CN110853763A - Fusion attribute-based miRNA-disease association identification method and system - Google Patents

Fusion attribute-based miRNA-disease association identification method and system Download PDF

Info

Publication number
CN110853763A
CN110853763A CN201911095171.6A CN201911095171A CN110853763A CN 110853763 A CN110853763 A CN 110853763A CN 201911095171 A CN201911095171 A CN 201911095171A CN 110853763 A CN110853763 A CN 110853763A
Authority
CN
China
Prior art keywords
mirna
disease
mirnas
weight
functional similarity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911095171.6A
Other languages
Chinese (zh)
Other versions
CN110853763B (en
Inventor
曹步文
邓曙光
阳王东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hongfujin Precision Industry Shenzhen Co Ltd
Original Assignee
Hongfujin Precision Industry Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hongfujin Precision Industry Shenzhen Co Ltd filed Critical Hongfujin Precision Industry Shenzhen Co Ltd
Priority to CN201911095171.6A priority Critical patent/CN110853763B/en
Publication of CN110853763A publication Critical patent/CN110853763A/en
Application granted granted Critical
Publication of CN110853763B publication Critical patent/CN110853763B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Biotechnology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Chemical & Material Sciences (AREA)
  • Biophysics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Analytical Chemistry (AREA)
  • Biomedical Technology (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)

Abstract

The invention discloses a miRNA-disease association identification method and system based on fusion attributes. The method comprises the steps of firstly calculating the functional similarity between any two miRNAs in a disease database, constructing an miRNA network undirected graph according to the functional similarity, calculating the aggregation coefficient between any two different miRNAs, and fusing the functional similarity and the aggregation coefficient between the two miRNAs to obtain a combined weight; and calculating the weight of each miRNA vertex according to the combined weight, performing descending sorting on each miRNA according to the weight, and screening out potential miRNAs related to diseases according to the descending sorting result. The method is simple to realize, miRNA nodes are subjected to weighting and descending sequencing on the basis of fusing two attributes of topological attributes and functional similarity, and then the association of potential miRNA and diseases is predicted by means of the disease database related to human miRNA and the sequenced result, so that the accuracy of miRNA-disease association identification is improved, and valuable clues can be provided for medical diagnosis.

Description

Fusion attribute-based miRNA-disease association identification method and system
Technical Field
The invention relates to the technical field of information biology, in particular to a miRNA-disease association identification method and system based on fusion attributes.
Background
mirnas (micrornas) are non-coding small-molecule RNAs, containing about 22 amino acids, and play important roles in post-transcriptional regulation of gene expression and various biological activities. Studies have demonstrated that dysfunction of mirnas is associated with a variety of complex human diseases. Therefore, identification of potential association of miRNA-diseases helps understanding the mechanisms of pathogens, providing valuable reference for medical diagnosis of human diseases. In 2010, Jiang et al quantified the functional similarity between miRNAs by using the overlapping degree of the target genes of the miRNAs for the first time, optimized disease-related miRNAs by hyper-geometric distribution, and further integrated the phenotypic similarity of diseases and the similarity of miRNAs to calculate the functional similarity between miRNA pairs.
Researchers have suggested that mirnas with similar functions often have some relationship to similar diseases and vice versa. Therefore, mirnas are widely used for prediction of disease-mirnas. Professor cheng proposes a RWRMDA (Random Walk with retrieval for MiRNA-Disease Association) method, and discovers some candidate Disease-related mirnas by Random Walk for the first time by utilizing global network information; xuan et al predict the association between miRNA and disease using a random walk model in miRNA functional similarity multithreading; based on the functional similarity of mirnas, professor luoca provides an unbalanced dichotomous random walk method in a heterogeneous network in 2017 to predict the association between miRNA and diseases. In 2018, Li et al predict the association of miRNA with disease by using label propagation based on linear neighborhood similarity.
Although various methods have been used to predict the association between miRNA and disease, there are methods that predict the association between miRNA-disease based only on the functional similarity of miRNA pairs and only on the topological properties of the miRNA similarity network. The literature shows that it is impractical to construct an absolutely reliable biological network by only one method. The existing miRNA-disease association identification method is often used for predicting the association between miRNA-diseases based on only one attribute, so that the identification of miRNA-disease association is not accurate enough.
Disclosure of Invention
The invention aims to provide a miRNA-disease association identification method and system based on fusion attributes, which are used for predicting potential association between miRNA and diseases based on topological attributes and functional similarities so as to solve the problem of low accuracy of the existing miRNA-disease association identification method.
In order to achieve the purpose, the invention provides the following scheme:
a method for miRNA-disease association identification based on fusion attributes, the method comprising:
acquiring a disease database to be analyzed; a plurality of disease types and a plurality of miRNAs are stored in the disease database;
calculating functional similarity between any two mirnas in the disease database;
constructing a miRNA network undirected graph according to the functional similarity between any two miRNAs;
calculating an aggregation coefficient between any two miRNAs according to the miRNA network undirected graph;
determining a combination weight between the any two miRNAs according to the functional similarity and the aggregation coefficient between the any two miRNAs;
calculating the weight of each miRNA vertex according to the combined weight between any two miRNAs;
sequencing the miRNAs in a descending order according to the weight of the peak of each miRNA;
and screening potential miRNA related to diseases according to the descending order sorting result.
Optionally, the calculating of the functional similarity between any two mirnas in the disease database specifically includes:
using a formulaCalculating functional similarity between any two mirnas in the disease database; wherein miRNA m1 and miRNA m2 are any two mirnas in the disease database; MSim (m1, m2) indicates functional similarity between miRNA m1 and miRNA m 2; DT1Represents a disease set associated with miRNA m1,
Figure BDA0002268102900000022
|DT1i denotes the length of the disease set associated with mirNam1, dt1iIndicating a set of diseases DT1The i disease of (1); DT2Represents a disease set associated with miRNA m2,|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j disease of (1); sim (dt)1i,DT2) Indicating disease dt1iAnd disease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
Optionally, the calculating an aggregation coefficient between any two mirnas according to the miRNA network undirected graph specifically includes:
obtaining a neighbor node set N of the miRNA m1 in the miRNA network undirected graphm1And a set of neighbor nodes N of the mirNam2m2；
Using a formulaCalculating an aggregation coefficient ECC (m1, m2) between miRNA m1 and miRNA m 2; wherein | Nm1∩Nm2I represents Nm1And Nm2The number of nodes in the intersection; | Nm1I represents Nm1The number of nodes in the node; | Nm2I represents Nm2The number of nodes in (1).
Optionally, the determining the combining weight between any two mirnas according to the functional similarity and the aggregation coefficient between any two mirnas specifically includes:
according to the functional similarity MSim (m1, m2) and the aggregation coefficient ECC (m1, m2) between any two miRNAs, adopting a formula
Figure BDA0002268102900000032
Determining a combined weight ω (m1, m2) between the miRNAs m1 and miRNAs m 2.
Optionally, the calculating the weight of each miRNA vertex according to the combined weight between any two mirnas specifically includes:
using formula Vωs＝kmax×dωsCalculating the weight V of any miRNA vertex Vωs(ii) a Wherein k ismaxRepresenting the weight of the edge in the weighted subgraph S; the weighted subgraph S is a set of the miRNA vertex V and direct neighbor nodes thereof;
Figure BDA0002268102900000033
the | V | represents the number of miRNA in the weighted subgraph S; ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S.
A system for miRNA-disease association identification based on fusion attributes, the system comprising:
the disease database acquisition module is used for acquiring a disease database to be analyzed; a plurality of disease types and a plurality of miRNAs are stored in the disease database;
a functional similarity calculation module for calculating the functional similarity between any two miRNAs in the disease database;
the miRNA network undirected graph construction module is used for constructing an miRNA network undirected graph according to the functional similarity between any two miRNAs;
an aggregation coefficient calculation module, configured to calculate an aggregation coefficient between any two mirnas according to the miRNA network undirected graph;
the attribute fusion module is used for determining the combined weight between any two miRNAs according to the functional similarity and the aggregation coefficient between any two miRNAs;
the peak weight calculation module is used for calculating the weight of each miRNA peak according to the combined weight between any two miRNAs;
the sequencing module is used for sequencing the miRNAs in a descending order according to the weight of the top point of each miRNA;
and the screening module is used for screening potential miRNA related to diseases according to the descending order sorting result.
Optionally, the functional similarity calculation module specifically includes:
a functional similarity calculation unit for employing a formula
Figure BDA0002268102900000041
Calculating functional similarity between any two mirnas in the disease database; wherein miRNA m1 and miRNA m2 are any two mirnas in the disease database; MSim (m1, m2) indicates functional similarity between miRNA m1 and miRNA m 2; DT1Represents a disease set associated with miRNA m1,
Figure BDA0002268102900000042
DT1length, dt, representing the disease set associated with miRNA m11iIndicating a set of diseases DT1The i disease of (1); DT2Represents a disease set associated with miRNA m2,|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j disease of (1); sim (dt)1i,DT2) Indicating disease dt1iAnd disease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
Optionally, the aggregation coefficient calculating module specifically includes:
a neighbor node set obtaining unit, configured to obtain a neighbor node set N of the miRNA m1 in the miRNA network undirected graphm1And the neighbor node set N of the miRNA m2m2；
An aggregation coefficient calculation unit for employing a formula
Figure BDA0002268102900000044
Calculating an aggregation coefficient ECC (m1, m2) between miRNAm1 and miRNA m 2; wherein | Nm1∩Nm2I represents Nm1And Nm2The number of nodes in the intersection; | Nm1I represents Nm1The number of nodes in the node; | Nm2I representsNm2The number of nodes in (1).
Optionally, the attribute fusion module specifically includes:
a combined weight calculation unit for adopting a formula according to the functional similarity MSim (m1, m2) and the aggregation coefficient ECC (m1, m2) between any two miRNAs
Figure BDA0002268102900000051
Determining a combined weight ω (m1, m2) between the miRNAs m1 and miRNAs m 2.
Optionally, the vertex weight calculating module specifically includes:
a vertex weight calculation unit for employing the formula Vωs＝kmax×dωsCalculating the weight V of any miRNA vertex Vωs(ii) a Wherein k ismaxRepresenting the weight of the edge in the weighted subgraph S; the weighted subgraph S is a set of the miRNA vertex V and direct neighbor nodes thereof;
Figure BDA0002268102900000052
the | V | represents the number of miRNA in the weighted subgraph S; ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a miRNA-disease association identification method and system based on fusion attributes, wherein the method comprises the steps of firstly calculating the functional similarity between any two miRNAs in a disease database, constructing a miRNA network undirected graph according to the functional similarity between any two miRNAs, calculating an aggregation coefficient between any two miRNAs according to the miRNA network undirected graph, and fusing the functional similarity and the aggregation coefficient between any two miRNAs to obtain the combined weight between any two miRNAs; calculating the weight of each miRNA vertex according to the combined weight between any two miRNAs; sequencing the miRNAs in a descending order according to the weight of the peak of each miRNA; and screening potential miRNA related to diseases according to the descending order sorting result. The method is simple to realize, the miRNA nodes are subjected to weighting and descending sequencing on the basis of fusing the two attributes, then the association of potential miRNA and diseases is predicted by means of the disease database related to human miRNA and the sequenced result, the accuracy of miRNA-disease association identification is improved, and valuable clues can be provided for medical diagnosis.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a flow chart of a miRNA-disease association identification method based on fusion attributes provided by the present invention;
FIG. 2 is a schematic diagram of a fusion attribute-based miRNA-disease association identification method provided by the invention;
fig. 3 is a structural diagram of the miRNA-disease association recognition system based on fusion attributes provided by the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a miRNA-disease association identification method and system based on fusion attributes, which are used for predicting potential association between miRNA and diseases based on topological attributes and functional similarities so as to solve the problem of low accuracy of the existing miRNA-disease association identification method.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a flow chart of the miRNA-disease association identification method based on fusion attributes provided by the present invention. Fig. 2 is a schematic diagram of the miRNA-disease association recognition method based on fusion attributes provided in the present invention. Referring to fig. 1 and fig. 2, the miRNA-disease association identification method based on fusion attributes provided by the present invention specifically includes:
step 101: a disease database to be analyzed is obtained.
The disease database stores a plurality of disease types and a plurality of miRNAs. Taking dbDEMC (database of differentially expressed miRNAs in tumors) (http:// www.picb.ac.cn/dbDEMC) as an example, it is a comprehensive database intended to detect the presence and display of differentially expressed miRNAs in human cancers by high-throughput methods. The current version contains 2224 differentially expressed mirnas in 36 cancer types, 49202 miRNA-cancer associations, providing an enhanced miRNA page to demonstrate basic miRNA descriptions, multiple expression profiles in each analysis type, and the results of related analysis experiments and low throughput validation assays. Differential expression profiles are shown in the form of heatmaps for a panel of mirnas and a variety of cancer types, in order for researchers to explore differences and similarities between cancers.
Step 102: calculating functional similarity between any two miRNAs in the disease database.
The functional similarity calculation method comprises the following steps:
Figure BDA0002268102900000071
calculating the functional similarity between any two miRNAs (a pair of miRNAs) in the disease database by adopting a formula (1); wherein miRNA m1 and miRNA m2 are any two different miRNAs in the disease database. MSim (m1, m2) indicates functional similarity between miRNA m1 and miRNA m 2. DT1Represents a disease set associated with miRNA m1,
Figure BDA0002268102900000072
|DT1i denotes the length of the disease set associated with miRNA m1, dt1iIndicating a set of diseases DT1The i-th disease of (1). DT2Represents a disease set associated with miRNA m2,|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j-th disease in (1). Sim (dt)1i,DT2) Indicating disease dt1iAnd disease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
Step 103: and constructing a miRNA network undirected graph according to the functional similarity between any two miRNAs.
Inputting the calculated functional similarity network information between each group of miRNA pairs (miRNA m1 and miRNA m2), filtering repeated interaction and self-interaction in the functional similarity network information, and establishing an undirected graph G of the miRNA network.
Step 104: and calculating the aggregation coefficient between any two miRNAs according to the miRNA network undirected graph.
The existing miRNA-disease association identification method usually uses only one functional similarity calculation to predict the association between miRNA and diseases, so that the identification of miRNA-disease association is not accurate enough. The invention identifies the correlation between miRNA and disease based on topological attribute and functional similarity, thus the identification accuracy is higher. The aggregation coefficients calculated in step 104 are the calculated topology attributes.
The aggregation coefficient between miRNA and miRNA can be calculated by the following formula:
Figure BDA0002268102900000074
ECC (m1, m2) is the edge aggregation coefficient between miRNA m1 and miRNA m 2. N is a radical ofm1Is the neighbor node set, N, of the miRNA m1 in the miRNA network undirected graphm2Is the neighbor node set of the miRNA m 2. | Nm1∩Nm2I represents Nm1And Nm2The number of neighbor nodes in the intersection. | Nm1I represents Nm1The number of neighbor nodes in the node; | Nm2I represents Nm2The number of neighbor nodes in (1). min (N)m1，Nm2) Represents the minimum number of neighbor nodes of miRNA m1 and miRNA m 2.
Step 105: determining a combining weight between the any two miRNAs according to the functional similarity and the aggregation coefficient between the any two miRNAs.
The functional similarity and aggregation coefficient of the miRNA pairs are fused, and the fusion method of the functional similarity and aggregation coefficient of the miRNA pairs can be obtained by the following formula:
wherein ω (m1, m2) represents the combined weight between miRNA m1 and miRNA m2, effectively fusing the topological properties and functional similarity of miRNA functional similarity networks. The denominator is increased by 1 to avoid the case of 0.
Step 106: calculating the weight of each miRNA vertex according to the combined weight between any two miRNAs.
In a weighted subgraph S, the weight of a miRNA vertex V can be calculated by:
Vωs＝kmax×dωs(4)
wherein, VωsIs the weight of any miRNA vertex V. k is a radical ofmaxRepresenting the weights of the edges in subgraph S, which is the set of miRNA vertices V and their direct neighbor nodes.
Figure BDA0002268102900000082
Representing the density of a weighted subgraph S. And | V | represents the number of miRNA in the weighted subgraph S. ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S. Σ ω denotes the summation of the combining weights for all edges in the weighted subgraph S.
The calculation of the weight of each miRNA vertex V is the fusion of the two topological attributes, namely the weight of each miRNA vertex V is calculated by combining the weights omega (m1, m 2).
Step 107: and sequencing the miRNAs in a descending order according to the weight magnitude of the peak of each miRNA.
Calculating the weight of each miRNA vertex by adopting a formula (4), then sorting the miRNAs in a descending order according to the weight of each miRNA vertex, and screening potential miRNAs related to diseases according to the sorting height.
Step 108: and screening potential miRNA related to diseases according to the descending order sorting result.
By means of a disease database related to human miRNA, the relevance of potential miRNA and disease is predicted by using the sequenced result, potential miRNA related to disease is screened out, and valuable clues are provided for medical diagnosis.
Taking miRNA related to lung cancer in PhenomiR2.0 database as an example, screening results obtained by the method are shown in Table 1:
TABLE 1 comparison of verified miRNA with screening results of the invention
Figure BDA0002268102900000091
Figure BDA0002268102900000101
In table 1, the 1 st column shows the mirnas that were verified by the phenomix 2.0 database, the 2 nd column TOP30 shows the mirnas that were screened by the method of the present invention and ranked 30 before the weight, and the 3 rd column sign shows whether the 30 mirnas screened by the present invention are related to diseases, the "Y" indicates related, "N" indicates unrelated. As can be seen from the screening results in table 1, 24 of the 30 mirnas screened by the method of the present invention are related to diseases, and only 6 of the 30 mirnas are not related to diseases, with an accuracy rate of 24/30 × 100% — 80%.
The pseudo code of the miRNA-disease association identification method based on the fusion attribute is as follows:
TABLE 2 pseudo code for fusion attribute based miRNA-disease association identification method
Figure BDA0002268102900000111
The invention provides a miRNA and disease association identification method based on topological attributes and functional similarities, starting from the biological significance of miRNA, by fusing the edge aggregation coefficients and the functional similarities of a miRNA similarity network. The method is simple to realize, the miRNA nodes are subjected to weighting and descending sequencing on the basis of fusing the two attributes, then the association of potential miRNA and diseases is predicted by means of the disease database related to human miRNA and the sequenced result, the accuracy of miRNA-disease association identification is improved, and valuable clues are provided for medical diagnosis.
Based on the miRNA-disease association recognition method based on the fusion attribute provided by the present invention, the present invention further provides a miRNA-disease association recognition system based on the fusion attribute, referring to fig. 3, the system includes:
a disease database acquisition module 301, configured to acquire a disease database to be analyzed; a plurality of disease types and a plurality of miRNAs are stored in the disease database;
a functional similarity calculation module 302, configured to calculate a functional similarity between any two mirnas in the disease database;
a miRNA network undirected graph constructing module 303, configured to construct a miRNA network undirected graph according to functional similarity between any two mirnas;
an aggregation coefficient calculation module 304, configured to calculate an aggregation coefficient between any two mirnas according to the miRNA network undirected graph;
an attribute fusion module 305, configured to determine a combining weight between the any two mirnas according to the functional similarity and the aggregation coefficient between the any two mirnas;
a vertex weight calculation module 306, configured to calculate a weight of each miRNA vertex according to the combined weight between any two mirnas;
a sorting module 307, configured to sort the mirnas in a descending order according to the weight of the top point of each miRNA;
a screening module 308 for screening potential mirnas related to the disease according to the descending order sorting result.
The functional similarity calculation module 302 specifically includes:
a functional similarity calculation unit for employing a formula
Figure BDA0002268102900000121
Calculating functional similarity between any two mirnas in the disease database; wherein miRNA m1 and miRNA m2 are any two mirnas in the disease database; MSim (m1, m2) indicates functional similarity between miRNA m1 and miRNA m 2; DT1Represents a disease set associated with miRNA m1,
Figure BDA0002268102900000131
|DT1i denotes the length of the disease set associated with miRNA m1, dt1iIndicating a set of diseases DT1The i disease of (1); DT2Represents a disease set associated with miRNA m2,
Figure BDA0002268102900000132
|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j disease of (1); sim (dt)1i,DT2) Indicating disease dt1iAnd disease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
The aggregation coefficient calculation module 304 specifically includes:
a neighbor node set obtaining unit, configured to obtain a neighbor node set N of the miRNA m1 in the miRNA network undirected graphm1And the neighbor node set N of the miRNA m2m2；
An aggregation coefficient calculation unit for employing a formulaCalculating an aggregation coefficient ECC (m1, m2) between miRNAm1 and miRNA m 2; wherein | Nm1∩Nm2I represents Nm1And Nm2The number of nodes in the intersection; | Nm1I represents Nm1The number of nodes in the node; | Nm2I represents Nm2The number of nodes in (1).
The attribute fusion module 305 specifically includes:
a combined weight calculation unit for adopting a formula according to the functional similarity MSim (m1, m2) and the aggregation coefficient ECC (m1, m2) between any two miRNAsDetermining a combined weight ω (m1, m2) between the miRNAs m1 and miRNAs m 2.
The vertex weight calculation module 306 specifically includes:
a vertex weight calculation unit for employing the formula Vωs＝kmax×dωsCalculating the weight V of any miRNA vertex Vωs(ii) a Wherein k ismaxRepresenting the weight of the edge in the weighted subgraph S; the weighted subgraph S is a set of the miRNA vertex V and direct neighbor nodes thereof;
Figure BDA0002268102900000135
the | V | represents the number of miRNA in the weighted subgraph S; ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (10)

1. A fusion attribute-based miRNA-disease association identification method is characterized by comprising the following steps:
acquiring a disease database to be analyzed; a plurality of disease types and a plurality of miRNAs are stored in the disease database;
calculating functional similarity between any two mirnas in the disease database;
constructing a miRNA network undirected graph according to the functional similarity between any two miRNAs;
calculating an aggregation coefficient between any two miRNAs according to the miRNA network undirected graph;
determining a combination weight between the any two miRNAs according to the functional similarity and the aggregation coefficient between the any two miRNAs;
calculating the weight of each miRNA vertex according to the combined weight between any two miRNAs;
sequencing the miRNAs in a descending order according to the weight of the peak of each miRNA;
and screening potential miRNA related to diseases according to the descending order sorting result.
2. The miRNA-disease association identification method according to claim 1, wherein the calculating of the functional similarity between any two mirnas in the disease database specifically comprises:
using a formula
Figure FDA0002268102890000011
Calculating functional similarity between any two mirnas in the disease database; wherein miRNA m1 and miRNA m2 are any two mirnas in the disease database; MSim (m1, m2) denotes miRNA mFunctional similarity between 1 and miRNA m 2; DT1Represents a disease set associated with miRNA m1,
Figure FDA0002268102890000012
|DT1i denotes the length of the disease set associated with mirNam1, dt1iIndicating a set of diseases DT1The i disease of (1); DT2Represents a disease set associated with miRNA m2,
Figure FDA0002268102890000013
|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j disease of (1); sim (dt)1i,DT2) Indicating disease dt1iAnd disease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
3. The miRNA-disease association identification method according to claim 2, wherein the calculating an aggregation coefficient between any two mirnas according to the miRNA network undirected graph specifically comprises:
obtaining a neighbor node set N of the miRNA m1 in the miRNA network undirected graphm1And the neighbor node set N of the miRNA m2m2；
Using a formula
Figure FDA0002268102890000021
Calculating an aggregation coefficient ECC (m1, m2) between miRNA m1 and miRNA m 2; wherein | Nm1∩Nm2I represents Nm1And Nm2The number of nodes in the intersection; | Nm1I represents Nm1The number of nodes in the node; | Nm2I represents Nm2The number of nodes in (1).
4. The miRNA-disease association identification method according to claim 3, wherein the determining the combined weight between any two mirnas according to the functional similarity and the aggregation coefficient between any two mirnas specifically comprises:
according to the functional similarity MSim (m1, m2) and the aggregation coefficient ECC (m1, m2) between any two miRNAs, adopting a formula
Figure FDA0002268102890000022
Determining a combined weight ω (m1, m2) between the miRNAs m1 and miRNAs m 2.
5. The miRNA-disease association identification method of claim 4, wherein the calculating the weight of each miRNA vertex according to the combined weight between any two miRNAs specifically comprises:
using formula Vωs＝kmax×dωsCalculating the weight V of any miRNA vertex Vωs(ii) a Wherein k ismaxRepresenting the weight of the edge in the weighted subgraph S; the weighted subgraph S is a set of the miRNA vertex V and direct neighbor nodes thereof;
Figure FDA0002268102890000023
the | V | represents the number of miRNA in the weighted subgraph S; ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S.
6. A system for miRNA-disease association identification based on fusion attributes, the system comprising:
the disease database acquisition module is used for acquiring a disease database to be analyzed; a plurality of disease types and a plurality of miRNAs are stored in the disease database;
a functional similarity calculation module for calculating the functional similarity between any two miRNAs in the disease database;
the miRNA network undirected graph construction module is used for constructing an miRNA network undirected graph according to the functional similarity between any two miRNAs;
an aggregation coefficient calculation module, configured to calculate an aggregation coefficient between any two mirnas according to the miRNA network undirected graph;
the attribute fusion module is used for determining the combined weight between any two miRNAs according to the functional similarity and the aggregation coefficient between any two miRNAs;
the peak weight calculation module is used for calculating the weight of each miRNA peak according to the combined weight between any two miRNAs;
the sequencing module is used for sequencing the miRNAs in a descending order according to the weight of the top point of each miRNA;
and the screening module is used for screening potential miRNA related to diseases according to the descending order sorting result.
7. The miRNA-disease association recognition system of claim 6, wherein the functional similarity calculation module specifically comprises:
a functional similarity calculation unit for employing a formulaCalculating functional similarity between any two mirnas in the disease database; wherein miRNA m1 and miRNA m2 are any two mirnas in the disease database; MSim (m1, m2) indicates functional similarity between miRNA m1 and miRNA m 2; DT1Represents a disease set associated with miRNA m1,
Figure FDA0002268102890000032
|DT1i denotes the length of the disease set associated with miRNA m1, dt1iIndicating a set of diseases DT1The i disease of (1); DT2Represents a disease set associated with miRNA m2,
Figure FDA0002268102890000033
|DT2i denotes the length of the disease set associated with miRNA m2, dt2jIndicating a set of diseases DT2The j disease of (1); sim (dt)1i,DT2) Indicating disease dt1iAnd disease of the heartDisease set DT2Similarity between them, Sim (dt)2j,DT1) Indicating disease dt2jAnd disease set DT1The similarity between them.
8. The miRNA-disease association recognition system of claim 7, wherein the aggregation coefficient calculation module specifically comprises:
a neighbor node set obtaining unit, configured to obtain a neighbor node set N of the miRNA m1 in the miRNA network undirected graphm1And the neighbor node set N of the miRNA m2m2；
An aggregation coefficient calculation unit for employing a formula
Figure FDA0002268102890000034
Calculating an aggregation coefficient ECC (m1, m2) between miRNA m1 and miRNA m 2; wherein | Nm1∩Nm2I represents Nm1And Nm2The number of nodes in the intersection; | Nm1I represents Nm1The number of nodes in the node; | Nm2I represents Nm2The number of nodes in (1).
9. The miRNA-disease association recognition system of claim 8, wherein the attribute fusion module specifically comprises:
a combined weight calculation unit for adopting a formula according to the functional similarity MSim (m1, m2) and the aggregation coefficient ECC (m1, m2) between any two miRNAsDetermining a combined weight ω (m1, m2) between the miRNAs m1 and miRNAs m 2.
10. The miRNA-disease association recognition system of claim 9, wherein the vertex weight calculation module specifically comprises:
a vertex weight calculation unit for employing the formula Vωs＝kmax×dωsCalculating the weight of any miRNA vertex VVωs(ii) a Wherein k ismaxRepresenting the weight of the edge in the weighted subgraph S; the weighted subgraph S is a set of the miRNA vertex V and direct neighbor nodes thereof;
Figure FDA0002268102890000042
the | V | represents the number of miRNA in the weighted subgraph S; ω represents the combined weight of the edges in the weighted subgraph S, i.e. the combined weight between any two mirnas in the weighted subgraph S.
CN201911095171.6A 2019-11-11 2019-11-11 Fusion attribute-based miRNA-disease association identification method and system Active CN110853763B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911095171.6A CN110853763B (en) 2019-11-11 2019-11-11 Fusion attribute-based miRNA-disease association identification method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911095171.6A CN110853763B (en) 2019-11-11 2019-11-11 Fusion attribute-based miRNA-disease association identification method and system

Publications (2)

Publication Number Publication Date
CN110853763A true CN110853763A (en) 2020-02-28
CN110853763B CN110853763B (en) 2022-07-19

Family

ID=69601321

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911095171.6A Active CN110853763B (en) 2019-11-11 2019-11-11 Fusion attribute-based miRNA-disease association identification method and system

Country Status (1)

Country Link
CN (1) CN110853763B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111681705A (en) * 2020-05-21 2020-09-18 中国科学院深圳先进技术研究院 miRNA-disease association prediction method, system, terminal and storage medium
CN114783597A (en) * 2022-06-23 2022-07-22 瀚依科技(杭州)有限公司 Method and device for diagnosing multi-class diseases, electronic equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107506608A (en) * 2017-09-29 2017-12-22 杭州电子科技大学 A kind of improved miRNA disease association Forecasting Methodologies based on collaborative filtering
CN107526937A (en) * 2017-09-29 2017-12-29 杭州电子科技大学 A kind of MiRNA disease association Forecasting Methodologies based on collaboration filtering
CN107862179A (en) * 2017-11-06 2018-03-30 中南大学 A kind of miRNA disease association Relationship Prediction methods decomposed based on similitude and logic matrix
CN109712670A (en) * 2018-12-25 2019-05-03 湖南城市学院 A kind of recognition methods and system of miRNA functional module
CN109935332A (en) * 2019-03-01 2019-06-25 桂林电子科技大学 A kind of miRNA- disease association prediction technique based on double random walk models
CN110428899A (en) * 2019-08-02 2019-11-08 陕西师范大学 The more Data Integration circular rnas restarted based on double random walks and disease associated prediction technique

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107506608A (en) * 2017-09-29 2017-12-22 杭州电子科技大学 A kind of improved miRNA disease association Forecasting Methodologies based on collaborative filtering
CN107526937A (en) * 2017-09-29 2017-12-29 杭州电子科技大学 A kind of MiRNA disease association Forecasting Methodologies based on collaboration filtering
CN107862179A (en) * 2017-11-06 2018-03-30 中南大学 A kind of miRNA disease association Relationship Prediction methods decomposed based on similitude and logic matrix
CN109712670A (en) * 2018-12-25 2019-05-03 湖南城市学院 A kind of recognition methods and system of miRNA functional module
CN109935332A (en) * 2019-03-01 2019-06-25 桂林电子科技大学 A kind of miRNA- disease association prediction technique based on double random walk models
CN110428899A (en) * 2019-08-02 2019-11-08 陕西师范大学 The more Data Integration circular rnas restarted based on double random walks and disease associated prediction technique

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
LUO ET AL.: "A novel approach for predicting microRNA-disease associations by unbalanced bi-random walk on heterogeneous network", 《JOURNAL OF BIOMEDICAL INFORMATICS》 *
XING, CHEN ET AL.: "RWRMDA: predicting novel human microRNA-disease associations", 《MOLECULAR BIOSYSTEMS》 *
刘晓燕等: "一种预测miRNA与疾病关联关系的矩阵分解算法", 《智能系统学报》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111681705A (en) * 2020-05-21 2020-09-18 中国科学院深圳先进技术研究院 miRNA-disease association prediction method, system, terminal and storage medium
WO2021232789A1 (en) * 2020-05-21 2021-11-25 中国科学院深圳先进技术研究院 Mirna-disease association prediction method, system, terminal, and storage medium
CN111681705B (en) * 2020-05-21 2024-05-24 中国科学院深圳先进技术研究院 MiRNA-disease association prediction method, system, terminal and storage medium
CN114783597A (en) * 2022-06-23 2022-07-22 瀚依科技(杭州)有限公司 Method and device for diagnosing multi-class diseases, electronic equipment and storage medium
CN114783597B (en) * 2022-06-23 2022-11-29 瀚依科技(杭州)有限公司 Method and device for diagnosing multi-class diseases, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN110853763B (en) 2022-07-19

Similar Documents

Publication Publication Date Title
Wang et al. LDGRNMF: LncRNA-disease associations prediction based on graph regularized non-negative matrix factorization
CN107506608B (en) Improved miRNA-disease association prediction method based on collaborative filtering
Kong et al. Spot: Locating social media users based on social network context
CN110459264B (en) Method for predicting relevance of circular RNA and diseases based on gradient enhanced decision tree
Zhu et al. A two-stage geometric method for pruning unreliable links in protein-protein networks
Wang et al. DeepDRK: a deep learning framework for drug repurposing through kernel-based multi-omics integration
Ou-Yang et al. Detecting protein complexes from signed protein-protein interaction networks
CN108121896B (en) Disease relation analysis method and device based on miRNA
CN110853763B (en) Fusion attribute-based miRNA-disease association identification method and system
Su et al. Recognition of small molecule–RNA binding sites using RNA sequence and structure
Zheng et al. CGMDA: an approach to predict and validate MicroRNA-disease associations by utilizing chaos game representation and LightGBM
CN103455612B (en) Based on two-stage policy non-overlapped with overlapping network community detection method
Gao et al. Graph regularized L 2, 1-nonnegative matrix factorization for miRNA-disease association prediction
Zhang et al. SGNNMD: signed graph neural network for predicting deregulation types of miRNA-disease associations
CN110782948A (en) Method for predicting potential association of miRNA and disease based on constraint probability matrix decomposition method
CN106202999A (en) Microorganism high-pass sequencing data based on different scale tuple word frequency analyzes agreement
Baba et al. Predicting large RNA-like topologies by a knowledge-based clustering approach
Pizzuti et al. An evolutionary restricted neighborhood search clustering approach for PPI networks
CN110534153B (en) Target prediction system and method based on deep learning
Li et al. Towards semi-supervised ensemble clustering using a new membership similarity measure
Wu On biological validity indices for soft clustering algorithms for gene expression data
Chandrasekhar et al. Performance analysis of enhanced clustering algorithm for gene expression data
Hui et al. A Review of Random Walk-Based Method for the Identification of Disease Genes and Disease Modules
Zhang et al. Two-Stage Inference for LncRNA-Disease Associations Based on Diverse Heterogeneous Information Sources
Li et al. IMC-MDA: Prediction of miRNA-disease association based on induction matrix completion

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant