CN108875305A - A kind of leukaemia cancer cell detector of colony intelligence optimizing - Google Patents
A kind of leukaemia cancer cell detector of colony intelligence optimizing Download PDFInfo
- Publication number
- CN108875305A CN108875305A CN201810458511.6A CN201810458511A CN108875305A CN 108875305 A CN108875305 A CN 108875305A CN 201810458511 A CN201810458511 A CN 201810458511A CN 108875305 A CN108875305 A CN 108875305A
- Authority
- CN
- China
- Prior art keywords
- gene
- value
- module
- sample
- algorithm
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The invention discloses the leukaemia cancer cell detector of colony intelligence optimizing, which reads in module, data prediction and feature ordering module, parameter optimization module, model output module by gene microarray and forms.System first pre-processes the gene microarray data of input, then carries out importance ranking to remaining each gene, calculates correlation by statistics score, recycles classifier criterion function to calculate contribution degree, all gene importance are ranked up.Improved optimization method joined under original intelligent optimizing algorithm fitness detection with population disturb, can prevent population diversity scatter and disappear and optimization process fall into local optimum.Then the optimized parameter searched is completed into model construction as classifier parameters and exports result.System speed is fast, is suitble to on-line checking.
Description
Technical field
The present invention relates to gene microarray data applied technical fields, and in particular, to a kind of white blood of colony intelligence optimizing
Sick cancer cell detector.
Background technique
Biochip technology is by microphotography, according to the intermolecular principle specifically to interact, by life section
Discontinuous analytic process is integrated in the miniature biochemical analysis system on silicon chip or glass-chip surface in field, with reality
Now to accurate, quick, large information capacity the detection of cell, protein, gene and other biological components.According to cured on chip
Biochip can be divided into genetic chip, protein-chip, polysaccharide chip and neuron chip by the difference of biomaterial.
Currently, most successful biochip form is also to be claimed with gene order for analysis object " microarray (microarray) "
For genetic chip (Gene chip) or DNA chip (DNA chip).Leukaemia can be divided into acute and chronic white blood by the emergency of onset
Disease.Acute leukemia cells differentiation is stuck in early stage, and based on original and prorubricyte, disease is quickly grown, course of disease number
Month.Chronic leukemia cell differentiation is preferable, based on inmature or mature cell, develops slowly, the course of disease several years.By sick cell system
Column classification, the T and B cell system of grain, single, red, macronucleus system and Lymphatic System including medullary system.Leukaemia is often clinically divided into lymph
Chronic myeloid leukemia, myelocytic leukemia, cell mixing leukaemia etc..Wherein acute leukemia is great to anthropogenic influence, Huan Zhetong
It often will be dead in 3 months after the onset.And it is very difficult for the detection of acute leukemia in history and be difficult to accurate.Now
Using DNA microarray technology, scientist is expected to be broken through in the field.And this kind of data have the characteristics of higher-dimension small sample,
For general classifier and parameter testing, there is the typical case of dimension disaster difficult.How to overcome it, is instantly one big
Research hotspot.
Summary of the invention
In order to overcome the shortcomings of to be difficult to search gene microarray data optimal feature subset and classification optimal parameter at present,
The purpose of the present invention is to provide a kind of leukaemia cancer cell detectors of colony intelligence optimizing.
The technical solution adopted by the present invention to solve the technical problems is：A kind of leukaemia cancer cell inspection of colony intelligence optimizing
Instrument is surveyed, the system is defeated by gene microarray reading module, data prediction and feature ordering module, parameter optimization module, model
Module forms out；Wherein：
That gene microarray reads in module reading is the class label Y=[y of all gene microarrays1,y2,...,ym],
Middle yiThe gene microarray expression value of=k, k ∈ (- 1,1) and all samples：
Wherein every a line xiRepresent the expression value of all genes of sample, corresponding each column xjA gene is represented to exist
Expression value in all samples, subscript i indicate that i-th of sample, in total m, subscript j indicate j-th of gene, in total n.
Data prediction and feature ordering module be the original microarray data of reading is normalized and feature row
The treatment process of sequence.Wherein normalization operation is：
Wherein, Min, Max are respectively the minimum value of sample gene expression values, maximum value.And feature ordering selection uses often
The contribution degree of a gene pairs classification accuracy is given a mark to realize, in support vector machines theory, by defining a contribution degree letter
Number：
Wherein, α=[α1,...,αn],Hij=yiyjK(xi,xj), α is the corresponding coefficient of normal vector, H is intermediary matrix, J
For cost function, I be unit matrix, K is kernel function, y is label value, x is sample characteristics, the transposition of subscript T representing matrix,
Subscript i, j respectively indicates i-th of sample and j-th of gene.In fact, the formula represents the square value of classification boundaries size, support
The objective function of vector machine is to minimize it.In the case where using kernel function of the linear kernel as support vector machines, then have：
Wherein,w*It is normal vector, α for optimal supporting vector, w*Optimal supporting vector is corresponding
Coefficient.Above formula is observed, it is available：The significance level of each feature is according to this feature for the contribution of this cost function
It determines, i.e., the contribution margin of each feature is：Wherein, δ indicates contribution degree.
It, generally can following approximate calculation when using Non-linear Kernel as kernel function：
Wherein, α value is constant after some feature of reasonable assumption eliminates, and H (- i) indicates the H-matrix after this feature cancellation
Value.And when using the hypothesis, the result of obtained result and linear kernel is not much different.It can cycle calculations feature using the formula
Contribution degree carries out gene importance ranking.
For one group of training sample set { xn,tn(n=1,2 ... N, x ∈ Rd, t ∈ { 0,1 }), wherein x indicates training sample
This, t indicates sample moment point.Classification function y (the x of Method Using Relevance Vector Machine；Wr it) is defined as：
Wherein K (x, xi) it is kernel function, wriFor weight, wr0For initial weight.By logistic sigmoid contiguous function
It is applied to former formula, the likelihood probability p (t | wr) that can obtain the data set is estimated as：
Wherein, σ is variance, and in order to avoid over-fitting, Method Using Relevance Vector Machine is that each wr is provided with Gaussian prior probability distribution
Constraint condition：
The approximation of posterior probability can be obtained according to Laplace theory：α fixed first, the correspondence maximum possible sought is seemingly
The weight w r of right probability MPMP, since p (wr | t, α) is proportional to p (t | wr) p (wr | α), wr is sought using second order Newton methodMP：
Wherein, yn=σ { y (xn；) }, wr A=diag (α0,α1,...,αN).Then Laplace method is utilized, to above formula
Carrying out secondary derivation can obtain：
Wherein, Φ is N × (N+1) structure matrix, Φ=[φ (x1),φ(x2),...,φ(xN),]T, φ (xn)=
[1,K(xn,x1),K(xn,x2),...,K(xn,xN)]T, B=diag (β1,β2,...,βN), B is a diagonal matrix, βn=σ { y
(Xn)}[1-σ{y(Xn)}].Negative sign is taken to the right formula of above formula, then inverts, covariance matrix Σ can be obtained.Followed by, using Σ and
wrMP, hyper parameter α is updated：
wrMP=Σ ΦTBt
In an iterative process, most αiMeeting is close to infinity, therefore their corresponding wriIt is just substantially equal to 0,
Its basic function can be left out and reach sparsity.
Parameter optimization module design increases the diversity of population using a kind of improved colony intelligence optimizing algorithm, specifically sets
It counts as follows：
1) species information of DE algorithm is initialized：
In population, it is randomly generated：
X in above formulai(0) represent the expression value of i-th of individual chromosome gene primary, xj,i(0) i-th primary is represented
In body the expression value of j-th of chromosomal gene, rand (0,1) be uniform random number, NP in (0,1) section be Population Size,
Subscript L, U respectively indicate floor value, upper dividing value.
2) mutation operation (Mutation)：DE algorithm distinguishes over the characteristics of genetic algorithm (GA) and is that it uses plan of checking the mark
It slightly makes a variation, by the difference of two individuals of random selection, carries out vector sum with target individual after scaling, that is,
vi(g+1)=xr1(g)+F·(xr2(g)-xr3(g)),i≠r1≠r2≠r3
In above formula, g represents g generation, and F is the zoom factor of two random vector differences, viIt (g+1) is variation intermediate variable,
xr1(g)、xr2(g)、xr3(g) expression value of g the 1st, 2,3 r-th of chromosomal gene of individual of generation is respectively indicated.
3) crossover operation (Crossover)：By g for population xi(g) the intermediate variable v generated with step 2)i(g+1) into
Row intersects, and generates
In above formula, CR is the crossing-over rate of setting, uj,iIt (g+1) is to intersect intermediate variable.
4) selection operation (Selection)：Differential evolution algorithm carries out retaining the next generation using common greedy algorithm, such as
Fruit intersects the population's fitness f (u generatedi(g+1) it is greater than the population's fitness f (x of previous generationi(g)), then retain, otherwise population
It is constant, i.e.,
In order to avoid there is precocious situation, an adaptive operator λ is designed：
G in above formulamaxMaximum number of iterations is represented, G represents current iteration number, F0For mutation operator, initial stage the value compared with
Greatly, guarantee sample diversity, the later period gradually becomes smaller, it is intended that protects the excellent information of evolutionary process.In differential evolution algorithm, such as
Fruit fitness can not be more than that history is optimal after certain the number of iterations always, then it is assumed that fall into local optimum, utilize gunz at this time
Energy algorithm jumps out differential evolution algorithm：
5) current location point information initializing is given to ant colony intelligence algorithm, wherein ant individual amount is：M, pheromones are dense
Degree is：τij=c (c>0).
6) all ants 1,2... are simulated, m is mobile to terminal, and each ant moves from current location i to next position j
Dynamic probabilityFor：
7) when an iteration is completed, i.e., when all ants cover path, current information element concentration is updated：
ρ is pheromone concentration volatility coefficient in above formula,Represent the pheromone concentration that ant k leaves on the ij of path, root
It is inversely proportional to the relationship of path length according to pheromone concentration, can be defined as follows：
In above formula, C is proportionality constant, and L is path length.
8) it after obtaining new candidate solution, is most preferably compared with history and more new historical is best.
9) iteration operation above procedure is until reach maximum algebra.Then most using history optimal parameter as parameter optimization
Terminate fruit input model output module.
Model output module, the model obtained using above procedure, directly inputs patient data, can be obtained according to label value
Result out.
Beneficial effects of the present invention are mainly manifested in：The present invention sets monitored parameters increase in intelligent searching process
Population diversity, to increase the probability for searching optimized parameter, speed is fast, is suitble to on-line checking.
Detailed description of the invention
Fig. 1 is structural schematic diagram of the invention；
Fig. 2 is flow chart of the invention.
Specific embodiment
The present invention is illustrated below according to attached drawing.
Referring to Fig.1, a kind of leukaemia cancer cell detector of colony intelligence optimizing, the system read in module by gene microarray
1, data prediction and feature ordering module 2, parameter optimization module 3, model output module 4 form；Wherein：
That gene microarray reads in the reading of module 1 is the class label Y=[y of all gene microarrays1,y2,...,ym],
Wherein yiThe gene microarray expression value of=k, k ∈ (- 1,1) and all samples：
Wherein every a line xiRepresent the expression value of all genes of sample, corresponding each column xjA gene is represented to exist
Expression value in all samples, subscript i indicate that i-th of sample, in total m, subscript j indicate j-th of gene, in total n.
Data prediction and feature ordering module 2 be the original microarray data of reading is normalized and feature row
The treatment process of sequence.Wherein normalization operation is：
Wherein, Min, Max are respectively the minimum value of sample gene expression values, maximum value.And feature ordering selection uses often
The contribution degree of a gene pairs classification accuracy is given a mark to realize, in support vector machines theory, by defining a contribution degree letter
Number：
Wherein, α=[α1,...,αn],Hij=yiyjK(xi,xj), α is the corresponding coefficient of normal vector, H is intermediary matrix, J
For cost function, I be unit matrix, K is kernel function, y is label value, x is sample characteristics, the transposition of subscript T representing matrix,
Subscript i, j respectively indicates i-th of sample and j-th of gene.In fact, the formula represents the square value of classification boundaries size, support
The objective function of vector machine is to minimize it.In the case where using kernel function of the linear kernel as support vector machines, then have：
Wherein,w*It is normal vector, α for optimal supporting vector, w*Optimal supporting vector is corresponding
Coefficient.Above formula is observed, it is available：The significance level of each feature is according to this feature for the contribution of this cost function
It determines, i.e., the contribution margin of each feature is：Wherein, δ indicates contribution degree.
It, generally can following approximate calculation when using Non-linear Kernel as kernel function：
Wherein, α value is constant after some feature of reasonable assumption eliminates, and H (- i) indicates the H-matrix after this feature cancellation
Value.And when using the hypothesis, the result of obtained result and linear kernel is not much different.It can cycle calculations feature using the formula
Contribution degree carries out gene importance ranking.
For one group of training sample set { xn,tn(n=1,2 ... N, x ∈ Rd, t ∈ { 0,1 }), wherein x indicates training sample
This, t indicates sample moment point.Classification function y (the x of Method Using Relevance Vector Machine；Wr it) is defined as：
Wherein K (x, xi) it is kernel function, wriFor weight, wr0For initial weight.By logistic sigmoid contiguous function
It is applied to former formula, the likelihood probability p (t | wr) that can obtain the data set is estimated as：
Wherein, σ is variance, and in order to avoid over-fitting, Method Using Relevance Vector Machine is that each wr is provided with Gaussian prior probability distribution
Constraint condition：
The approximation of posterior probability can be obtained according to Laplace theory：α fixed first, the correspondence maximum possible sought is seemingly
The weight w r of right probability MPMP, since p (wr | t, α) is proportional to p (t | wr) p (wr | α), wr is sought using second order Newton methodMP：
Wherein, yn=σ { y (xn；) }, wr A=diag (α0,α1,...,αN).Then Laplace method is utilized, to above formula
Carrying out secondary derivation can obtain：
Wherein, Φ is N × (N+1) structure matrix, Φ=[φ (x1),φ(x2),...,φ(xN),]T, φ (xn)=
[1,K(xn,x1),K(xn,x2),...,K(xn,xN)]T, B=diag (β1,β2,...,βN), B is a diagonal matrix, βn=σ { y
(Xn)}[1-σ{y(Xn)}].Negative sign is taken to the right formula of above formula, then inverts, covariance matrix Σ can be obtained.Followed by, using Σ and
wrMP, hyper parameter α is updated：
wrMP=Σ ΦTBt
In an iterative process, most αiMeeting is close to infinity, therefore their corresponding wriIt is just substantially equal to 0,
Its basic function can be left out and reach sparsity.
Parameter optimization module 3 designs the diversity for increasing population using a kind of improved colony intelligence optimizing algorithm, specifically
It designs as follows：
1) species information of DE algorithm is initialized：
In population, it is randomly generated：
X in above formulai(0) represent the expression value of i-th of individual chromosome gene primary, xj,i(0) i-th primary is represented
In body the expression value of j-th of chromosomal gene, rand (0,1) be uniform random number, NP in (0,1) section be Population Size,
Subscript L, U respectively indicate floor value, upper dividing value.
2) mutation operation (Mutation)：DE algorithm distinguishes over the characteristics of genetic algorithm (GA) and is that it uses plan of checking the mark
It slightly makes a variation, by the difference of two individuals of random selection, carries out vector sum with target individual after scaling, that is,
vi(g+1)=xr1(g)+F·(xr2(g)-xr3(g)),i≠r1≠r2≠r3
In above formula, g represents g generation, and F is the zoom factor of two random vector differences, viIt (g+1) is variation intermediate variable,
xr1(g)、xr2(g)、xr3(g) expression value of g the 1st, 2,3 r-th of chromosomal gene of individual of generation is respectively indicated.
3) crossover operation (Crossover)：By g for population xi(g) the intermediate variable v generated with step 2)i(g+1) into
Row intersects, and generates
In above formula, CR is the crossing-over rate of setting, uj,iIt (g+1) is to intersect intermediate variable.
4) selection operation (Selection)：Differential evolution algorithm carries out retaining the next generation using common greedy algorithm, such as
Fruit intersects the population's fitness f (u generatedi(g+1) it is greater than the population's fitness f (x of previous generationi(g)), then retain, otherwise population
It is constant, i.e.,
In order to avoid there is precocious situation, an adaptive operator λ is designed：
G in above formulamaxMaximum number of iterations is represented, G represents current iteration number, F0For mutation operator, initial stage the value compared with
Greatly, guarantee sample diversity, the later period gradually becomes smaller, it is intended that protects the excellent information of evolutionary process.In differential evolution algorithm, such as
Fruit fitness can not be more than that history is optimal after certain the number of iterations always, then it is assumed that fall into local optimum, utilize gunz at this time
Energy algorithm jumps out differential evolution algorithm：
5) current location point information initializing is given to ant colony intelligence algorithm, wherein ant individual amount is：M, pheromones are dense
Degree is：τij=c (c>0).
6) all ants 1,2... are simulated, m is mobile to terminal, and each ant moves from current location i to next position j
Dynamic probabilityFor：
7) when an iteration is completed, i.e., when all ants cover path, current information element concentration is updated：
ρ is pheromone concentration volatility coefficient in above formula,Represent the pheromone concentration that ant k leaves on the ij of path, root
It is inversely proportional to the relationship of path length according to pheromone concentration, can be defined as follows：
In above formula, C is proportionality constant, and L is path length.
8) it after obtaining new candidate solution, is most preferably compared with history and more new historical is best.
9) iteration operation above procedure is until reach maximum algebra.Then most using history optimal parameter as parameter optimization
The fruit that terminates is defeated
Enter model output module.
The parameter that parameter optimization module 3 exports enters model output module 4, and the parameter as diagnostic equipment uses.Model is defeated
Module analyzes the practical leukaemia sufferer gene microarray data of subsequent input out.
Above-described embodiment is used to illustrate the present invention, rather than limits the invention, in spirit of the invention and
In scope of protection of the claims, to any modifications and changes that the present invention makes, protection scope of the present invention is both fallen within.
Claims (5)
1. a kind of leukaemia cancer cell detector of colony intelligence optimizing, it is characterised in that：The system reads in mould by gene microarray
Block, data prediction and feature ordering module, parameter optimization module, model output module composition.
2. the leukaemia cancer cell detector of colony intelligence optimizing according to claim 1, it is characterised in that：The micro- battle array of gene
Column read in the class label Y=[y that module reads in all gene microarrays1,y2,...,ym], wherein yi=k, k ∈ (- 1,1), with
And the gene microarray expression value of all samples：
Wherein every a line xiRepresent the expression value of all genes of sample, corresponding each column xjA gene is represented all
Expression value in sample, subscript i indicate that i-th of sample, in total m, subscript j indicate j-th of gene, in total n.
3. the leukaemia cancer cell detector of colony intelligence optimizing according to claim 1, it is characterised in that：The data are located in advance
It manages and the original microarray data of reading is normalized feature ordering module and the treatment process of feature ordering.Wherein return
One, which changes operation, is：
Wherein, Min, Max are respectively the minimum value of sample gene expression values, maximum value.And feature ordering selection uses each base
Because being realized to the contribution degree marking of classification accuracy, in support vector machines theory, by defining a contribution degree function：
Wherein, α=[α1,...,αn],Hij=yiyjK(xi,xj), α is the corresponding coefficient of normal vector, H is intermediary matrix, and J is generation
Valence function, I are unit matrix, K is kernel function, y is label value, x is sample characteristics, the transposition of subscript T representing matrix, subscript
I, j respectively indicates i-th of sample and j-th of gene.In fact, the formula represents the square value of classification boundaries size, supporting vector
The objective function of machine is to minimize it.In the case where using kernel function of the linear kernel as support vector machines, then have：
Wherein,w*It is normal vector, α for optimal supporting vector, w*The corresponding system of optimal supporting vector
Number.Above formula is observed, it is available：The significance level of each feature according to this feature for this cost function contribution come
It determines, i.e., the contribution margin of each feature is：Wherein, δ indicates contribution degree.
It, generally can following approximate calculation when using Non-linear Kernel as kernel function：
Wherein, α value is constant after some feature of reasonable assumption eliminates, and H (- i) indicates the H-matrix value after this feature cancellation.And
When using the hypothesis, the result of obtained result and linear kernel is not much different.It can cycle calculations signature contributions using the formula
Degree carries out gene importance ranking.
For one group of training sample set { xn,tn(n=1,2 ... N, x ∈ Rd, t ∈ { 0,1 }), wherein x indicates training sample, t
Indicate sample moment point.Classification function y (the x of Method Using Relevance Vector Machine；Wr it) is defined as：
Wherein K (x, xi) it is kernel function, wriFor weight, wr0For initial weight.By logistic sigmoid contiguous function application
To former formula, the likelihood probability p (t | wr) that can obtain the data set is estimated as：
Wherein, σ is variance, and in order to avoid over-fitting, Method Using Relevance Vector Machine is that each wr is constrained provided with Gaussian prior probability distribution
Condition：
The approximation of posterior probability can be obtained according to Laplace theory：α fixed first, the correspondence maximum possible likelihood sought are general
The weight w r of rate MPMP, since p (wr | t, α) is proportional to p (t | wr) p (wr | α), wr is sought using second order Newton methodMP：
Wherein, yn=σ { y (xn；) }, wr A=diag (α0,α1,...,αN).Then Laplace method is utilized, above formula is carried out
Secondary derivation can obtain：
Wherein, Φ is N × (N+1) structure matrix, Φ=[φ (x1),φ(x2),...,φ(xN),]T, φ (xn)=[1, K
(xn,x1),K(xn,x2),...,K(xn,xN)]T, B=diag (β1,β2,...,βN), B is a diagonal matrix, βn=σ { y
(Xn)}[1-σ{y(Xn)}].Negative sign is taken to the right formula of above formula, then inverts, covariance matrix Σ can be obtained.Followed by, using Σ and
wrMP, hyper parameter α is updated：
wrMP=Σ ΦTBt
In an iterative process, most αiMeeting is close to infinity, therefore their corresponding wriIt is just substantially equal to 0, it can be with
Leave out its basic function and reaches sparsity.
4. the leukaemia cancer cell detector of colony intelligence optimizing according to claim 1, it is characterised in that：The parameter optimization
Module increases the diversity of population using improved colony intelligence optimizing algorithm, specific as follows：
1) species information of DE algorithm is initialized：
In population, it is randomly generated：
X in above formulai(0) represent the expression value of i-th of individual chromosome gene primary, xj,i(0) it represents in i-th of individual primary
The expression value of j-th of chromosomal gene, rand (0,1) are that uniform random number, the NP in (0,1) section are Population Size, subscript
L, U respectively indicates floor value, upper dividing value.
2) mutation operation (Mutation)：DE algorithm distinguish over the characteristics of genetic algorithm (GA) be that its use check the mark strategy into
Row variation carries out vector sum with target individual after scaling by the difference of two individuals of random selection, that is,
vi(g+1)=xr1(g)+F·(xr2(g)-xr3(g)),i≠r1≠r2≠r3
In above formula, g represents g generation, and F is the zoom factor of two random vector differences, viIt (g+1) is variation intermediate variable, xr1
(g)、xr2(g)、xr3(g) expression value of g the 1st, 2,3 r-th of chromosomal gene of individual of generation is respectively indicated.
3) crossover operation (Crossover)：By g for population xi(g) the intermediate variable v generated with step 2)i(g+1) it is handed over
Fork generates
In above formula, CR is the crossing-over rate of setting, uj,iIt (g+1) is to intersect intermediate variable.
4) selection operation (Selection)：Differential evolution algorithm carries out retaining the next generation using common greedy algorithm, if handed over
Pitch the population's fitness f (u generatedi(g+1) it is greater than the population's fitness f (x of previous generationi(g)), then retain, otherwise population is constant,
I.e.
In order to avoid there is precocious situation, an adaptive operator λ is designed：
G in above formulamaxMaximum number of iterations is represented, G represents current iteration number, F0For mutation operator, initial stage, the value was larger, guaranteed
Sample diversity, later period gradually become smaller, it is intended that protect the excellent information of evolutionary process.In differential evolution algorithm, if adapted to
Degree can not be more than that history is optimal after certain the number of iterations always, then it is assumed that fall into local optimum, utilize swarm intelligence algorithm at this time
Jump out differential evolution algorithm：
5) current location point information initializing is given to ant colony intelligence algorithm, wherein ant individual amount is：M, pheromone concentration are：
τij=c (c>0).
6) all ants 1,2... are simulated, m is moved to terminal, what each ant was moved from current location i to next position j
ProbabilityFor：
7) when an iteration is completed, i.e., when all ants cover path, current information element concentration is updated：
ρ is pheromone concentration volatility coefficient in above formula,The pheromone concentration that ant k leaves on the ij of path is represented, according to letter
The relationship that plain concentration is inversely proportional to path length is ceased, can be defined as follows：
In above formula, C is proportionality constant, and L is path length.
8) it after obtaining new candidate solution, is most preferably compared with history and more new historical is best.
9) iteration operation above procedure is until reach maximum algebra.Then using history optimal parameter as the most termination of parameter optimization
Fruit input model output module.
5. the leukaemia cancer cell detector of colony intelligence optimizing according to claim 1, it is characterised in that：The model output
The model that module is obtained using parameter optimization module, directly inputs patient data, can be obtained a result according to label value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810458511.6A CN108875305A (en) | 2018-05-14 | 2018-05-14 | A kind of leukaemia cancer cell detector of colony intelligence optimizing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810458511.6A CN108875305A (en) | 2018-05-14 | 2018-05-14 | A kind of leukaemia cancer cell detector of colony intelligence optimizing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108875305A true CN108875305A (en) | 2018-11-23 |
Family
ID=64333928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810458511.6A Pending CN108875305A (en) | 2018-05-14 | 2018-05-14 | A kind of leukaemia cancer cell detector of colony intelligence optimizing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108875305A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145171A (en) * | 2007-09-15 | 2008-03-19 | 中国科学院合肥物质科学研究院 | Gene microarray data predication method based on independent component integrated study |
| CN106971091A (en) * | 2017-03-03 | 2017-07-21 | 江苏大学 | A kind of tumour recognition methods based on certainty particle group optimizing and SVMs |
| CN107103332A (en) * | 2017-04-07 | 2017-08-29 | 武汉理工大学 | A kind of Method Using Relevance Vector Machine sorting technique towards large-scale dataset |
| CN107292128A (en) * | 2017-06-27 | 2017-10-24 | 湖南农业大学 | One kind pairing interacting genes detection method and forecast model |
-
2018
- 2018-05-14 CN CN201810458511.6A patent/CN108875305A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101145171A (en) * | 2007-09-15 | 2008-03-19 | 中国科学院合肥物质科学研究院 | Gene microarray data predication method based on independent component integrated study |
| CN106971091A (en) * | 2017-03-03 | 2017-07-21 | 江苏大学 | A kind of tumour recognition methods based on certainty particle group optimizing and SVMs |
| CN107103332A (en) * | 2017-04-07 | 2017-08-29 | 武汉理工大学 | A kind of Method Using Relevance Vector Machine sorting technique towards large-scale dataset |
| CN107292128A (en) * | 2017-06-27 | 2017-10-24 | 湖南农业大学 | One kind pairing interacting genes detection method and forecast model |
Non-Patent Citations (4)
| Title |
|---|
| XINTENG GAO 等: "A novel effective diagnosis model based on optimized least squares support machine for gene microarray", 《APPLIED SOFT COMPUTING》 * |
| 刘志斌 等: "《最优化方法及应用案例》", 30 November 2013, 石油工业出版社 * |
| 曹涛: "基于相关向量机的污水软测量建模研究", 《中国优秀硕士学位论文全文数据库》 * |
| 熊伟丽 等: "基于改进差分进化算法的非线性系统模型参数辨识", 《计算机应用研究》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Örkcü et al. | Estimating the parameters of 3-p Weibull distribution using particle swarm optimization: A comprehensive experimental comparison | |
| Hassanien et al. | Computational intelligence techniques in bioinformatics | |
| CN116982113A (en) | Machine learning driven plant gene discovery and gene editing | |
| Maulik | Analysis of gene microarray data in a soft computing framework | |
| Alok et al. | Semi-supervised clustering for gene-expression data in multiobjective optimization framework | |
| CN104966106B (en) | A kind of biological age substep Forecasting Methodology based on support vector machines | |
| Arowolo et al. | A survey of dimension reduction and classification methods for RNA-Seq data on malaria vector | |
| Fonseca et al. | Phylogeographic model selection using convolutional neural networks | |
| CN105808976A (en) | Recommendation model based miRNA target gene prediction method | |
| CN108595909A (en) | TA targeting proteins prediction techniques based on integrated classifier | |
| Babu et al. | A comparative study of gene selection methods for cancer classification using microarray data | |
| US20220245188A1 (en) | A system and method for processing biology-related data, a system and method for controlling a microscope and a microscope | |
| CN112837743B (en) | Drug repositioning method based on machine learning | |
| Saha et al. | Aggregation of multi-objective fuzzy symmetry-based clustering techniques for improving gene and cancer classification | |
| Salem et al. | Mgs-cm: a multiple scoring gene selection technique for cancer classification using microarrays | |
| CN108664763A (en) | A kind of lung cancer carcinoma cell detection instrument that parameter is optimal | |
| Ramakrishna et al. | Evolutionary Optimization Algorithm for Classification of Microarray Datasets with Mayfly and Whale Survival. | |
| CN108875305A (en) | A kind of leukaemia cancer cell detector of colony intelligence optimizing | |
| JP2004355174A (en) | Data analysis method and system | |
| CN110739028B (en) | Cell line drug response prediction method based on K-nearest neighbor constraint matrix decomposition | |
| CN109308934A (en) | A kind of gene regulatory network construction method based on integration characteristic importance and chicken group's algorithm | |
| CN108897988A (en) | A kind of colon cancer cancer cell detector of colony intelligence optimizing | |
| Shukla et al. | Supervised learning of Plasmodium falciparum life cycle stages using single-cell transcriptomes identifies crucial proteins | |
| CN108695002A (en) | A kind of intelligent Colon cancer cancer cell detector | |
| Muhammad et al. | Gvdeepnet: Unsupervised deep learning techniques for effective genetic variant classification |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181123 |