Gene transcription regulation mechanism prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a gene transcription regulation mechanism prediction method and device, electronic equipment and a storage medium, and relates to the technical field of biology, the method comprises the following steps: obtaining transcription factors of different species and target genes corresponding to the transcription factors; calculating a regulation capability parameter matrix of each transcription factor and the target gene pair; each transcription factor and the corresponding target gene serve as a transcription factor and target gene pair; determining a transcription factor expression matrix and a target gene expression matrix according to the regulation capability parameter matrix; constructing a linear relation between the two expression matrixes; a plurality of transcription factors are selected to minimize the difference value of the linear relation, and then a sparse optimization problem is constructed; searching a sparse solution according to the sparse optimization problem, and determining transcriptional regulation networks of different species according to the sparse solution; and predicting a gene transcriptional regulation mechanism of the corresponding species according to each transcriptional regulation network. The transcriptional regulation network is determined according to the linear relation between the transcription factor and the target gene, and then the gene transcriptional regulation mechanism is predicted.

Description

Technical Field

[0001] The present application relates to the field of biological technology, and in particular to a method, device, electronic device and storage medium for predicting gene transcription regulation mechanism. Background Art

[0002] Transcriptional regulation refers to the process by which transcription factors (TFs) bind to DNA regulatory elements to activate or inhibit the transcription of their target genes. Epigenetic factors are also important regulatory factors, such as histone modifications, which can control the accessibility of DNA regulatory elements to TFs. The interactions between all TFs, DNA regulatory elements and their target genes form transcriptional regulatory networks (TRNs). Uncovering the regulatory mechanism of TRNs is crucial for understanding the molecular mechanisms of biological processes in various species.

[0003] With the development of high-throughput sequencing technology, more and more genomes, transcriptomes and functional genomics data of species have been sequenced, which provides key basic data for further exploration of TRNs. However, how to reveal the transcriptional regulatory mechanisms (TRMs) of species from these multi-omics data is still a huge challenge. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a gene transcription regulation mechanism prediction method, device, electronic device and storage medium to predict the gene transcription regulation mechanism.

[0005] To achieve the above purpose, one aspect of the embodiments of the present application proposes a method for predicting gene transcription regulation mechanism, the method comprising the following steps:

[0006] Obtaining transcription factors of different species and target genes corresponding to the transcription factors;

[0007] Calculating the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene is regarded as a transcription factor and target gene pair;

[0008] Determine a transcription factor expression matrix and a target gene expression matrix according to the regulatory ability parameter matrix;

[0009] Constructing a linear relationship between the transcription factor expression matrix and the target gene expression matrix;

[0010] Selecting a number of the transcription factors to minimize the difference in the linear relationship, thereby constructing a sparse optimization problem;

[0011] Searching for a sparse solution according to the sparse optimization problem, and then determining the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factor, the DNA regulatory element and the target gene corresponding to the transcription factor;

[0012] The gene transcriptional regulatory mechanism corresponding to the species is predicted based on each of the transcriptional regulatory networks.

[0013] In some embodiments, the step of obtaining transcription factors of different species and target genes corresponding to the transcription factors comprises the following steps:

[0014] Obtain the transcription factors and binding motifs of the transcription factors in different tissues and developmental stages of model organisms and non-model organisms from public databases;

[0015] identifying active promoters and enhancers of each of the transcription factors based on transcription-related histone modifications;

[0016] The binding sites of each of the transcription factors are predicted based on the active promoter and the enhancer; wherein the target gene corresponding to each of the transcription factors includes the corresponding binding motif and the binding site.

[0017] In some embodiments, the step of calculating the regulatory ability parameter matrix of each transcription factor and target gene pair comprises the following steps:

[0018] Calculate the regulatory ability parameters of each transcription factor and target gene pair according to the regulatory ability parameter expression;

[0019] Determine the control capability parameter matrix according to each of the control capability parameters;

[0020] The control capability parameter expression is:

[0021]

[0022] Among them, PRS i,j represents the regulatory ability parameter of the i-th transcription factor and the j-th target gene; d l / d0 represents the ratio of the distance from the lth binding site of the ith transcription factor to the transcription start site of the jth target gene to the constant distance; p l is the intensity of the transcription-related histone modification mark, which is used to indicate the accessibility of the DNA regulatory element to the binding of the corresponding transcription factor; n is the number of the target genes, and l is the serial number of the binding site.

[0023] In some embodiments, determining the transcription factor expression matrix and the target gene expression matrix according to the regulatory ability parameter matrix comprises the following steps:

[0024] Merging the transcription factor expression data of all the species in the regulatory ability parameter matrix into the transcription factor expression matrix;

[0025] The target gene expression data of all the species in the regulatory ability parameter matrix are combined into the target gene expression matrix.

[0026] In some embodiments, constructing the linear relationship between the transcription factor expression matrix and the target gene expression matrix comprises the following steps:

[0027] The linear relationship between the transcription factor expression matrix and the target gene expression matrix is ​​constructed as follows:

[0028] AX = B + ε;

[0029] Among them, A∈R m×r , A represents the transcription factor expression matrix; B∈R m×n , B represents the target gene expression matrix; ε∈R m×n , ε represents the error matrix; X∈R r×n , X represents the regulatory matrix; m is the number of the species, r is the number of the transcription factors, and n is the number of the target genes.

[0030] In some embodiments, the selecting of a plurality of the transcription factors to minimize the difference in the linear relationship, thereby constructing a sparse optimization problem, comprises the following steps:

[0031] The transcription factors less than a threshold number are selected to minimize the difference of the linear relationship, and then the sparse optimization problem is constructed as follows:

[0032] min||AX .,j -B .,j ||2;

[0033] st||X .,j ||0≤K;

[0034] Among them, ||·||2 means that the Euclidean norm is ||X .,j ||0 means X .,j The number of non-zeros in ||X .,j ||0 means X .,j The sparsity of is used to represent the number of the i-th transcription factor regulating the j-th target gene.

[0035] In some embodiments, searching for a sparse solution according to the sparse optimization problem comprises the following steps:

[0036] Convert the sparse optimization problem into an L1 regularization problem;

[0037] The conversion to the L1 regularization problem is:

[0038]

[0039] in,

[0040] Searching and obtaining the sparse solution according to the L1 regularization problem;

[0041] Determining the transcriptional regulatory networks of different species according to the sparse solution comprises the following steps:

[0042] Determining the transcriptional regulatory networks of different species according to the sparse solution using a recognition model;

[0043] The expression of the recognition model is:

[0044]

[0045] Among them, A t represents the transcription factor expression matrix of the t-th species, B t represents the target gene expression matrix of the t-th species; PRS t A matrix representing the regulatory ability parameters of all the transcription factor and target gene pairs in the t-th species; X represents the species with the shortest divergence time in the phylogenetic tree among the t-th species; t represents the transcriptional regulatory network in the t-th species; represents the inferred regulatory strength of the i-th transcription factor on the j-th target gene in the t-th species. When it is not 0, the i-th transcription factor is used as a regulatory factor; η is a regulatory parameter used to balance the relative importance of prior knowledge; λ t is the regularization parameter of the t-th species; τ t is the irrelevance penalty parameter of the transcriptional regulatory network of the tth species; k is the number of species, which include model organisms and non-model organisms.

[0046] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application provides a gene transcription regulation mechanism prediction device, the device comprising:

[0047] A data acquisition unit, used to acquire transcription factors of different species and target genes corresponding to the transcription factors;

[0048] A matrix calculation unit, used to calculate the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene are regarded as a transcription factor and target gene pair;

[0049] A matrix determination unit, used to determine a transcription factor expression matrix and a target gene expression matrix according to the regulation capability parameter matrix;

[0050] A linear relationship construction unit, used to construct a linear relationship between the transcription factor expression matrix and the target gene expression matrix;

[0051] A problem construction unit, used for selecting a plurality of the transcription factors to minimize the difference of the linear relationship, thereby constructing a sparse optimization problem;

[0052] A network determination unit, used for searching for a sparse solution according to the sparse optimization problem, and then determining the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factor, the DNA regulatory element and the target gene corresponding to the transcription factor;

[0053] A mechanism prediction unit is used to predict the gene transcriptional regulation mechanism corresponding to the species based on each of the transcriptional regulation networks.

[0054] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.

[0055] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0056] The embodiments of the present application include at least the following beneficial effects:

[0057] The present application can obtain the transcription factors of different species and the target genes corresponding to the transcription factors; calculate the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene are a transcription factor and target gene pair; determine the transcription factor expression matrix and the target gene expression matrix according to the regulatory ability parameter matrix; construct the linear relationship between the transcription factor expression matrix and the target gene expression matrix; select several transcription factors to minimize the difference in the linear relationship, and then construct a sparse optimization problem; search for sparse solutions according to the sparse optimization problem, and then determine the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes transcription factors, DNA regulatory elements and target genes corresponding to the transcription factors; predict the gene transcriptional regulatory mechanism of the corresponding species according to each transcriptional regulatory network. The present application uses the transcription factors and corresponding target genes of different species to reconstruct the linear relationship between the transcription factor expression matrix and the target gene expression matrix, and determines the transcriptional regulatory network according to the linear relationship, and then can determine the gene transcriptional regulatory mechanism, which can be used to reveal the gene transcriptional regulatory mechanism of the species. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 A schematic diagram of a method for predicting a gene transcription regulation mechanism provided in an embodiment of the present application;

[0060] Figure 2 A flow chart of a bioinformatics method for predicting TRN in non-model organisms using multi-omics data provided in the embodiments of the present application;

[0061] Figure 3 Performance diagram of fused pLASSO provided in the examples of this application;

[0062] Figure 4 A schematic diagram of the structure of a gene transcription regulation mechanism prediction device provided in an embodiment of the present application;

[0063] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0065] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0066] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0068] Before describing the embodiments of the present application in detail, some related technologies that may be involved in the embodiments of the present application are first described as follows:

[0069] Transcriptional regulation refers to the process by which transcription factors (TFs) bind to DNA regulatory elements to activate or inhibit the transcription of their target genes. Epigenetic factors are also important regulatory factors, such as histone modifications, which can control the accessibility of DNA regulatory elements to TFs. The interactions between all TFs, DNA regulatory elements and their target genes form transcriptional regulatory networks (TRNs). Uncovering the regulatory mechanisms of TRNs is crucial to understanding the molecular mechanisms of biological processes in various species.

[0070] With the development of high-throughput sequencing technology, more and more genome, transcriptome and functional genomic data of species have been sequenced, which provides key basic data for further exploration of TRNs. However, how to reveal the transcriptional regulatory mechanisms (TRMs) of species from these multi-omics data remains a huge challenge, especially for non-model organisms. Non-model organisms have many characteristics that model organisms do not have, and lack sufficient prior knowledge of gene regulation; the low sequence similarity between phylogenetically distant species hinders the functional annotation of gene regulatory elements; the noise and sequencing bias of multi-omics data may also affect the exploration of TRMs. In order to understand TRMs, it is necessary to identify TFs, DNA regulatory elements and their target genes from these omics data. Currently, there are bioinformatics tools that can predict TFs from the genome, but the research on DNA regulatory elements and TRNs is mainly focused on some model organisms. For non-model organisms, two methods are usually used for research: (1) comparative analysis, detecting conserved non-coding regions of multiple species, which may be functional elements; (2) functional analysis, using high-throughput sequencing technology such as ChIP-seq to detect TF binding signals or transcription-related epigenetic marks to identify active DNA regulatory elements in vivo. However, many studies have shown that conserved non-coding regions detected by comparative analysis are not necessarily functional, and functional DNA regulatory elements are not necessarily conserved. Therefore, although researchers have tried to use the conservation of intergenic sequences to identify DNA regulatory elements in phylogenetically similar species, they may only identify a small number of them with a high false positive rate.

[0071] In recent years, a large number of TFs-DNA interaction groups and epigenomic data from ChIP-seq sequencing have also deepened our understanding of TRNs in model organisms. However, the results of ChIP-seq are often complicated by the noise generated during the immunoprecipitation process; and because the detected DNA regulatory elements only have sequence information, it is not possible to directly determine the interaction between TFs, DNA regulatory elements and downstream targets, because the DNA regulatory elements and target genes may be thousands of bases apart.

[0072] Therefore, some embodiments of the present application provide a bioinformatics solution (fused pLASSO) to solve the above problems and explore the potential TRMs of organisms based on multi-omic data from multiple species or multiple sources. Various types of omics data from different sources can cross-validate each other, reduce false signals between them, and can interpret the complex interactions of TRNs by studying the cross-information between different types of regulatory factors. Although the TRNs of non-model organisms are largely unknown, they can be inferred by considering conserved genes, gene expression, and the interactions between TRNs of well-studied model species and target non-model species. Studies have shown that transcriptional regulatory interactions are more conservative than sequences. For example, in eukaryotes, the binding domains of TFs and DNA and their corresponding TFs binding motifs co-evolve and are highly conserved. Other studies have also shown that in different mammals, the conservation of interactions between TFs, DNA regulatory elements and their target genes (ie, TRNs) is also stronger than sequence conservation.

[0073] Therefore, the inventors of the present application believe that conserved TRNs are helpful in converting the TRMs knowledge of model organisms into the TRMs knowledge of non-model organisms. Therefore, some embodiments of the present application provide a bioinformatics method that combines multi-omics data of model organisms and non-model organisms and utilizes a large amount of knowledge related to TRNs of current model organisms to infer TRNs of non-model organisms.

[0074] The embodiment of the present application provides a gene transcription regulation mechanism prediction method, device, electronic device and storage medium. The technical solution of the present application includes: obtaining transcription factors of different species and target genes corresponding to the transcription factors; calculating the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene are a transcription factor and target gene pair; determining the transcription factor expression matrix and the target gene expression matrix according to the regulatory ability parameter matrix; constructing the linear relationship between the transcription factor expression matrix and the target gene expression matrix; selecting a number of transcription factors to minimize the difference in the linear relationship, and then constructing a sparse optimization problem; searching for a sparse solution according to the sparse optimization problem, and then determining the transcription regulation network of different species according to the sparse solution; wherein the transcription regulation network includes transcription factors, DNA regulatory elements and target genes corresponding to the transcription factors; predicting the gene transcription regulation mechanism of the corresponding species according to each transcription regulation network. The present application uses the transcription factors and corresponding target genes of different species to reconstruct the linear relationship between the transcription factor expression matrix and the target gene expression matrix, and determines the transcription regulation network according to the linear relationship, and then can determine the gene transcription regulation mechanism, which can be used to reveal the gene transcription regulation mechanism of the species.

[0075] The present application embodiment provides a gene transcription regulation mechanism prediction method, device, electronic device and storage medium, and relates to the field of biological technology. The gene transcription regulation mechanism prediction method, device, electronic device and storage medium provided in the present application embodiment can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements a knowledge extraction method, etc., but is not limited to the above forms.

[0076] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0077] Reference Figure 1 , the present application embodiment provides a method for predicting gene transcription regulation mechanism, which may include but is not limited to S100 to S160, as follows:

[0078] S100: Obtaining transcription factors of different species and target genes corresponding to the transcription factors.

[0079] Furthermore, S100 may include S101 to S103:

[0080] S101: Obtain the transcription factors and binding motifs of the transcription factors in different tissues and developmental stages of model organisms and non-model organisms from public databases;

[0081] S102: identifying active promoters and enhancers of each of the transcription factors according to transcription-related histone modifications;

[0082] S103: predicting the binding sites of each of the transcription factors according to the active promoter and the enhancer; wherein the target gene corresponding to each of the transcription factors includes the corresponding binding motif and the binding site.

[0083] S110: Calculate the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene is regarded as a transcription factor and target gene pair.

[0084] Furthermore, S110 may include S111-S112:

[0085] S111: calculating the regulatory ability parameters of each transcription factor and target gene pair according to the regulatory ability parameter expression;

[0086] S112: Determine the control capability parameter matrix according to each of the control capability parameters;

[0087] The control capability parameter expression is:

[0088]

[0089] Among them, PRS i,j represents the regulatory ability parameter of the i-th transcription factor and the j-th target gene; d l / d0 represents the ratio of the distance from the lth binding site of the ith transcription factor to the transcription start site of the jth target gene to the constant distance; p l is the intensity of the transcription-related histone modification mark, which is used to indicate the accessibility of the DNA regulatory element to the binding of the corresponding transcription factor; n is the number of the target genes, and l is the serial number of the binding site.

[0090] S120: Determine a transcription factor expression matrix and a target gene expression matrix according to the regulatory capability parameter matrix.

[0091] Further, S120 may include S121-S122:

[0092] S121: merging the transcription factor expression data of all the species in the regulatory ability parameter matrix into the transcription factor expression matrix;

[0093] S122: merging the target gene expression data of all the species in the regulatory ability parameter matrix into the target gene expression matrix.

[0094] S130: constructing a linear relationship between the transcription factor expression matrix and the target gene expression matrix.

[0095] Further, S130 may include S131:

[0096] S131: Constructing the linear relationship between the transcription factor expression matrix and the target gene expression matrix as follows:

[0097] AX = B + ε;

[0098] Among them, A∈R m×r , A represents the transcription factor expression matrix; B∈R m×n , B represents the target gene expression matrix; ε∈R m×n , ε represents the error matrix; X∈R r×n , X represents the regulatory matrix; m is the number of the species, r is the number of the transcription factors, and n is the number of the target genes.

[0099] S140: Selecting a number of the transcription factors to minimize the difference of the linear relationship, thereby constructing a sparse optimization problem.

[0100] Further, S140 may include the following steps S141:

[0101] S141: Select the transcription factors whose number is less than a threshold value to minimize the difference of the linear relationship, and then construct the sparse optimization problem as follows:

[0102] min||AX .,j -B .,j ||2;

[0103] st||X .,j ||0≤K;

[0104] Among them, ||·||2 means that the Euclidean norm is ||X .,j ||0 means X .,j The number of non-zeros in ||X .,j ||0 means X .,j The sparsity of is used to represent the number of the i-th transcription factor regulating the j-th target gene.

[0105] S150: searching for a sparse solution according to the sparse optimization problem, and then determining the transcriptional regulatory network of the different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factors, DNA regulatory elements and the target genes corresponding to the transcription factors.

[0106] Furthermore, searching for a sparse solution according to the sparse optimization problem in S150 includes the following steps S151-S152:

[0107] S151: Convert the sparse optimization problem into an L1 regularization problem;

[0108] The conversion to the L1 regularization problem is:

[0109]

[0110] in,

[0111] S152: Search and obtain the sparse solution according to the L1 regularization problem.

[0112] Determining the transcriptional regulatory networks of different species according to the sparse solution in S150 includes the following steps S153:

[0113] S153: Determine the transcriptional regulatory network of different species according to the sparse solution using a recognition model;

[0114] The expression of the recognition model is:

[0115]

[0116] Among them, A t represents the transcription factor expression matrix of the t-th species, B t represents the target gene expression matrix of the t-th species; PRS t A matrix representing the regulatory ability parameters of all the transcription factor and target gene pairs in the t-th species; X represents the species with the shortest divergence time in the phylogenetic tree among the t-th species; t represents the transcriptional regulatory network in the t-th species; represents the inferred regulatory strength of the i-th transcription factor on the j-th target gene in the t-th species. When it is not 0, the i-th transcription factor is used as a regulatory factor; η is a regulatory parameter used to balance the relative importance of prior knowledge; λ t is the regularization parameter of the t-th species; τ t is the irrelevance penalty parameter of the transcriptional regulatory network of the tth species; k is the number of species, which include model organisms and non-model organisms.

[0117] S160: Predicting the gene transcriptional regulatory mechanism corresponding to the species based on each of the transcriptional regulatory networks.

[0118] Next, the solution of the embodiment of the present application will be introduced and explained in detail with reference to specific application examples.

[0119] Based on the bioinformatics method for constructing TRNs, this embodiment comprehensively analyzes the multi-omics data of model organisms such as humans, mice, rats, fruit flies, worms and zebrafish, providing most of the model organism omics data for the inference of TRNs of non-model organisms in this embodiment.

[0120] This embodiment proposes a mathematical optimization model as a recognition model, which combines multiple omics data from multiple species and applies efficient algorithms to solve problems existing in the prior art, such as alternating direction method of multipliers (ADMM), proximal gradient algorithm (PGA), linearized Bregman method (LBM) and other large-scale optimization algorithms.

[0121] like Figure 2 As shown, this example proposes a bioinformatics method fused prior least absolute shrinkage and selection operator (fused pLASSO) to predict TRNs of non-model organisms that lack prior knowledge of transcriptional regulation. Starting from the identification of non-model organism TFs, TFs binding motifs and TFs binding sites (TFBSs), the genomes, transcriptomes, epigenomes and predicted TFBSs of model organisms and non-model organisms are then integrated into the fused pLASSO model with the best prior prediction performance for TRN inference.

[0122] Figure 2 Flowchart of the bioinformatics approach to predict TRNs in non-model organisms for multi-omics data. The pipeline starts with the identification of putative TFs, TF binding motifs, and TFBSs, and then integrates the genomes, predicted TFBSs, epigenomes (matrix PRS), and transcriptomes (matrix A and matrix B) of all model and non-model organisms into the fused pLASSO model for TRN inference, where matrix X represents the TRNs to be inferred. This approach is able to predict TRNs that are conserved among different animals as well as TRNs that are specific to each species.

[0123] Specifically, this embodiment may include the following steps:

[0124] 1. Processing and integration of omics data.

[0125] The 'p' in fused pLASSO represents the prior knowledge of transcriptional regulation obtained from the TFBSs predicted from active promoters and enhancers. TFs and TF binding motifs of different species can be obtained through databases such as AnimalTFDB and CIS-BP. Active promoters and enhancers can be identified through transcription-related histone modifications, i.e., ChIP-seq of the epigenome, and TFBSs of promoters and enhancers can be predicted (Qin et al., 2011). Using the epigenome and predicted TFBSs, the potential regulatory strength (PRS) of TF i on gene j is calculated:

[0126]

[0127] Among them, d l / d0 represents the ratio of the distance from the lth TFBS of transcription factor i to the transcription start site (TSS) of target gene j to the constant distance (10 kb). This distance is artificially set to define a region around the TSS where TFBS is considered to have a regulatory effect on gene transcription; p l is the intensity of histone modification marks, indicating the accessibility of DNA regulatory elements to TF binding. For model and non-model organisms, this example will calculate the PRS for each TF-target gene pair. If there is no histone modification data for a specific species, the p l All are 1. In this embodiment, PRS t represents the PRS matrix of all TF-target gene pairs in species t, which is the prior information of potential transcriptional regulation.

[0128] Since transcriptomes from different tissues and developmental stages show the expression relationship between TF and target genes, combining TF binding information with transcriptomes can improve TRN inference. Therefore, the transcriptomes of each species will be merged into two expression matrices. Matrix A t Contains TFs expression data of species t, matrix B t These transcriptome data and PRS matrices containing genomic, epigenomic, and TF binding information from model and non-model organisms will be projected into the fused pLASSO model to infer TRNs.

[0129] 2. Use fused pLASSO to infer TRN.

[0130] The relationship between TFs expression data and target gene expression data can be approximated by a linear relationship:

[0131] AX = B + ε;

[0132] Among them, A∈R m×r Represents the transcription factor expression matrix of the species, B∈R m×n represents the target gene expression matrix, ε∈R m×n represents the error matrix, X∈R r×n represents the regulatory matrix, which describes the regulatory relationship between the expression of TFs and target genes of a species; m is the number of samples, r is the number of transcription factors, and n is the number of target genes. In the gene transcription regulatory network inference, for each target gene j, this example wants to select a small number of TFs to make AX .,j and B .,j Minimize the difference between them, which is a sparse optimization problem, expressed as:

[0133] min||AX .,j -B .,j ||2;

[0134] st||X .,j ||0≤K;

[0135] Where ||·||2 represents the Euclidean norm. ||X .,j ||0 means X .,j The number of non-zeros in ||X .,j ||0 means X .,j The high sparsity of indicates the number of transcription factors i that regulate target gene j.

[0136] According to the above theoretical description, this embodiment can generally convert the sparse optimization problem into an L1 regularization problem, L1 regularization model (LASSO), which is expressed as:

[0137]

[0138] in

[0139] According to the above research basic theory, this study is based on the existing multi-omics data of model organisms to infer the TRN of non-model organisms. The algorithm needs to meet the following conditions: (1) integrated PRSs as prior information of potential targets regulated by TF; (2) search for sparse solutions, because each gene is usually regulated by only a few TFs; (3) identify conserved TRNs to improve the accuracy of TRN. In order to meet the above conditions, this example proposes the following fused pLASSO model to achieve simultaneous inference of TRN of k species:

[0140]

[0141] The matrix A t Contains TFs expression data of species t, matrix B t Contains expression data of target genes, PRS t represents the PRS matrix of all TF-target gene pairs in species t; Indicates the species with the shortest divergence time in the phylogenetic tree among t species; X t represents the TRN of species t, which is the variable to be estimated; represents the inferred regulatory intensity of TF i on target gene j in species t, when When it is not 0, TF i is considered as a regulatory factor; η, λ t , τ t As a tuning parameter, k is the number of all species including model organisms and non-model organisms.

[0142] The following three penalty terms are important for inferring the three characteristics of TRN:

[0143] (1) Evaluate the difference between the prior information PRSs and the inferred TRN, which was originally proposed in prior LASSO (pLASSO); (2) It is the sparse induced L1 norm, first proposed in LASSO and widely used to approximate the sparse solution of linear systems; (3) The differences between the regulatory vectors for each pair of adjacent species in the phylogenetic tree are presented for the first time in fused pLASSO.

[0144] In summary, fused pLASSO limits the differences in regulatory information from both PRS and transcriptome, and the number of predicted TFs and regulatory differences between neighboring species are relatively small.

[0145] In the fused pLASSO model proposed in this embodiment, η, λ t , τ t To adjust the parameters, balance the importance of the prior information, sparsity and conservatism of the TRN information. The choice of parameters is crucial to the quality of the inferred TRN. In the study of this embodiment, this embodiment will adopt the following scheme to update the above three adjustment parameters:

[0146] (1) The subproblem of the fused LASSO penalty term in the mathematical algorithms (PGA and ADMM) can be analytically expressed using a thresholding operator. Due to the predefined sparsity level and the difference in the regulatory vectors between the two species, the parameter λ t , τ t Can be adjusted through dynamic schemes.

[0147] where λ t >0 is the regularization parameter of species t, which provides a balance between accuracy and sparsity. In this study, this embodiment uses an iterative soft thresholding algorithm to solve the L1 regularization model of gene regulatory network inference based on omics data. In each iteration, the gradient step size of the algorithm is:

[0148]

[0149] Then perform the threshold calculation:

[0150]

[0151] The superscript n of Z and X is the number of iterations, ν represents the step size, usually 1 / 2, and the regularization parameter λ is iteratively updated using the above algorithm. t , to maintain The sparsity of .

[0152] τ t For the TRN irrelevance penalty parameter of species t, set:

[0153]

[0154] is the sth largest absolute value, so as to control the correlation of TRN not to exceed s.

[0155] (2) In this embodiment, η is selected by cross-validation technology.

[0156] η is a tuning parameter that balances the relative importance of data and prior knowledge. When η = 0, pLASOO is reduced to LASSO, which is completely independent of prior knowledge; when η = ∞, pLASSO completely relies on prior knowledge to fit the model.

[0157] This embodiment will use cross-validation technology to automatically adjust the value of η according to the reliability of prior knowledge to improve the efficiency of the result evaluator. Since the inferred TRN of non-model organisms is not necessarily conserved in all species, the balance of the four calculation items in the model can detect both the TRN unique to each species and the TRN conserved between different species.

[0158] This example will use existing real data to evaluate the performance of this method. This example uses previously published mESC TRN gold standard data and uses the proximal gradient algorithm (PGA) to conduct preliminary tests on human data for mESC TRN.

[0159] This application also provides more specific experimental examples, as follows:

[0160] This embodiment uses a benchmark dataset to verify that TRNs conserved between species can improve the prediction performance of TRNs. In the experiment, this embodiment used human and mouse data to predict TRNs of mouse embryonic stem cells (mESC), and used two independent mESC gold standard TRNs (Qin et al.) to evaluate the prediction performance of fused pLASSO. LASSO only integrates the transcriptome data of mice, pLASSO integrates the ChIP-seq and transcriptome data of mice, and fused pLASSO integrates the ChIP-seq and transcriptome data of mice and humans at the same time. The three trained LASSO, pLASSO and fused pLASSO models were verified on the mouse mESC gold standard TRN dataset. This embodiment uses the proximal gradient algorithm (PGA) to calculate the prediction performance of the three models, and uses the area under the working curve (AUC) as the evaluation indicator to evaluate the prediction accuracy of the three model TRNs. Figure 3 As shown in the figure, on two independent gold standard data sets, the AUC of fused pLASSO was about 0.70, which was significantly higher than the other two models. After integrating human data, fused pLASSO significantly improved the accuracy of TRN prediction in mice.

[0161] Specifically, Figure 3 Figure 2 is the performance graph of fused pLASSO. The comparison of different models was performed by predicting mESC TRN and evaluating different models using two independent mESC gold standard TRNs: fused pLASSO integrates mouse and human ChIP-seq and transcriptome data; pLASSO integrates mouse ChIP-seq and transcriptome data; LASSO uses only mouse transcriptome data. The proximal gradient algorithm (PGA) was used to calculate the performance of the three models, and the area under the working curve (AUC) represents the accuracy of TRN prediction. The results show that the evaluation results of the two independent gold standard TRNs based on mESC are consistent, and the integration of human data by fused pLASSO improves the accuracy of mouse TRN prediction.

[0162] Reference Figure 4 The present application also provides a gene transcription regulation mechanism prediction device, which can implement the above-mentioned gene transcription regulation mechanism prediction method, and the device includes:

[0163] A data acquisition unit, used to acquire transcription factors of different species and target genes corresponding to the transcription factors;

[0164] A matrix calculation unit, used to calculate the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene are regarded as a transcription factor and target gene pair;

[0165] A matrix determination unit, used to determine a transcription factor expression matrix and a target gene expression matrix according to the regulation capability parameter matrix;

[0166] A linear relationship construction unit, used to construct a linear relationship between the transcription factor expression matrix and the target gene expression matrix;

[0167] A problem construction unit, used for selecting a plurality of the transcription factors to minimize the difference of the linear relationship, thereby constructing a sparse optimization problem;

[0168] A network determination unit, used for searching for a sparse solution according to the sparse optimization problem, and then determining the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factor, the DNA regulatory element and the target gene corresponding to the transcription factor;

[0169] A mechanism prediction unit is used to predict the gene transcriptional regulation mechanism corresponding to the species based on each of the transcriptional regulation networks.

[0170] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0171] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above gene transcription regulation mechanism prediction method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0172] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0173] See also Figure 5 , Figure 5 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0174] The processor 501 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0175] The memory 502 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 502 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 502, and the processor 501 calls and executes the gene transcription regulation mechanism prediction method of the embodiment of this application;

[0176] Input / output interface 503, used to implement information input and output;

[0177] Communication interface 504, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0178] A bus 505 that transmits information between the various components of the device (e.g., the processor 501, the memory 502, the input / output interface 503, and the communication interface 504);

[0179] The processor 501 , the memory 502 , the input / output interface 503 and the communication interface 504 are connected to each other in communication within the device via the bus 505 .

[0180] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned gene transcription regulation mechanism prediction method is implemented.

[0181] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0182] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0183] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0184] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0185] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0186] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0187] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application 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 interchangeable where appropriate, so that the embodiments of the present application 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 comprising 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.

[0188] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0189] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0190] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0191] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0192] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0193] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for predicting gene transcription regulation mechanism, characterized in that: The method comprises the following steps: Obtaining transcription factors of different species and target genes corresponding to the transcription factors; Calculating the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene is regarded as a transcription factor and target gene pair; Determine a transcription factor expression matrix and a target gene expression matrix according to the regulatory ability parameter matrix; Constructing a linear relationship between the transcription factor expression matrix and the target gene expression matrix; Selecting a number of the transcription factors to minimize the difference in the linear relationship, thereby constructing a sparse optimization problem; Searching for a sparse solution according to the sparse optimization problem, and then determining the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factor, the DNA regulatory element and the target gene corresponding to the transcription factor; The gene transcriptional regulatory mechanism corresponding to the species is predicted based on each of the transcriptional regulatory networks.

2. The method for predicting gene transcription regulation mechanism according to claim 1, characterized in that: The method of obtaining transcription factors of different species and target genes corresponding to the transcription factors comprises the following steps: Obtain the transcription factors and binding motifs of the transcription factors in different tissues and developmental stages of model organisms and non-model organisms from public databases; identifying active promoters and enhancers of each of the transcription factors based on transcription-related histone modifications; The binding sites of each of the transcription factors are predicted based on the active promoter and the enhancer; wherein the target gene corresponding to each of the transcription factors includes the corresponding binding motif and the binding site.

3. The method for predicting gene transcription regulation mechanism according to claim 1, characterized in that: The step of calculating the regulatory ability parameter matrix of each transcription factor and target gene pair comprises the following steps: Calculate the regulatory ability parameters of each transcription factor and target gene pair according to the regulatory ability parameter expression; Determine the control capability parameter matrix according to each of the control capability parameters; The control capability parameter expression is: Among them, PRS i,j represents the regulatory ability parameter of the i-th transcription factor and the j-th target gene; d l / d0 represents the ratio of the distance from the lth binding site of the ith transcription factor to the transcription start site of the jth target gene to the constant distance; p l is the intensity of the transcription-related histone modification mark, which is used to indicate the accessibility of the DNA regulatory element to the binding of the corresponding transcription factor; n is the number of the target genes, and l is the serial number of the binding site.

4. The method for predicting gene transcription regulation mechanism according to claim 1, characterized in that: Determining the transcription factor expression matrix and the target gene expression matrix according to the regulation capability parameter matrix comprises the following steps: Merging the transcription factor expression data of all the species in the regulatory ability parameter matrix into the transcription factor expression matrix; The target gene expression data of all the species in the regulatory ability parameter matrix are combined into the target gene expression matrix.

5. The method for predicting gene transcription regulation mechanism according to claim 1, characterized in that: The step of constructing a linear relationship between the transcription factor expression matrix and the target gene expression matrix comprises the following steps: The linear relationship between the transcription factor expression matrix and the target gene expression matrix is ​​constructed as follows: AX = B + ε; Among them, A∈R m×r , A represents the transcription factor expression matrix; B∈R m×n , B represents the target gene expression matrix; ε∈R m×n , ε represents the error matrix; X∈R r×n , X represents the regulatory matrix; m is the number of the species, r is the number of the transcription factors, and n is the number of the target genes.

6. The method for predicting gene transcription regulation mechanism according to claim 5, characterized in that: The step of selecting a plurality of the transcription factors to minimize the difference in the linear relationship, and then constructing a sparse optimization problem, comprises the following steps: The transcription factors less than a threshold number are selected to minimize the difference of the linear relationship, and then the sparse optimization problem is constructed as follows: min||AX .,j -B .,j ||2; s.t.||X .,j ||0≤K; Among them, ||·||2 means that the Euclidean norm is ||X .,j ||0 means X .,j The number of non-zeros in ||X .,j ||0 means X .,j The sparsity of is used to represent the number of the i-th transcription factor regulating the j-th target gene.

7. The method for predicting gene transcription regulation mechanism according to claim 6, characterized in that: The step of searching for a sparse solution according to the sparse optimization problem comprises the following steps: Convert the sparse optimization problem into an L1 regularization problem; The conversion to the L1 regularization problem is: in, Searching and obtaining the sparse solution according to the L1 regularization problem; Determining the transcriptional regulatory networks of different species according to the sparse solution comprises the following steps: Determining the transcriptional regulatory networks of different species according to the sparse solution using a recognition model; The expression of the recognition model is: Among them, A t represents the transcription factor expression matrix of the t-th species, B t represents the target gene expression matrix of the t-th species; PRS t A matrix representing the regulatory ability parameters of all the transcription factor and target gene pairs in the t-th species; X represents the species with the shortest divergence time in the phylogenetic tree among the t-th species; t represents the transcriptional regulatory network in the t-th species; represents the inferred regulatory strength of the i-th transcription factor on the j-th target gene in the t-th species. When it is not 0, the i-th transcription factor is used as a regulatory factor; η is a regulatory parameter used to balance the relative importance of prior knowledge; λ t is the regularization parameter of the t-th species; τ t is the irrelevance penalty parameter of the transcriptional regulatory network of the tth species; k is the number of species, which include model organisms and non-model organisms.

8. A gene transcription regulation mechanism prediction device, characterized in that: The device comprises: A data acquisition unit, used to acquire transcription factors of different species and target genes corresponding to the transcription factors; A matrix calculation unit, used to calculate the regulatory ability parameter matrix of each transcription factor and target gene pair; wherein each transcription factor and the corresponding target gene are regarded as a transcription factor and target gene pair; A matrix determination unit, used to determine a transcription factor expression matrix and a target gene expression matrix according to the regulation capability parameter matrix; A linear relationship construction unit, used to construct a linear relationship between the transcription factor expression matrix and the target gene expression matrix; A problem construction unit, used for selecting a plurality of the transcription factors to minimize the difference of the linear relationship, thereby constructing a sparse optimization problem; A network determination unit, used to search for a sparse solution according to the sparse optimization problem, and then determine the transcriptional regulatory network of different species according to the sparse solution; wherein the transcriptional regulatory network includes the transcription factor, the DNA regulatory element and the target gene corresponding to the transcription factor; A mechanism prediction unit is used to predict the gene transcriptional regulation mechanism corresponding to the species based on each of the transcriptional regulation networks.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images