CN107887023A - A kind of microbial diseases Relationship Prediction method based on similitude and double random walks - Google Patents
A kind of microbial diseases Relationship Prediction method based on similitude and double random walks Download PDFInfo
- Publication number
- CN107887023A CN107887023A CN201711293802.6A CN201711293802A CN107887023A CN 107887023 A CN107887023 A CN 107887023A CN 201711293802 A CN201711293802 A CN 201711293802A CN 107887023 A CN107887023 A CN 107887023A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- disease
- microorganism
- mtd
- 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.)
- Pending
Links
Classifications
-
- 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
- G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
Abstract
The invention discloses a kind of microbial diseases Relationship Prediction method based on similitude and double random walks, function of diseases similitude is built first with disease gene relation and gene function affinity information；Disease Gaussian kernel similitude is built further according to known microbial diseases relation；The final similitude of disease is integrated on the basis of function of diseases similitude and Gaussian kernel similitude.The similitude of microorganism derives from the Gaussian kernel similitude of known microorganisms disease relationship.Again by microorganism affinity information, disease affinity information, it is known that microbial diseases relation information be integrated into a double-deck heterogeneous network.Microbial diseases Relationship Prediction is carried out in heterogeneous network using double random walk methods.The present invention can be predicted to the relation between microbial diseases, and basic basis are provided for Bioexperiment, save its manpower and materials cost.
Description
Technical field
The invention belongs to system biology field, is related to a kind of microorganism-disease based on similitude and double random walks
Relationship Prediction method.
Background technology
Increasing research shows that microorganism plays very important effect in many mankind's complex diseases.With mesh
The fast development of preceding DNA sequencing technology of future generation promotes microorganism and the discovery of disease association relation between human body, than
Such as micropopulation and various Cancerous diseases, angiocardiopathy, metabolic syndrome (such as obesity and diabetes), central nervous system
System disease and auto-inflammatory disease etc..These researchs not only help the understanding to disease mechanisms, are also beneficial to disease
New treatment and the development of diagnosis scheme.For example confirm that fecal microorganism group transplanting is to treat safely and effectively controlling for clostridium infection
Treatment scheme, by being reintroduced back to normal flora to donor excrement, correction imbalance, which is laid equal stress on, newly-built founds normal gut function for it.Institute
To become more and more urgent to biology and the system understanding of the relation of disease.
What is be commonly used is by the discovery microorganism of the method based on experiment of routine and the relation of disease, its shortcoming
It is to be time-consuming and expensive, while is also limited by experimental situation, for example some bacteriums can not be in existing planting experiment room environmental
In cultivated.At the same time, mode is predicted to the relation between microorganism and disease by computation model not having
To application development energetically.Up to the present, it is few by computation model the relation of microorganism and disease to be predicted
Method occurs.KATZHMDA methods be at present by the relation of known microorganism and disease come be predicted new microorganism with
The first model of the relation of disease, it characterizes similitude, Gaussian kernel similitude and microorganism Gaussian kernel by integrated disease
Similitude, the relation of new microorganism and disease is predicted using KATZ degree information.Research of the present microorganism in terms of calculating,
In terms of mostly concentrating on microorganism classification, and it is seriously inadequate with the concern of the relation of disease to it.Currently to microorganism and disease
Relation calculating forecast model development degree and prediction result to be also not enough to allow Bioexperiment personnel to recognize computation model pre-
The validity of survey, and further in this, as the basis of subsequent experimental research.
It is limited by the efficiency of Bioexperiment checking, by the concern of computation model predictive microbiology and the relation of disease and enters
Exhibition is not enough and its prediction result needs further to be improved, and currently still has to the system understanding of microorganism and the relation of disease
Limit.There is an urgent need to propose significantly more efficient forecast model, existing biological information is made full use of, by way of more science
It was found that new microorganism and the relation of disease, the research for its follow-up Relationship Prediction lays the foundation, and is further Bioexperiment
Research provides important basic foundation.In addition, with the development of current calculation biology and DNA sequencing technology of future generation, understand
Significance level more and more higher to microorganism to disease, and then the development to microorganism and the Relationship Prediction model of disease proposes
Urgent need.Therefore, in order to further confirm that the importance of the relation of microorganism and disease, for its follow-up Relationship Prediction model
Development and Bioexperiment checking help is provided, it is necessary to design the side that a kind of effective microbial diseases incidence relation is predicted
Method.
The content of the invention
The technical problems to be solved by the invention are, in view of the shortcomings of the prior art, there is provided one kind is based on similitude and double
The microorganism of random walk-disease relationship Forecasting Methodology, be capable of the relation of accurate predictive microbiology and disease, for it is follow-up its
The development of Relationship Prediction model provides basis and a large amount of manpower and materials for further effectively avoiding Biochemistry Experiment from being consumed.
The technical solution of invention is as follows：
A kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks, comprises the following steps：
Step 1：Function of diseases similarity matrix D is built respectivelyfunsim, disease Gaussian kernel similarity matrix KGIP,dWith micro- life
Thing Gaussian kernel similarity matrix KGIP,m；
Step 2：Integrated function of diseases similarity matrix DfunsimWith disease Gaussian kernel similarity matrix KGIP,dObtain disease most
Whole similarity matrix Sd；By microorganism Gaussian kernel similarity matrix KGIP,mAs the final similarity matrix S of microorganismm；
Step 3：According to known microorganism-disease relationship, the final similarity matrix S of microorganismmIt is finally similar with disease
Property matrix Sd, double-deck heterogeneous network is built, using double random walk methods to microorganism-disease to being associated fraction
Prediction；Associated score is bigger, then corresponding microorganism-disease is to there is a possibility that relation is bigger.
Further, in the step 1, the functional similarity meter of disease-gene relationship and gene first known to
The functional similarity between two kinds of diseases is calculated, it is similar then to build function of diseases by the functional similarity of all diseases between any two
Property matrix Dfunsim；
It is as follows for any two kinds of diseases A and B, its functional similarity calculation formula：
Wherein, GA={ gA1,gA2,......,gAmIt is the gene sets associated with disease A, equally, GB={ gB1,
gB2,......,gBnIt is the gene sets associated with disease B, m and n are respectively gene sets GAAnd GBIn number gene；For gene gAiWith gene sets GBFunctional similarity value,For gene gBjWith gene sets GAWork(
Energy similarity, corresponding calculation formula are as follows：
Wherein F (gAi,gBj) it is gene gAiAnd gBjBetween Semantic Similarity value, HumanNet databases, which provide, to be based on
The Semantic Similarity value calculated value of log-likelihood function, specific calculation are as follows：
F(gAi,gBj)=LLS (gAi,gBj).
Wherein LLS represents that log-likelihood function (in HumanNet databases, gene language is calculated using log-likelihood function
Adopted similarity is prior art).
Further, in the step 1, according to known microorganism-disease relationship, it is similar that disease Gaussian kernel is built respectively
Property matrix KGIP,dWith microorganism Gaussian kernel similarity matrix KGIP,m, process is as follows：
First, defineFor the set of microorganism, NmFor the quantity of microorganism；
For the set of all diseases, NdFor the quantity of disease；Adjacency matrix Y ∈ Nm×NdRepresent whether deposited between each microorganism and disease
In known relation；If microorganism miWith disease djIn the presence of known incidence relation then yijIt is worth for 1, otherwise value is 0；
Then, the Gaussian kernel similitude of all diseases between any two is calculated；For any two kinds of disease d1And d2, its Gauss
Core similitude calculation is as follows：
KGIP,d(d1,d2)=exp (- γd||yd1-yd2||2)
Wherein,
γdTo control the adjustment parameter of core width, γ 'dFor disease bandwidth parameter, 1 is arranged to according to (Gaussian kernel use) experience；
The Gaussian kernel similitude of all microorganisms between any two is calculated again；For any two kinds of microorganism m1And m2, its Gauss
Core similitude calculation is as follows：
KGIP,m(m1,m2)=exp (- γm||ym1-ym2||2).
Wherein,γm
To control the adjustment parameter of core width, γ 'mFor microorganism bandwidth parameter, 1 is empirically arranged to；、
Finally, disease Gaussian kernel similarity matrix K is built by the Gaussian kernel similitude of all diseases between any twoGIP,d, by
The Gaussian kernel similitude structure microorganism Gaussian kernel similarity matrix K of all microorganisms between any twoGIP,m。
Further, in the step 2, function of diseases similarity matrix D is integratedfunsimWith disease Gaussian kernel similitude square
Battle array KGIP,dObtain the final similarity matrix S of diseased, specific integration mode is calculated as follows：
I.e. the final similitude of disease is functional similarity and the average value of Gaussian kernel similitude.
Further, in the step 5, according to the final similitude S of microorganismm, the final similitude S of diseased, it is known that it is micro-
Biology-disease data adjacency matrix Y integrates double-deck heterogeneous network, continues to predict using double random walk methods, its is pre-
Flow gauge is as follows：
First, similarity matrix S final to microorganismmData do row normalization processing, obtain the microorganism of random walk
Similarity relationships matrix MM ∈ Nm×Nm, its calculation is as follows：
Equally, similarity matrix S final to diseasedData do row normalization processing, and the disease for obtaining random walk is similar
Sexual intercourse matrix MD ∈ Nd×Nd, its calculation is as follows：
Then, migration simultaneously, process are as follows in this double-deck heterogeneous network：
Iteration carries out left migration in microorganism network：
Iteration carries out right migration in disease network：
Wherein, t be current iteration number, Pt∈Nm×NdRepresent that microorganism-disease that the t times iteration is predicted to obtain is closed
Join score matrix, Pt(i, j) represents microorganism i and disease j associated score (correlation degree)；L_PtRepresent on microorganism network
Carry out new microorganism-disease association score matrix that the t times iteration is predicted to obtain, R_PtExpression carries out the on disease network
T iteration predicts obtained microorganism-disease association score matrix；P0For adjacency matrix Y ∈ Nm×NdNormalization matrix,
α is attenuation parameter, IlAnd IrRespectively microorganism network and disease network be most
Big iterations parameter, α, IlAnd IrValue rule of thumb or cross validation determine (set attenuation parameter value be 0.1, IlAnd Ir
Value be respectively 2 and 1)；LnumAnd RnumRespectively microorganism network and disease network have completed the number of iteration,
Work as PtRestrain (Pt+1-PtLess than some very little threshold value when (such as 10-10), it is believed that migration reaches stable state) or
When iteration migration of the person in microorganism network and disease network reaches maximum iteration, terminate iteration, final PtI.e.
To predict obtained microorganism-disease association score matrix.
Beneficial effect：
The present invention, which proposes, a kind of to be predicted based on similitude and microorganism-disease relationship Forecasting Methodology of double random walks
New microorganism and the relation of disease.Calculate first by disease gene relation and gene function affinity information to calculate disease
Functional similarity, the Gaussian kernel similitude of disease is calculated further according to known microorganism-disease relationship, is further integrated
The final similitude of disease.Equally, microorganism Gaussian kernel similitude and conduct are calculated also according to known microorganism-disease relationship
Final microorganism similitude.Again by the final similitude of microorganism, the final similitude of disease and known microorganism-disease relationship
It is integrated into double-deck heterogeneous network.Finally set in microorganism similitude network and disease similitude network it is different with
Migration step number is walked, final microbial diseases relation pair incidence relation fraction is predicted by iterating to certain stable state.It is logical
Cross five times to intersect and stay the prediction result of a checking and other method relatively to show, the present invention can be between microorganism and disease
Relation more effectively predicted.It can be provided for the development of the computation model of subsequent prediction microorganism-disease relationship important
Basis, basic directive function is provided for biomedicine experiment, save its manpower and materials cost.
The present invention enters during to the double-deck heterogeneous network of structure to microorganism similarity matrix and disease similarity matrix
Row normalization processing is gone.To being respectively provided with the iteration of random walk in microorganism and disease network in random walk process
Step number limits.It is similar by being associated with reference to the similar microorganism of disease association similar in two networks with similar microorganism
Disease predicts final associated score.And take and five times of cross validations same in KATZHMDA methods and stay a checking
Method has carried out the comparison of estimated performance, passes through the estimated performance of the analysis shows present invention to AUC indexs.
The present invention is directed to microorganism-disease relationship field, there is provided a kind of to predict the effective of its relation by computation model
Microorganism-disease association Relationship Prediction method, important foundation can be provided for the research of the computation model in this follow-up field,
Overall understanding to disease mechanisms provides help, and further promotes the exploitation of medicine and the diagnoses and treatment of complex disease.
Brief description of the drawings
Fig. 1 is overview flow chart of the present invention；
For the present invention, five times of cross validations on data set compare figure to Fig. 2；
Fig. 3 stays a cross validation to compare figure for the present invention on data set；
Embodiment
The present invention is described in further details below with reference to the drawings and specific embodiments：
Embodiment 1：
Function of diseases similitude is calculated first with disease gene relation and gene affinity information；Based on known micro- life
Thing-disease relationship calculates disease Gaussian kernel similitude and microorganism Gaussian kernel similitude；Utilize function of diseases similitude and Gauss
Core similitude integrates the final similitude of disease, and specific integration mode is to take disease Gaussian kernel similitude and function of diseases similitude equal
Value.Using microorganism Gaussian kernel similitude as the final similitude of microorganism.Microorganism affinity information, disease similitude are believed again
Breath and known microorganism-disease relationship information integration are into a double-deck heterogeneous network.Phase is associated based on similar microorganism
As disease and the starting point of the similar microorganism of similar disease association entered using double random walk methods in heterogeneous network
Row microorganism-disease relationship prediction.To the critical process of random walk method in microorganism similitude network and disease similitude
Difference is set to iterate to certain stable state with migration step number is walked and obtain final microorganism-disease relationship pair in network
Associated score.
Known microorganisms-disease relationship that the present invention uses comes from HMDAD (http://www.cuilab.cn/
Hmdad) database, altogether including 39 kinds of diseases and 292 kinds of microorganisms, its known microorganism-disease relationship number is 483.
Handled by duplicate removal, final relation number is 450, and disease and microbe number are respectively 39 and 292.Disease gene relation number
According to coming from DisGeNET databases.
The whole flow process of microorganism based on similitude and double random walks-disease relationship prediction is as shown in figure 1, can draw
It is divided into following steps：
(1) function of diseases similitude D is calculatedfunsimDetailed process be：
Firstly, for disease to A and B, it is as follows to define its functional similarity calculation formula：
Wherein, GA={ gA1,gA2,......,gAmIt is the gene sets associated with disease A, equally, GB={ gB1,
gB2,......,gBnIt is the gene sets associated with disease B, m and n are respectively gene sets GAAnd GBIn number gene；For gene gAiWith gene sets GBFunctional similarity value,For gene gBjWith gene sets GAFunction
Similarity, corresponding calculation formula are as follows：
Wherein F (gAi,gBj) it is gene gAiAnd gBjBetween Semantic Similarity value, HumanNet databases, which provide, to be based on
The Semantic Similarity value calculated value of log-likelihood function, specific calculation are as follows：
F(gAi,gBj)=LLS (gAi,gBj).
In HumanNet databases, the functional similarity value of gene 6188 and 6209 provided is 0.9697, according to disease
The gene of association, and the functional similarity of gene, disease Gastric and duodenal ulcer and Gastro-
Oesophageal reflux functional similarity value is 0.1655.
(2) microorganism-disease relationship known to, the process for building microorganism Gaussian kernel similitude are as follows：
First, defineFor the set of microorganism, NmFor the quantity of microorganism；
For the set of all diseases, NdFor the quantity of disease；Adjacency matrix Y ∈ Nm*NdRepresent to whether there is between each microorganism and disease
Known relation.If microorganism miWith disease djIn the presence of known incidence relation then yijIt is worth for 1, otherwise value is 0.Such as micro- life
Thing m1And m2Gaussian kernel Similarity measures mode be defined as follows：
KGIP,m(m1,m2)=exp (- γm||ym1-ym2||2).
Wherein,γm
It is as follows for the adjustment parameter of control core width, its calculation：
Wherein γ 'mAccording to Gaussian kernel 1 is arranged to using experience.According to above-mentioned calculation formula, microorganism
Actinobacillus and microorganism Actinobacteria Gaussian kernel similarity is 0.0390.
Equally, disease d is defined1And d2Gaussian kernel Similarity measures mode it is as follows：
KGIP,d(d1,d2)=exp (- γd||yd1-yd2||2)
Wherein, γ 'dEmpirically it is also configured as 1. ；According to above-mentioned calculation formula, disease Allergic asthma and disease Atopic
Dermatitis Gaussian kernel similarity is 0.4274.
(3) according to the function of diseases similitude D of calculatingfunsimWith disease Gaussian kernel similitude KGIP,dIntegrated final disease
Similitude process, specific integration mode are calculated as follows：
Final disease similitude is functional similarity and the average value of Gaussian kernel similitude.
(4) by microorganism Gaussian kernel similarity matrix KGIP,mAs the final similarity matrix S of microorganismm：
Sm=KGIP,m
Microorganism only has a Gaussian kernel similitude, therefore its final similitude SmFor Gaussian kernel similitude.
(5) according to the final similitude S of microorganismm, the final similitude S of diseased, it is known that microorganism-disease data integrated one
The individual double-deck heterogeneous network, continue to predict using double random walk methods, its pre- flow gauge is：
First, similarity matrix S final to microorganismmData do row normalization and handle to obtain the microbial of random walk
Like sexual intercourse matrix MM ∈ Nm*Nm, its calculation is as follows：
Equally, similarity matrix S final to diseasedData do row normalization and handle to obtain the disease similitude of random walk
Relational matrix MD ∈ Nd*Nd, its calculation is as follows：
We define matrix P ∈ Nm*NdMicroorganism-disease relationship of prediction is represented, P (i, j) represents microorganism i and disease j
Associated score (correlation degree).The prediction process of random walk model is the migration simultaneously in this double-deck heterogeneous network,
Still maximum iteration parameter I is respectively provided with to microorganism and disease networklAnd Ir.Walk process in heterogeneous network
It is as follows：
Left migration in microorganism network：
Right migration in disease network：
Wherein, t be current iteration number, Pt∈Nm×NdRepresent that microorganism-disease that the t times iteration is predicted to obtain is closed
Join score matrix, Pt(i, j) represents microorganism i and disease j associated score (correlation degree)；L_PtRepresent on microorganism network
Carry out new microorganism-disease association score matrix that the t times iteration is predicted to obtain, R_PtExpression carries out the on disease network
T iteration predicts obtained microorganism-disease association score matrix；P0For adjacency matrix Y ∈ Nm×NdNormalization matrix,
α is attenuation parameter, IlAnd IrRespectively microorganism network and disease network be most
Big iterations parameter, α, IlAnd IrValue rule of thumb or cross validation determine (in the present embodiment, rule of thumb tested with intersection
Card, the value for setting attenuation parameter are 0.1, maximum iteration parameter IlAnd IrValue be respectively 2 and 1)；LnumAnd RnumRespectively
Microorganism network and disease network have completed the number of iteration,
Work as PtRestrain (Pt+1-PtLess than some very little threshold value when (such as 10-10), it is believed that migration reaches stable state) or
When iteration migration of the person in microorganism network and disease network reaches maximum iteration, terminate iteration, final PtI.e.
To predict obtained microorganism-disease association score matrix.It is descending to each associated score in matrix to be ranked up, ranking
More forward microorganism-disease is to there is a possibility that incidence relation is bigger.
In order to verify effectiveness of the invention, we refer to the validation criteria of other algorithms, employ two kinds of verification modes：
(1) 5 times of cross validation；(2) checking is stayed.In five times of cross validations, known microorganism-disease relationship is randomly divided into 5
Part, it is test set to select 1 part in turn successively, and remaining 4 parts are training set, and it is 100 times that it, which tests checking number,.Staying a checking
In, it is test set to select a known microorganisms-disease relationship from known microorganism-disease relationship successively, remaining to be
Training set.The evaluation index used is AUC (the areas under ROC curves) value.
Fig. 2 shows the AUC figures of integrated function of diseases similitude and Gaussian kernel similitude in five times of cross validations.From figure
In as can be seen that the present invention (Predicting microbe-disease interactions based on
Similarities and bi-random walk on the heterogeneous network, referred to as BRWH-MDI)
AUC be 0.8676, similitude and microorganism Gaussian kernel are characterized based on disease Gaussian kernel similitude, disease better than other 3
Method (the KATZHMDA of similitude：0.8567, HGBI：0.7762, NBI：0.5622).It is particularly low in error rate (FPR values)
When, accuracy (TPR values) is higher, it was demonstrated that ranking preceding microorganism-disease relationship is got in the prediction result of the present invention
Correctly.
Fig. 3 describes the performance comparision of integrated function of diseases similitude and Gaussian kernel the similitude each method in a checking is stayed
Figure.It can also be seen that BRWH of the present invention AUC is 0.8780 from figure, the similarly performance due to other 3 methods
(KATZHMDA：0.8644, HGBI：0.7866, NBI：5553).Equally when error rate (FPR values) is low, accuracy (TPR
Value) it is higher, it also show the more high accuracy of the preceding microorganism-disease relationship of ranking in the prediction result of the present invention.
By the performance of above-mentioned application case, the present invention accurate can predict new microorganism-disease relationship, after being
Continuous biomedicine experiment provides directive function, improves medical diagnosis on disease and treatment level.
Claims (5)
1. a kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks, it is characterised in that including following
Step：
Step 1：Function of diseases similarity matrix D is built respectivelyfunsim, disease Gaussian kernel similarity matrix KGIP,dWith microorganism height
This core similitude matrix KGIP,m；
Step 2：Integrated function of diseases similarity matrix DfunsimWith disease Gaussian kernel similarity matrix KGIP,dObtain disease most last phase
Like property matrix Sd；By microorganism Gaussian kernel similarity matrix KGIP,mAs the final similarity matrix S of microorganismm；
Step 3：According to known microorganism-disease relationship, the final similarity matrix S of microorganismmWith the final similarity matrix of disease
Sd, double-deck heterogeneous network is built, using double random walk methods to microorganism-disease to being associated Score on Prediction；
Associated score is bigger, then corresponding microorganism-disease is to there is a possibility that relation is bigger.
2. a kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks according to claim 1,
Characterized in that, in the step 1, the functional similarity of disease-gene relationship and gene calculates two kinds known to first
Functional similarity between disease, function of diseases similarity matrix is then built by the functional similarity of all diseases between any two
Dfunsim；
It is as follows for any two kinds of diseases A and B, its functional similarity calculation formula：
<mrow>
<msub>
<mi>D</mi>
<mrow>
<mi>f</mi>
<mi>u</mi>
<mi>n</mi>
<mi>s</mi>
<mi>i</mi>
<mi>m</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>A</mi>
<mo>,</mo>
<mi>B</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>i</mi>
<mo>&le;</mo>
<mi>m</mi>
</mrow>
</munder>
<msub>
<mi>FS</mi>
<msub>
<mi>G</mi>
<mi>B</mi>
</msub>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>+</mo>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>j</mi>
<mo>&le;</mo>
<mi>n</mi>
</mrow>
</munder>
<msub>
<mi>FS</mi>
<msub>
<mi>G</mi>
<mi>A</mi>
</msub>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>B</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>m</mi>
<mo>+</mo>
<mi>n</mi>
</mrow>
</mfrac>
<mo>,</mo>
</mrow>
Wherein, GA={ gA1,gA2,......,gAmIt is the gene sets associated with disease A, equally, GB={ gB1,
gB2,......,gBnIt is the gene sets associated with disease B, m and n are respectively gene sets GAAnd GBIn number gene；For gene gAiWith gene sets GBFunctional similarity value,For gene gBjWith gene sets GAFunction
Similarity, corresponding calculation formula are as follows：
<mrow>
<msub>
<mi>FS</mi>
<msub>
<mi>G</mi>
<mi>B</mi>
</msub>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
</mrow>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>j</mi>
<mo>&le;</mo>
<mi>n</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mi>F</mi>
<mo>(</mo>
<mrow>
<msub>
<mi>g</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>B</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
<mrow>
<msub>
<mi>FS</mi>
<msub>
<mi>G</mi>
<mi>A</mi>
</msub>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>B</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munder>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
</mrow>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>i</mi>
<mo>&le;</mo>
<mi>m</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<mi>F</mi>
<mo>(</mo>
<mrow>
<msub>
<mi>g</mi>
<mrow>
<mi>A</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>g</mi>
<mrow>
<mi>B</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
Wherein F (gAi,gBj) it is gene gAiAnd gBjBetween Semantic Similarity value, HumanNet databases are provided based on logarithm
The Semantic Similarity value calculated value of likelihood function, specific calculation are as follows：
F(gAi,gBj)=LLS (gAi,gBj).
Wherein LLS represents that log-likelihood function (in HumanNet databases, gene semantic phase is calculated using log-likelihood function
It is prior art like property value).
3. a kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks according to claim 1,
Characterized in that, in the step 1, according to known microorganism-disease relationship, disease Gaussian kernel similarity matrix is built respectively
KGIP,dWith microorganism Gaussian kernel similarity matrix KGIP,m, process is as follows：
First, defineFor the set of microorganism, NmFor the quantity of microorganism；
For the set of all diseases, NdFor the quantity of disease；Adjacency matrix Y ∈ Nm×NdRepresent whether deposited between each microorganism and disease
In known relation；If microorganism miWith disease djIn the presence of known incidence relation then yijIt is worth for 1, otherwise value is 0；
Then, the Gaussian kernel similitude of all diseases between any two is calculated；For any two kinds of disease d1And d2, its Gauss nuclear phase
It is as follows like property calculation：
KGIP,d(d1,d2)=exp (- γd||yd1-yd2||2)
<mrow>
<msub>
<mi>&gamma;</mi>
<mi>d</mi>
</msub>
<mo>=</mo>
<msub>
<msup>
<mi>&gamma;</mi>
<mo>&prime;</mo>
</msup>
<mi>d</mi>
</msub>
<mo>/</mo>
<mrow>
<mo>(</mo>
<mfrac>
<mn>1</mn>
<msub>
<mi>N</mi>
<mi>d</mi>
</msub>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>d</mi>
</msub>
</munderover>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>yd</mi>
<mi>j</mi>
</msub>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
Wherein,γdFor
Control the adjustment parameter of core width, γ 'dFor disease bandwidth parameter, 1 is arranged to according to (Gaussian kernel use) experience；
The Gaussian kernel similitude of all microorganisms between any two is calculated again；For any two kinds of microorganism m1And m2, its Gauss nuclear phase
It is as follows like property calculation：
KGIP,m(m1,m2)=exp (- γm||ym1-ym2||2).
<mrow>
<msub>
<mi>&gamma;</mi>
<mi>m</mi>
</msub>
<mo>=</mo>
<msub>
<msup>
<mi>&gamma;</mi>
<mo>&prime;</mo>
</msup>
<mi>m</mi>
</msub>
<mo>/</mo>
<mrow>
<mo>(</mo>
<mfrac>
<mn>1</mn>
<msub>
<mi>N</mi>
<mi>m</mi>
</msub>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>m</mi>
</msub>
</munderover>
<mo>|</mo>
<mo>|</mo>
<msub>
<mi>ym</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>)</mo>
</mrow>
<mo>,</mo>
</mrow>
Wherein,γmFor control
The adjustment parameter of core width processed, γ 'mFor microorganism bandwidth parameter, 1 is empirically arranged to；、
Finally, disease Gaussian kernel similarity matrix K is built by the Gaussian kernel similitude of all diseases between any twoGIP,d, by owning
The Gaussian kernel similitude structure microorganism Gaussian kernel similarity matrix K of microorganism between any twoGIP,m。
4. a kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks according to claim 1,
Characterized in that, in the step 2, function of diseases similarity matrix D is integratedfunsimWith disease Gaussian kernel similarity matrix KGIP,d
Obtain the final similarity matrix S of diseased, specific integration mode is calculated as follows：
<mrow>
<msub>
<mi>S</mi>
<mi>d</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>D</mi>
<mrow>
<mi>f</mi>
<mi>u</mi>
<mi>n</mi>
<mi>s</mi>
<mi>i</mi>
<mi>m</mi>
</mrow>
</msub>
<mo>+</mo>
<msub>
<mi>K</mi>
<mrow>
<mi>G</mi>
<mi>I</mi>
<mi>P</mi>
<mo>,</mo>
<mi>d</mi>
</mrow>
</msub>
</mrow>
<mn>2</mn>
</mfrac>
</mrow>
I.e. the final similitude of disease is functional similarity and the average value of Gaussian kernel similitude.
5. a kind of microorganism-disease relationship Forecasting Methodology based on similitude and double random walks according to claim 1,
Characterized in that, in the step 5, according to the final similitude S of microorganismm, the final similitude S of diseased, it is known that microorganism-disease
Sick data adjacency matrix Y integrates double-deck heterogeneous network, continues to predict using double random walk methods, its pre- flow gauge is such as
Under：
First, similarity matrix S final to microorganismmData do row normalization processing, obtain the microorganism similitude of random walk
Relational matrix MM ∈ Nm×Nm, its calculation is as follows：
<mrow>
<mi>M</mi>
<mi>M</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mi>S</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>k</mi>
<mo>=</mo>
<msub>
<mi>N</mi>
<mi>m</mi>
</msub>
</mrow>
</msubsup>
<msub>
<mi>S</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mtd>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>k</mi>
<mo>=</mo>
<msub>
<mi>N</mi>
<mi>m</mi>
</msub>
</mrow>
</msubsup>
<msub>
<mi>S</mi>
<mi>m</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>></mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>o</mi>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>r</mi>
<mi>w</mi>
<mi>i</mi>
<mi>s</mi>
<mi>e</mi>
<mo>.</mo>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Equally, similarity matrix S final to diseasedData do row normalization processing, obtain the disease similarity relationships of random walk
Matrix MD ∈ Nd×Nd, its calculation is as follows：
<mrow>
<mi>M</mi>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mi>S</mi>
<mi>d</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>k</mi>
<mo>=</mo>
<msub>
<mi>N</mi>
<mi>d</mi>
</msub>
</mrow>
</msubsup>
<msub>
<mi>S</mi>
<mi>d</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mtd>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>k</mi>
<mo>=</mo>
<msub>
<mi>N</mi>
<mi>d</mi>
</msub>
</mrow>
</msubsup>
<msub>
<mi>S</mi>
<mi>d</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>></mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>o</mi>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>r</mi>
<mi>w</mi>
<mi>i</mi>
<mi>s</mi>
<mi>e</mi>
<mo>.</mo>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Then, migration simultaneously, process are as follows in this double-deck heterogeneous network：
Iteration carries out left migration in microorganism network：
<mrow>
<mi>L</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<mi>t</mi>
</msub>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>&alpha;</mi>
<mo>&times;</mo>
<mi>M</mi>
<mi>M</mi>
<mo>&times;</mo>
<msub>
<mi>P</mi>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>+</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<mi>&alpha;</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>Y</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>t</mi>
<mo>&le;</mo>
<msub>
<mi>I</mi>
<mi>l</mi>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>L</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<msub>
<mi>I</mi>
<mi>l</mi>
</msub>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>t</mi>
<mo>></mo>
<msub>
<mi>I</mi>
<mi>l</mi>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Iteration carries out right migration in disease network：
<mrow>
<mi>R</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<mi>t</mi>
</msub>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<mi>&alpha;</mi>
<mo>)</mo>
<mo>&times;</mo>
<msub>
<mi>P</mi>
<mrow>
<mi>t</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>&times;</mo>
<mi>M</mi>
<mi>D</mi>
<mo>+</mo>
<mi>&alpha;</mi>
<mo>&times;</mo>
<mi>Y</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>t</mi>
<mo>&le;</mo>
<msub>
<mi>I</mi>
<mi>r</mi>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>R</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<msub>
<mi>I</mi>
<mi>r</mi>
</msub>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>t</mi>
<mo>></mo>
<msub>
<mi>I</mi>
<mi>r</mi>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
<mrow>
<msub>
<mi>P</mi>
<mi>t</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mi>L</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<mi>t</mi>
</msub>
<mo>&times;</mo>
<msub>
<mi>L</mi>
<mrow>
<mi>n</mi>
<mi>u</mi>
<mi>m</mi>
</mrow>
</msub>
<mo>+</mo>
<mi>R</mi>
<mo>_</mo>
<msub>
<mi>P</mi>
<mi>t</mi>
</msub>
<mo>&times;</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>n</mi>
<mi>u</mi>
<mi>m</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>L</mi>
<mrow>
<mi>n</mi>
<mi>u</mi>
<mi>m</mi>
</mrow>
</msub>
<mo>+</mo>
<msub>
<mi>R</mi>
<mrow>
<mi>n</mi>
<mi>u</mi>
<mi>m</mi>
</mrow>
</msub>
</mrow>
</mfrac>
</mrow>
Wherein, t be current iteration number, Pt∈Nm×NdRepresent that the t times iteration predicts obtained microorganism-disease association point
Matrix number, Pt(i, j) represents microorganism i and disease j associated score (correlation degree)；L_PtRepresent to carry out on microorganism network
The t times iteration predicts obtained new microorganism-disease association score matrix, R_PtExpression is carried out the t times on disease network
Iteration predicts obtained microorganism-disease association score matrix；P0For adjacency matrix Y ∈ Nm×NdNormalization matrix,α is attenuation parameter, IlAnd IrRespectively microorganism network and disease network greatest iteration time
Number parameter, α, IlAnd IrValue rule of thumb or cross validation determine (set attenuation parameter value be 0.1, IlAnd IrValue difference
For 2 and 1)；LnumAnd RnumRespectively microorganism network and disease network have completed the number of iteration,
Work as PtRestrain (Pt+1-PtLess than some very little threshold value when (such as 10-10), it is believed that migration reaches stable state) or
When iteration migration in microorganism network and disease network reaches maximum iteration, terminate iteration, final PtIt is as pre-
The microorganism measured-disease association score matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711293802.6A CN107887023A (en) | 2017-12-08 | 2017-12-08 | A kind of microbial diseases Relationship Prediction method based on similitude and double random walks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711293802.6A CN107887023A (en) | 2017-12-08 | 2017-12-08 | A kind of microbial diseases Relationship Prediction method based on similitude and double random walks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107887023A true CN107887023A (en) | 2018-04-06 |
Family
ID=61773209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711293802.6A Pending CN107887023A (en) | 2017-12-08 | 2017-12-08 | A kind of microbial diseases Relationship Prediction method based on similitude and double random walks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107887023A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647484A (en) * | 2018-05-17 | 2018-10-12 | 中南大学 | A kind of drug relationship prediction technique integrated based on multiple information with least square method |
| CN108877953A (en) * | 2018-06-06 | 2018-11-23 | 中南大学 | A kind of drug sensitivity prediction method based on more similitude networks |
| CN109920478A (en) * | 2019-03-07 | 2019-06-21 | 中南大学 | A kind of microorganism-disease relationship prediction technique filled based on similitude and low-rank matrix |
| 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 |
| CN110349619A (en) * | 2019-01-17 | 2019-10-18 | 哈尔滨工业大学 | Disease and metabolin network establishing method based on similarity calculation |
| CN110610765A (en) * | 2019-09-10 | 2019-12-24 | 陕西师范大学 | Method for predicting disease-related metabolites by using double random walks |
| CN110610763A (en) * | 2019-09-10 | 2019-12-24 | 陕西师范大学 | KaTZ model-based metabolite and disease association relation prediction method |
| CN111653359A (en) * | 2020-05-30 | 2020-09-11 | 吾征智能技术(北京)有限公司 | Intelligent prediction model construction method and prediction system for hemorrhagic diseases |
| CN112151191A (en) * | 2020-10-16 | 2020-12-29 | 山东管理学院 | Microorganism and disease association relation prediction method and system based on attention mechanism |
| CN112599202A (en) * | 2020-12-24 | 2021-04-02 | 南华大学 | Disease-related miRNA prediction system |
| CN112820347A (en) * | 2021-02-02 | 2021-05-18 | 中南大学 | Disease gene prediction method based on multiple protein network pulse dynamics process |
| CN113192562A (en) * | 2021-05-07 | 2021-07-30 | 中南大学 | Pathogenic gene identification method and system fusing multi-scale module structure information |
| CN116052873A (en) * | 2023-01-18 | 2023-05-02 | 齐齐哈尔大学 | Disease-metabolite association prediction system based on weight k-nearest neighbor |
| CN112599202B (en) * | 2020-12-24 | 2024-04-26 | 南华大学 | Disease-related miRNA prediction system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007100834A3 (en) * | 2006-02-27 | 2008-07-31 | Univ California | Graph querying, graph motif mining and the discovery of clusters |
| US20130064439A1 (en) * | 2011-02-23 | 2013-03-14 | Siemens Corporation | Systems and Method for Automatic Prostate Localization in MR Images Using Random Walker Segmentation Initialized Via Boosted Classifiers |
| CN105653846A (en) * | 2015-12-25 | 2016-06-08 | 中南大学 | Integrated similarity measurement and bi-directional random walk based pharmaceutical relocation method |
-
2017
- 2017-12-08 CN CN201711293802.6A patent/CN107887023A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007100834A3 (en) * | 2006-02-27 | 2008-07-31 | Univ California | Graph querying, graph motif mining and the discovery of clusters |
| US20130064439A1 (en) * | 2011-02-23 | 2013-03-14 | Siemens Corporation | Systems and Method for Automatic Prostate Localization in MR Images Using Random Walker Segmentation Initialized Via Boosted Classifiers |
| CN105653846A (en) * | 2015-12-25 | 2016-06-08 | 中南大学 | Integrated similarity measurement and bi-directional random walk based pharmaceutical relocation method |
Non-Patent Citations (2)
| Title |
|---|
| CHENG YAN ETAL: ""DNRLMF-MDA:Predicting microRNA-Disease Associations Based on Similarities of microRNAs and Diseases"", 《HTTPS://IEEEXPLORE.IEEE.ORG/DOCUMENT/8118134》 * |
| SHUAIZOU ETAL: ""A novel approach for predicting microbe_disease associations by bi-random walk on the heterogeneous network"", 《PLOS ONE》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647484B (en) * | 2018-05-17 | 2020-10-23 | 中南大学 | Medicine relation prediction method based on multivariate information integration and least square method |
| CN108647484A (en) * | 2018-05-17 | 2018-10-12 | 中南大学 | A kind of drug relationship prediction technique integrated based on multiple information with least square method |
| CN108877953B (en) * | 2018-06-06 | 2021-09-03 | 中南大学 | Drug sensitivity prediction method based on multi-similarity network |
| CN108877953A (en) * | 2018-06-06 | 2018-11-23 | 中南大学 | A kind of drug sensitivity prediction method based on more similitude networks |
| CN110349619A (en) * | 2019-01-17 | 2019-10-18 | 哈尔滨工业大学 | Disease and metabolin network establishing method based on similarity calculation |
| 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 |
| CN109920478A (en) * | 2019-03-07 | 2019-06-21 | 中南大学 | A kind of microorganism-disease relationship prediction technique filled based on similitude and low-rank matrix |
| CN109920478B (en) * | 2019-03-07 | 2020-12-08 | 中南大学 | Microorganism-disease relation prediction method based on similarity and low-rank matrix filling |
| CN110610765A (en) * | 2019-09-10 | 2019-12-24 | 陕西师范大学 | Method for predicting disease-related metabolites by using double random walks |
| CN110610763A (en) * | 2019-09-10 | 2019-12-24 | 陕西师范大学 | KaTZ model-based metabolite and disease association relation prediction method |
| CN111653359A (en) * | 2020-05-30 | 2020-09-11 | 吾征智能技术(北京)有限公司 | Intelligent prediction model construction method and prediction system for hemorrhagic diseases |
| CN111653359B (en) * | 2020-05-30 | 2023-09-08 | 吾征智能技术(北京)有限公司 | Intelligent prediction model construction method and prediction system for hemorrhagic disease |
| CN112151191A (en) * | 2020-10-16 | 2020-12-29 | 山东管理学院 | Microorganism and disease association relation prediction method and system based on attention mechanism |
| CN112151191B (en) * | 2020-10-16 | 2023-02-03 | 山东管理学院 | Microorganism and disease association relation prediction method and system based on attention mechanism |
| CN112599202A (en) * | 2020-12-24 | 2021-04-02 | 南华大学 | Disease-related miRNA prediction system |
| CN112599202B (en) * | 2020-12-24 | 2024-04-26 | 南华大学 | Disease-related miRNA prediction system |
| CN112820347A (en) * | 2021-02-02 | 2021-05-18 | 中南大学 | Disease gene prediction method based on multiple protein network pulse dynamics process |
| CN112820347B (en) * | 2021-02-02 | 2023-09-22 | 中南大学 | Disease gene prediction method based on multiple protein network pulse dynamics process |
| CN113192562A (en) * | 2021-05-07 | 2021-07-30 | 中南大学 | Pathogenic gene identification method and system fusing multi-scale module structure information |
| CN113192562B (en) * | 2021-05-07 | 2022-05-13 | 中南大学 | Pathogenic gene identification method and system fusing multi-scale module structure information |
| CN116052873A (en) * | 2023-01-18 | 2023-05-02 | 齐齐哈尔大学 | Disease-metabolite association prediction system based on weight k-nearest neighbor |
| CN116052873B (en) * | 2023-01-18 | 2024-01-26 | 齐齐哈尔大学 | Disease-metabolite association prediction system based on weight k-nearest neighbor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107887023A (en) | A kind of microbial diseases Relationship Prediction method based on similitude and double random walks | |
| CN107862179A (en) | A kind of miRNA disease association Relationship Prediction methods decomposed based on similitude and logic matrix | |
| CN105653846B (en) | Drug method for relocating based on integrated similarity measurement and random two-way migration | |
| CN111696678B (en) | Drug administration decision method and system based on deep learning | |
| Mandal et al. | Cancer classification using neural network | |
| CN106485096B (en) | The miRNA- Relationship To Environmental Factors prediction technique learnt based on random two-way migration and multi-tag | |
| CN107610784A (en) | A kind of method of predictive microbiology and disease relationship | |
| Zhao et al. | Microbes and complex diseases: from experimental results to computational models | |
| CN106815486A (en) | A kind of system pharmacology method of personalized medicine | |
| CN106055922A (en) | Hybrid network gene screening method based on gene expression data | |
| CN109935337A (en) | A kind of medical record lookup method and system based on similarity measurement | |
| Sapna et al. | Implementation of genetic algorithm in predicting diabetes | |
| CN108647484A (en) | A kind of drug relationship prediction technique integrated based on multiple information with least square method | |
| CN114334038A (en) | Disease drug prediction method based on heterogeneous network embedded model | |
| CN111599485A (en) | Infectious disease propagation law prediction method, device, equipment and storage medium | |
| CN105808976A (en) | Recommendation model based miRNA target gene prediction method | |
| CN113178234B (en) | Compound function prediction method based on neural network and connection graph algorithm | |
| CN106326688B (en) | A kind of higher-dimension lacks the choosing method of the gene, signal path and related protein of sample | |
| Liu et al. | Combination of neuro-fuzzy network models with biological knowledge for reconstructing gene regulatory networks | |
| CN107644678A (en) | A kind of method that algorithm predictive microbiology and disease relationship are inferred based on network | |
| Song et al. | Prediction of drug-related diseases through integrating pairwise attributes and neighbor topological structures | |
| CN109920478A (en) | A kind of microorganism-disease relationship prediction technique filled based on similitude and low-rank matrix | |
| Liu et al. | Application of Training Load Prediction Model based on Improved BP Neural Network in Sports Training of Athletes | |
| Morell et al. | Latent variable regression discontinuity design with cluster level treatment assignment | |
| US20240136025A1 (en) | Compound function prediction method based on neural network and connectivity map algorithm |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180406 |