CN108763862A - A kind of active method of derivation gene pathway - Google Patents
A kind of active method of derivation gene pathway Download PDFInfo
- Publication number
- CN108763862A CN108763862A CN201810422205.7A CN201810422205A CN108763862A CN 108763862 A CN108763862 A CN 108763862A CN 201810422205 A CN201810422205 A CN 201810422205A CN 108763862 A CN108763862 A CN 108763862A
- Authority
- CN
- China
- Prior art keywords
- gene
- expression value
- sample
- expression
- interaction
- 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
Links
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The present invention provides a kind of active method of derivation gene pathway, including obtains the expression value of sample and its corresponding passage way network and each gene, and is weighted processing to the expression value of each gene as the weight of gene using the gene t t values examined and Pearson correlation coefficients;According to the topological structure of passage way network, interaction type and its corresponding intensity between gene are obtained in same access, and using the intensity of interaction type between gene and each gene expression values after weighting, obtain the interaction expression value between each gene;It integrates the expression value of each gene and interaction expression value and is analyzed using Principal Component Analysis, and obtained first principal component is further defined as to the active fraction of access.Implement the present invention, while considering that the importance of gene and the importance of gene interaction infer the activity of access, to realize the evaluation to biological pathway sample state.
Description
Technical field
The present invention relates to technical field of gene detection more particularly to a kind of active methods of derivation gene pathway.
Background technology
Many research methods propose that it is single to break through to have more the biological marker of robustness for searching on functional plane recently
The unstable problem of gene label.Due to gene be not it is individual participate in bioprocess, gene outcome usually with function module or
The modes such as signal cascade act synergistically, and the function module lacked of proper care in high level may be more more stable than single-gene as biomarker,
Influence very little of the various noises to it.The biological marker of functional plane can effectively reduce heterogeneity and the heredity of sample of tissue
Heterogeneity, while the relationship effectively between analysis critical function access and disease.Therefore, the expression of the relevant gene of integration function
Spectrum, and it is beneficial to obtain the biological marker of more robustness in functional plane extraction characteristic of division.Function module is often embedding
Enter in classical access and protein network, these high-throughput information can from Gene Ontology, KEGG databases or its
Gene sets defined in his microarray expression profile research experiment, as obtained in molecular label database MSigDB.
Since path information height embodies chemical effect and functional expression between gene, the gene expression water in access
It is flat inseparable with the function that access embodies, once the expression of the notable gene in access will be disorderly, passage portion work(
It can also lack of proper care.Therefore, Classification and Identification experiment is carried out by the activity of the gene expression profile passage in analysis path, with
Obtain accurate biomarker.Such as Su researchers devise one to solve the problems, such as Duplication in different accesses
Log-likelihood function finds the linear sub-channel with classification capacity, and obtained linear sub-channel has higher classification capacity,
Classifying quality is further promoted;For another example, the researchers such as Breslin by pathway members gene expression values and infer
Pathway activity;For another example, the average value (Mean) or median that the researchers such as Guo pass through calculating pathway members' gene expression values
(Median) infer pathway activity;For another example, the researchers such as Bild by principal component analysis pathway members gene expression profile with
First principal component infer pathway activity, this method can also identification disorder flow pattern and carcinogenic lead label, this also be phase
The targeted therapy for closing cancer subtypes puies forward important foundation much of that;For another example, the researchers such as Lee think the CORGs in access
(condition-responsive genes) gene pairs pathway activity plays a major role and is not gene all in access.
The above achievement in research shows to consider that the function module of gene can identify more stable biological marker, and has obtained more accurate
Classifying quality.
However, above-mentioned pathway activity estimating method is only with the notable gene in access, there is no consider gene
Between interaction information, access is only only regarded as to the simple set of individual gene, but ignore in passage way network
Gene topology information, this is just lost the information exchanged between many important genes.
Invention content
Technical problem to be solved of the embodiment of the present invention is, provides a kind of derivation gene pathway active method, together
When consider that the importance of gene and the importance of gene interaction infer the activity of access, to realize to biological pathway
The evaluation of sample state.
In order to solve the above-mentioned technical problem, an embodiment of the present invention provides a kind of active method of derivation gene pathway, packets
Include following steps:
Step S1, sample and its corresponding passage way network are obtained, and obtains contained each gene in the access network
Expression value, and the t values of the t inspections of the expression value with gene between two kinds of different manifestations types and gene expression values are showed with sample
The Pearson correlation coefficients of type are the weight of gene, and processing is weighted to the expression value of each gene in the access network;
Step S2, according to the topological structure of the passage way network, interaction type between gene is obtained in same access
And its corresponding intensity, and utilize the intensity and weighting treated each base corresponding to interaction type between obtained gene
The expression value of cause obtains the interaction expression value between each gene in the access network;
Step S3, it integrates the interaction expression value in the passage way network between the expression value and each gene of each gene and adopts
It is analyzed with Principal Component Analysis, and obtained first principal component is further defined as to the active fraction of access.
Wherein, in the step S1, the expression value of contained each gene in the access network is standardized,
Specifically formula isWherein, giJ represents expression values of the gene i in sample j, and mean and std are respectively represented
The average value and standard deviation of gene expression value in all samples.
Wherein, in the step S1, the expression value of power treated gene is z'ij=tscore(gi)2*ρ(gi)*zij;
Wherein, z 'ijFor gene g in sample jijExpression value after weighting;Gene tscore(gi) it is gene giDouble two tables of tail t check analyses
The statistical value of gene expression values between type;ρ(gi) it is skin of the gene between the expression value and sample phenotype of all samples
Ademilson related coefficient.
Wherein, in the step S2, the interaction expression value between each gene isWherein, ehj
For gene gijWith gene gkjThe expression value of interaction;βikFor gene giWith gene gkThe corresponding β value of interaction type;ρikFor gene giWith
Gene gkThe Pearson correlation coefficients of expression value;z′ijFor gene g in sample jijExpression value after weighting;z′kjFor in sample j
Gene gkjExpression value after weighting.
Wherein, in the step S3, the calculation formula of the active fraction of each gene pathway is:
a(Pj)=w1jz′1j+w2jz'2j+…+wijz′ij+…+wnjz'nj+w(n+1)je1j+…+w(n+h)ehj+…w(n+l)elj;
Wherein, a (Pj) be sample j pathway activity score, w1jIt is first gene in sample j in the weight of first principal component, wijFor
Gene i is in the weight of first principal component, w in sample j(n+1)jIt is first Interaction among genes in sample j in the power of first principal component
Weight, n are the total number of gene, and l is the number of Interaction among genes.
Implement the embodiment of the present invention, has the advantages that:
The present invention uses in passage way network of the Principal Component Analysis to incorporating each sample the expression value of each gene and each
Interaction expression value between gene is analyzed, and the first principal component that each sample obtains is defined as to the active fraction of access,
The importance of gene is considered not only, it is also considered that arrived the importance of gene interaction to infer the activity of access, had
There is very wide practicability.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, according to
These attached drawings obtain other attached drawings and still fall within scope of the invention.
Fig. 1 is the flow chart provided in an embodiment of the present invention for deriving the active method of gene pathway;
Fig. 2 is the application scenario diagram provided in an embodiment of the present invention for deriving the active method of gene pathway;
Specific implementation mode
To make the object, technical solutions and advantages of the present invention clearer, the present invention is made into one below in conjunction with attached drawing
Step ground detailed description.
As shown in Figure 1, in the embodiment of the present invention, a kind of active method of derivation gene pathway of proposition, including it is following
Step:
Step S1, sample and its corresponding passage way network are obtained, and obtains contained each gene in the access network
Expression value, and the t values of the t inspections of the expression value with gene between two kinds of different manifestations types and gene expression values are showed with sample
The Pearson correlation coefficients of type are the weight of gene, and processing is weighted to the expression value of each gene in the access network;
Step S2, according to the topological structure of the passage way network, interaction type between gene is obtained in same access
And its corresponding intensity, and utilize the intensity and weighting treated each base corresponding to interaction type between obtained gene
The expression value of cause obtains the interaction expression value between each gene in the access network;
Step S3, it integrates the interaction expression value in the passage way network between the expression value and each gene of each gene and adopts
It is analyzed with Principal Component Analysis, and obtained first principal component is further defined as to the active fraction of access.
Detailed process is in step sl, in order to make gene expression values in same grade, to avoid each gene expression values
Not under same dimension, unreasonable classification results are obtained.First to the expression value of contained each gene in the access network into
Row standardization, specific formula are:
In formula (1), gijExpression values of the gene i in sample j is represented, mean and std respectively represent the gene in all samples
The average value and standard deviation of expression value in this.If the expression value of some gene has missing in sample j, then other samples are used
Average value of gene expression values is as filling missing values in this.
Since differential expression of the gene in two kinds of phenotypes can be intuitively depicted by t test values come if the t inspections of gene
It is higher to test value, then more showing that gene differential expression under two kinds of phenotypes is bigger, therefore this characteristic pair of t test values can be utilized
Gene expression values are weighted, this is just exaggerated the gene in not isophenic gene expression values difference.
The expression value of each sample power treated gene is:
z'ij=tscore(gi)2*ρ(gi)*zij(2);
In formula (2), z 'ijFor gene g in sample jijExpression value after weighting;Gene tscore(gi) it is gene giDouble tail t inspections
Test the statistical value of the gene expression values between two phenotypes of analysis;ρ(gi) it is expression value and sample of the gene in all samples
Pearson correlation coefficients between phenotype.
It should be noted that t inspections can be divided into single gross examination and double gross examinations, single totality t is examined, is mainly examined
The difference of the average of one sample and the average of population sample.Look at whether this difference is notable.The system of single-sample t-test
Metering is:
WhereinIt is the average value of population sample, n is the population sample number of sample, σXIt is the standard deviation of sample.
It is the difference in respective aggregate level for weighing two samples that double totality t, which are examined,.Double totality t inspections can be segmented
For independent samples t test and paired-sample t test.The most commonly used is independent samples t tests in cancer classification experiment.Existed by gene
Two not isophenic t test values describe differential expression of the gene under the two different phenotypes.Its gene is in two different tables
The t test statistics of type is:
Wherein n1And n2It is positive sample and the overall number of negative sample respectively,WithIt is two samples, the base in this respectively
Because of the variance of expression value,WithThe average value of the gene expression values in respectively two samples.Two samples of null hypothesis are obeyed
The average value being just distributed very much and variance be identical.Only when the variance of two populations is equal, usually this method is called
Student t is examined.When this null hypothesis is invalid, this method is sometimes referred to as Welch's t-test.T inspections can also be used to
Same statistic existing difference when measuring twice is examined, judges whether the difference between them is zero, in this case
Inspection commonly known as " is matched " or " duplicate measurements " t is examined.
It should be noted that Pearson correlation coefficients be often used in portray gene expression values and sample phenotype correlation and
, there is interaction between two of which gene in the correlation between two genes, Pearson correlation coefficients can be retouched intuitively
State the interaction strength of the two genes.Then there is the calculation formula of the Pearson correlation coefficients of the gene i and gene k of interaction
For:
The value of Pearson correlation coefficients indicates two between 1 and -1 when the Pearson correlation coefficients of two genes are 1
The complete linear positive of gene is related, and correlation is stronger;The Pearson correlation coefficients of two genes indicate that the two variables do not have when being 0
Linear dependence, it is related weaker;The Pearson correlation coefficients of two variables indicate that the two genes are completely negative when being -1
Correlation can also illustrate that two gene associations are stronger.
Pearson correlation coefficients be it is symmetrical, i.e.,:Corr (X, Y)=cor (Y, X).One pass of Pearson correlation coefficients
The mathematical property of key is:It is constant in the case of the different variations of the position of two variables and scale.That is, I
X can be transformed into a+bX, Y is become c+dY, wherein a, b, c and d are constant b, d > 0, and this variation of variable does not change
Become the related coefficient between them.
In step s 2, such as fruit gene i and gene k, there are interaction relationships in access, can be according to two genes
Expression value define the expression values of the two intergenic interactions.Using the intensity and type to interact between them to the gene
Interaction is weighted.Therefore, the interaction expression value between gene i and gene k is expressed as:
In formula (3), ehjFor gene gijWith gene gkjThe expression value of interaction;βikFor gene giWith gene gkInteraction type pair
The β value answered;ρikFor gene giWith gene gkThe Pearson correlation coefficients of expression value;z′ijFor gene g in sample jijAfter weighting
Expression value;z′kjFor gene g in sample jkjExpression value after weighting.
And so on, it may be determined that go out the interaction expression value between each gene in the access network.
In the step S3, the calculation formula of the active fraction of each gene pathway is:
a(Pj)=w1jz′1j+w2jz'2j+…+wijz′ij+…+wnjz'nj+w(n+1)je1j+…+w(n+h)ehj+…w(n+l)elj(4);
In formula (4), a (Pj) be sample j pathway activity score, w1jIt is first gene in sample j in first principal component
Weight, wijIt is gene i in sample j in the weight of first principal component, w(n+1)jIt is first Interaction among genes in sample j
The weight of one principal component, n are the total number of gene, and l is the number of Interaction among genes.
It should be noted that principal component analysis (Principal component analysis, PCA) is in machine learning
Important Feature Dimension Reduction algorithm, basic principle is will be in the dimension of the feature vector of original data projection to covariance matrix
Come.
The algorithm of PCA is shown in substantially steps are as follows:
1:Standardization, i.e. mean normalization are done to all sample datas;
2:Calculate the covariance matrix C of sample data:
Wherein, m is sample number, and n is the data volume of each sample;
3:Singular value decomposition is carried out to the covariance matrix that previous step acquires:
[U, S, V]=svd (C) (4-2)
4:Then according to the corresponding feature vector setting projection properties matrix P of characteristic value;
5:It will come on original data projection to eigenmatrix:
Z=PTX (4-3)
PCA technologies are usually used in each research field, its different names of application field are also not quite similar, for example it is being tied
Structure dynamics is called noise and vibration frequency specturm analysis, empirical modal analysis.It, usually can be into machine learning treatment classification problem
The processing of row feature selecting, in classification experiments, in the case of number of samples is limited, ten hundreds of genes is divided as feature
Class is clearly worthless, this will greatly reduce the performance of grader.It is a kind of feasible to carry out dimension-reduction treatment to biological data
Method.Gene data after PCA technology dimensionality reductions remains the information of legacy data, the wherein variance of first principal component data most
Greatly, it is commonly used to choose as important characteristic of division.
As shown in Fig. 2, for the application scenario diagram provided in an embodiment of the present invention for deriving the active method of gene pathway.It is first
First, standardize to gene expression values.Secondly, establish the Interaction among genes based on gene expression Value Data and access;?
In passage way network, each node represents a gene, and each edge indicates the interaction relationship between two genes;Third,
Pathway activity score is calculated using principal component analysis for each sample.
Implement the embodiment of the present invention, has the advantages that:
The present invention uses in passage way network of the Principal Component Analysis to incorporating each sample the expression value of each gene and each
Interaction expression value between gene is analyzed, and the first principal component that each sample obtains is defined as to the active fraction of access,
The importance of gene is considered not only, it is also considered that arrived the importance of gene interaction to infer the activity of access, had
There is very wide practicability.
It is above disclosed to be only a preferred embodiment of the present invention, the power of the present invention cannot be limited with this certainly
Sharp range, therefore equivalent changes made in accordance with the claims of the present invention, are still within the scope of the present invention.
Claims (5)
1. a kind of active method of derivation gene pathway, which is characterized in that include the following steps:
Step S1, sample and its corresponding passage way network are obtained, and obtains the expression of contained each gene in the access network
Value, and the t values of the t inspections of the expression value with gene between two kinds of different manifestations types and gene expression values and sample phenotype
Pearson correlation coefficients are the weight of gene, and processing is weighted to the expression value of each gene in the access network;
Step S2, according to the topological structure of the passage way network, obtain in same access between gene interaction type and its
Corresponding intensity, and utilize the intensity and weighting treated each gene corresponding to interaction type between obtained gene
Expression value obtains the interaction expression value between each gene in the access network;
Step S3, the interaction expression value in the passage way network between the expression value and each gene of each gene is integrated and using master
Componential analysis is analyzed, and obtained first principal component is further defined as to the active fraction of access.
2. deriving the active method of gene pathway as described in claim 1, which is characterized in that in the step S1, to institute
The expression value for stating contained each gene in access network is standardized, and specific formula isWherein, gij
Expression values of the gene i in sample j is represented, mean and std respectively represent the average value of gene expression value in all samples
And standard deviation.
3. deriving the active method of gene pathway as claimed in claim 2, which is characterized in that in the step S1, at power
The expression value of gene after reason is z 'ij=tscore(gi)2*ρ(gi)*zij;Wherein, z 'ijFor gene g in sample jijAfter weighting
Expression value;Gene tscore(gi) it is gene giThe statistical value of gene expression values between double two phenotypes of tail t check analyses;ρ(gi)
For Pearson correlation coefficients of the gene between the expression value and sample phenotype of all samples.
4. the active method of derivation gene pathway stated such as claim 3, which is characterized in that in the step S2, each gene
Between interaction expression value beWherein, ehjFor gene gijWith gene gkjThe expression value of interaction;βikFor
Gene giWith gene gkThe corresponding β value of interaction type;ρikFor gene giWith gene gkThe Pearson correlation coefficients of expression value;z′ijFor
Gene g in sample jijExpression value after weighting;z′kjFor gene g in sample jkjExpression value after weighting.
5. deriving the active method of gene pathway as claimed in claim 4, which is characterized in that in the step S3, each base
Because the calculation formula of the active fraction of access is:
a(Pj)=w1jz′1j+w2jz′2j+…+wijz′ij+…+wnjz′nj+w(n+1)je1j+…+w(n+h)ehj+…w(n+l)elj;Wherein,
a(Pj) be sample j pathway activity score, w1jIt is first gene in sample j in the weight of first principal component, wijFor sample j
Middle gene i is in the weight of first principal component, w(n+1)jIt is first Interaction among genes in sample j in the weight of first principal component, n
For the total number of gene, l is the number of Interaction among genes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810422205.7A CN108763862B (en) | 2018-05-04 | 2018-05-04 | Method for deducing gene pathway activity |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810422205.7A CN108763862B (en) | 2018-05-04 | 2018-05-04 | Method for deducing gene pathway activity |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108763862A true CN108763862A (en) | 2018-11-06 |
CN108763862B CN108763862B (en) | 2021-06-29 |
Family
ID=64009157
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810422205.7A Expired - Fee Related CN108763862B (en) | 2018-05-04 | 2018-05-04 | Method for deducing gene pathway activity |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108763862B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109817337A (en) * | 2019-01-30 | 2019-05-28 | 中南大学 | A kind of appraisal procedure and similar disorder differentiating method of single disease sample Pathway Activation degree |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004022711A2 (en) * | 2002-09-05 | 2004-03-18 | Bioseek, Inc. | Biomap analysis |
WO2013044354A1 (en) * | 2011-09-26 | 2013-04-04 | Trakadis John | Method and system for genetic trait search based on the phenotype and the genome of a human subject |
CN103045605A (en) * | 2012-12-26 | 2013-04-17 | 首都医科大学宣武医院 | I type neurofibroma NF1 gene mutation nucleotide sequence related to cerebrovascular stenosis and application thereof |
CN103180462A (en) * | 2010-10-06 | 2013-06-26 | 拜奥默里克斯公司 | Method for determining biological pathway activity |
CN104094266A (en) * | 2011-11-07 | 2014-10-08 | 独创系统公司 | Methods and systems for identification of causal genomic variants |
US20150038369A1 (en) * | 2007-04-09 | 2015-02-05 | George Washington University | Method and Kit for Diagnosing Autism Using Gene Expression Profiling |
US20150080249A1 (en) * | 2012-04-05 | 2015-03-19 | H. Lee Moffitt Cancer Center And Research Institute, Inc. | O-glycan pathway ovarian cancer signature |
CN105893731A (en) * | 2015-01-19 | 2016-08-24 | 大道安康(北京)科技发展有限公司 | Method for building expression detecting system of genetic health network |
CN107133492A (en) * | 2017-05-02 | 2017-09-05 | 温州大学 | A kind of method that gene pathway is recognized based on PAGIS |
CN107203704A (en) * | 2017-05-02 | 2017-09-26 | 温州大学 | A kind of method that gene pathway is recognized based on GSA |
CN107220526A (en) * | 2017-05-02 | 2017-09-29 | 温州大学 | A kind of method that gene pathway is recognized based on PADOG |
-
2018
- 2018-05-04 CN CN201810422205.7A patent/CN108763862B/en not_active Expired - Fee Related
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004022711A2 (en) * | 2002-09-05 | 2004-03-18 | Bioseek, Inc. | Biomap analysis |
US20150038369A1 (en) * | 2007-04-09 | 2015-02-05 | George Washington University | Method and Kit for Diagnosing Autism Using Gene Expression Profiling |
CN103180462A (en) * | 2010-10-06 | 2013-06-26 | 拜奥默里克斯公司 | Method for determining biological pathway activity |
WO2013044354A1 (en) * | 2011-09-26 | 2013-04-04 | Trakadis John | Method and system for genetic trait search based on the phenotype and the genome of a human subject |
CN104094266A (en) * | 2011-11-07 | 2014-10-08 | 独创系统公司 | Methods and systems for identification of causal genomic variants |
US20150080249A1 (en) * | 2012-04-05 | 2015-03-19 | H. Lee Moffitt Cancer Center And Research Institute, Inc. | O-glycan pathway ovarian cancer signature |
CN103045605A (en) * | 2012-12-26 | 2013-04-17 | 首都医科大学宣武医院 | I type neurofibroma NF1 gene mutation nucleotide sequence related to cerebrovascular stenosis and application thereof |
CN105893731A (en) * | 2015-01-19 | 2016-08-24 | 大道安康(北京)科技发展有限公司 | Method for building expression detecting system of genetic health network |
CN107133492A (en) * | 2017-05-02 | 2017-09-05 | 温州大学 | A kind of method that gene pathway is recognized based on PAGIS |
CN107203704A (en) * | 2017-05-02 | 2017-09-26 | 温州大学 | A kind of method that gene pathway is recognized based on GSA |
CN107220526A (en) * | 2017-05-02 | 2017-09-29 | 温州大学 | A kind of method that gene pathway is recognized based on PADOG |
Non-Patent Citations (4)
Title |
---|
WUFENG FAN 等: "Pathway Interaction Network Analysis Identifies Dysregulated Pathways in Human Monocytes Infected by Listeria monocytogenes", 《COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE》 * |
XIANBIN LI 等: "Subpathway Analysis based on Signaling-Pathway Impact Analysis of Signaling Pathway", 《PLOS ONE》 * |
向波 等: "注意力缺陷多动障碍和低出生体重共有的生物学通路研究", 《中华医学遗传学杂志》 * |
方宏源 等: "一种基于基因拓扑重要性的通路识别方法", 《生物信息学》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109817337A (en) * | 2019-01-30 | 2019-05-28 | 中南大学 | A kind of appraisal procedure and similar disorder differentiating method of single disease sample Pathway Activation degree |
Also Published As
Publication number | Publication date |
---|---|
CN108763862B (en) | 2021-06-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Aghaeepour et al. | Rapid cell population identification in flow cytometry data | |
CN112435714B (en) | Tumor immune subtype classification method and system | |
Hardin et al. | Outlier detection in the multiple cluster setting using the minimum covariance determinant estimator | |
Strauss | Discriminating groups of organisms | |
CN111899882B (en) | Method and system for predicting cancer | |
Balakrishnan et al. | Multivariate measures of skewness for the skew-normal distribution | |
WO2021139116A9 (en) | Method, apparatus and device for intelligently grouping similar patients, and storage medium | |
US20140040265A1 (en) | Method and apparatus for representing multidimensional data | |
CN108763864A (en) | A method of evaluation biological pathway sample state | |
Momal et al. | Tree-based inference of species interaction network from abundance data | |
Zeng et al. | Feature-guided clustering of multi-dimensional flow cytometry datasets | |
Leiva et al. | Linear discrimination for multi-level multivariate data with separable means and jointly equicorrelated covariance structure | |
WO2024131524A1 (en) | Depression diet management method based on food image segmentation | |
CN114596253A (en) | Alzheimer's disease identification method based on brain imaging genome features | |
CN108763862A (en) | A kind of active method of derivation gene pathway | |
CN113270191A (en) | Data correction and classification method and storage medium | |
Villarroel et al. | Cluster analysis using multivariate mixed effects models | |
Nouretdinov et al. | Multiprobabilistic venn predictors with logistic regression | |
Charytanowicz et al. | Discrimination of wheat grain varieties using x-ray images | |
CN110991517A (en) | Classification method and system for unbalanced data set in stroke | |
Leek | Surrogate variable analysis | |
Tai et al. | Detection of cell separation-induced gene expression through a penalized deconvolution approach | |
CN113296947A (en) | Resource demand prediction method based on improved XGboost model | |
del Barrio et al. | optimalflow: Optimal-transport approach to flow cytometry gating and population matching | |
Bhargavi et al. | A COMPARATIVE STUDY FOR STATISTICAL OUTLIER DETECTION USING COLON CANCER DATA |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210629 |
|
CF01 | Termination of patent right due to non-payment of annual fee |