposed matrix of the negated reference matrix and the negated matrix to be processed to obtain a fourth matching statistic matrix.

[0113] Based on the above embodiment, optionally, the device further includes:

[0114] A matching statistic determination module, used for determining a matching statistic according to the reference molecular fingerprint vector and the molecular fingerprint vector to be processed if it is determined that the number of the reference molecular fingerprint vector and the number of the molecular fingerprint vector to be processed are both one;

[0115] The vector similarity determination module is used to determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistics.

[0116] Based on the above embodiment, the matching statistics determination module includes:

[0117] A vector negation unit is used to negate each vector element in the reference molecular fingerprint vector to obtain a negated reference molecular fingerprint vector, and to negate each vector element in the to-be-processed molecular fingerprint vector to obtain a negated to-be-processed molecular fingerprint vector;

[0118] A first vector inner product operation unit is used to perform a vector inner product operation on a reference molecular fingerprint vector and a transposed vector of the inverted molecular fingerprint vector to be processed to obtain a first matching statistic;

[0119] A second vector inner product operation unit is used to perform a vector inner product operation on the transposed vector of the inverted reference molecular fingerprint vector and the molecular fingerprint vector to be processed to obtain a second matching statistic;

[0120] A third vector inner product operation unit is used to perform a vector inner product operation on the reference molecular fingerprint vector and the transposed vector of the molecular fingerprint vector to be processed to obtain a third matching statistic;

[0121] The fourth vector inner product operation unit is used to perform a vector inner product operation on the transposed vector of the negated reference molecular fingerprint vector and the negated molecular fingerprint vector to be processed to obtain a fourth matching statistic.

[0122] Based on the above embodiment, optionally, the similarity determination module 330 includes:

[0123] A similarity matrix determination unit, used to determine a similarity matrix between the to-be-processed molecular fingerprint vector and a reference molecular fingerprint vector according to a matching statistic matrix and at least one pre-selected similarity calculation formula;

[0124] The vector similarity determination module includes:

[0125] The vector similarity determination unit is used to determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic and at least one pre-selected similarity calculation formula.

[0126] Based on the above embodiment, optionally, the matching statistic matrix is ​​expressed by the following formula:

[0127] C=X·Y T ;

[0128] Where X represents an m×p reference matrix, m×p means that the reference matrix consists of m reference molecular fingerprint vectors, each of which includes p vector elements. represents the inverted reference matrix, Y represents the n×p matrix to be processed, n×p represents that the matrix to be processed is composed of n molecular fingerprint vectors to be processed, and each molecular fingerprint vector to be processed includes p vector elements, represents the inverse of the matrix to be processed, A represents the first matching statistic matrix, B represents the second matching statistic matrix, C represents the third matching statistic matrix, D represents the fourth matching statistic matrix, and the capacities of A, B, C and D are all m×n;

[0129] S represents the similarity matrix, the capacity of the similarity matrix is ​​m×n, S ij Represents the similarity between the i-th reference molecular fingerprint vector among the m reference molecular fingerprint vectors of the reference matrix X and the j-th to-be-processed molecular fingerprint vector among the n to-be-processed molecular fingerprint vectors of the to-be-processed matrix Y;

[0130] The matching statistic is expressed by the following formula:

[0131] c=x·y T ;

[0132] Where x represents the reference molecular fingerprint vector, represents the inverted reference molecular fingerprint vector, y represents the molecular fingerprint vector to be processed, represents the inversion of the molecular fingerprint vector to be processed, a represents the first matching statistic, b represents the second matching statistic, c represents the third matching statistic, and d represents the fourth matching statistic.

[0133] Based on the above embodiment, optionally,

[0134] The molecular fingerprint vector conversion module is used to determine the reference molecular structure in a preset reference molecular library and convert the reference molecular structure into a reference molecular fingerprint vector, and to determine the molecular structure to be processed and convert the molecular structure to be processed into a molecular fingerprint vector to be processed.

[0135] The apparatus for determining the similarity of molecular fingerprint vectors provided in the embodiment of the present invention can execute the method for determining the similarity of molecular fingerprint vectors provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0136] Embodiment 4

[0137] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0138] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0139] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0140] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 performs the various methods and processes described above, such as a method for determining the similarity of a molecular fingerprint vector.

[0141] In some embodiments, the method for determining vector similarity may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining the molecular fingerprint vector similarity described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the method for determining the molecular fingerprint vector similarity in any other appropriate manner (e.g., by means of firmware).

[0142] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0143] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0144] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0145] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0146] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0147] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0148] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0149] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining the similarity of molecular fingerprint vectors, It is characterized in that include: Determine a reference matrix that matches a preset reference molecular fingerprint vector, and determine a matrix to be processed that matches a molecular fingerprint vector to be processed; The number of the reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as the number of vector elements of the reference molecular fingerprint vector; Determine a matching statistic matrix according to a reference matrix and a matrix to be processed; According to the matching statistic matrix, the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector is determined.

2. The method according to claim 1, It is characterized in that The vector elements are in binary code form. According to the reference matrix and the matrix to be processed, a matching statistic matrix is ​​determined, including: Performing an inversion operation on each matrix element in the reference matrix to obtain an inverted reference matrix, and performing an inversion operation on each matrix element in the matrix to be processed to obtain an inverted matrix to be processed; Performing a matrix inner product operation on the reference matrix and the transposed matrix of the inverted matrix to be processed to obtain a first matching statistic matrix; Performing a matrix inner product operation on the transposed matrix of the inverted reference matrix and the matrix to be processed to obtain a second matching statistic matrix; Performing a matrix inner product operation on the reference matrix and the transposed matrix of the matrix to be processed to obtain a third matching statistic matrix; A matrix inner product operation is performed on the transposed matrix of the negated reference matrix and the negated matrix to be processed to obtain a fourth matching statistic matrix.

3. The method according to claim 1, It is characterized in that The method further comprises: If it is determined that the number of the reference molecular fingerprint vector and the number of the molecular fingerprint vector to be processed are both one, then determining a matching statistic according to the reference molecular fingerprint vector and the molecular fingerprint vector to be processed; According to the matching statistics, the similarity between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector is determined.

4. The method according to claim 3, It is characterized in that According to the reference molecular fingerprint vector and the molecular fingerprint vector to be processed, the matching statistics are determined, including: Performing a negation operation on each vector element in the reference molecular fingerprint vector to obtain a negated reference molecular fingerprint vector, and performing a negation operation on each vector element in the to-be-processed molecular fingerprint vector to obtain a negated to-be-processed molecular fingerprint vector; Performing a vector inner product operation on the reference molecular fingerprint vector and the transposed vector of the inverted molecular fingerprint vector to be processed to obtain a first matching statistic; Performing a vector inner product operation on the transposed vector of the inverted reference molecular fingerprint vector and the molecular fingerprint vector to be processed to obtain a second matching statistic; Performing a vector inner product operation on the reference molecular fingerprint vector and the transposed vector of the molecular fingerprint vector to be processed to obtain a third matching statistic; A vector inner product operation is performed on the transposed vector of the negated reference molecular fingerprint vector and the negated molecular fingerprint vector to be processed to obtain a fourth matching statistic.

5. The method according to claim 1 or 3, It is characterized in that According to the matching statistic matrix, the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector is determined, including: Determine a similarity matrix between the molecular fingerprint vector to be processed and the reference molecular fingerprint vector according to the matching statistic matrix and at least one pre-selected similarity calculation formula; Determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistics, including: The similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector is determined according to the matching statistic and at least one pre-selected similarity calculation formula.

6. The method according to claim 5, It is characterized in that The matching statistics matrix is ​​expressed by the following formula: C=X·Y T ; Where X represents an m×p reference matrix, m×p means that the reference matrix consists of m reference molecular fingerprint vectors, each of which includes p vector elements. represents the inverted reference matrix, Y represents the n×p matrix to be processed, n×p represents that the matrix to be processed is composed of n molecular fingerprint vectors to be processed, and each molecular fingerprint vector to be processed includes p vector elements, represents the inverse of the matrix to be processed, A represents the first matching statistic matrix, B represents the second matching statistic matrix, C represents the third matching statistic matrix, D represents the fourth matching statistic matrix, and the capacities of A, B, C and D are all m×n; S represents the similarity matrix, the capacity of the similarity matrix is ​​m×n, S ij Represents the similarity between the i-th reference molecular fingerprint vector among the m reference molecular fingerprint vectors of the reference matrix X and the j-th to-be-processed molecular fingerprint vector among the n to-be-processed molecular fingerprint vectors of the to-be-processed matrix Y; The matching statistic is expressed by the following formula: c=x·y T ; Where x represents the reference molecular fingerprint vector, represents the inverted reference molecular fingerprint vector, y represents the molecular fingerprint vector to be processed, represents the inversion of the molecular fingerprint vector to be processed, a represents the first matching statistic, b represents the second matching statistic, c represents the third matching statistic, and d represents the fourth matching statistic.

7. The method according to claim 1, It is characterized in that Before determining the reference matrix matching the preset reference molecular fingerprint vector, it also includes: A reference molecular structure is determined in a preset reference molecular library, and the reference molecular structure is converted into a reference molecular fingerprint vector; and a molecular structure to be processed is determined, and the molecular structure to be processed is converted into a molecular fingerprint vector to be processed.

8. A device for determining the similarity of molecular fingerprint vectors, It is characterized in that include: A matrix determination module, used to determine a reference matrix matching a preset reference molecular fingerprint vector, and to determine a matrix to be processed matching a molecular fingerprint vector to be processed; The number of the reference molecular fingerprint vectors and / or the molecular fingerprint vectors to be processed is at least two, the reference molecular fingerprint vector includes at least two vector elements, and the number of vector elements of the molecular fingerprint vector to be processed is the same as the number of vector elements of the reference molecular fingerprint vector; A matching statistic matrix determination module is used to determine a matching statistic matrix according to a reference matrix and a matrix to be processed; The similarity determination module is used to determine the similarity between the to-be-processed molecular fingerprint vector and the reference molecular fingerprint vector according to the matching statistic matrix.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the program, the method for determining the molecular fingerprint vector similarity as described in any one of claims 1-7 is implemented.

10. A storage medium storing computer executable instructions, It is characterized in that The computer executable instructions are used to perform the method for determining the similarity of molecular fingerprint vectors as claimed in any one of claims 1 to 7 when executed by a computer processor.

Citation Information

Cited By

  • Non-detention management and control data analysis method and management and control platform

    CN119720103A