CN109558493A - A kind of disease similarity calculating method based on disease ontology - Google Patents
A kind of disease similarity calculating method based on disease ontology Download PDFInfo
- Publication number
- CN109558493A CN109558493A CN201811255993.1A CN201811255993A CN109558493A CN 109558493 A CN109558493 A CN 109558493A CN 201811255993 A CN201811255993 A CN 201811255993A CN 109558493 A CN109558493 A CN 109558493A
- Authority
- CN
- China
- Prior art keywords
- disease
- node
- ontology
- similarity
- similitude
- 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
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention belongs to field of bioinformatics, specially a kind of disease similarity calculating method based on disease ontology.The method of the present invention is divided into two parts, and first part is the function of diseases Similarity measures based on gene ontology, and second part is that the disease Semantic Similarity based on disease ontology calculates.In terms of algorithm evaluation, two kinds of appraisal procedures of rate are shared using ROC curve and PTC, are as a result superior to existing disease to similarity assessment algorithm.Complicated metabolism and vital movement are in close relations often and in vivo for the pathogenesis of disease, this brings huge challenge to the mankind in the understanding of disease incidence mechanism and the research of diagnosis and treatment means, and the method for the present invention facilitates the research of disease incidence mechanism, diagnosis and treatment means and disease prevention etc..
Description
Technical field
The invention belongs to field of bioinformatics, and for assessing the similarity of disease pair, correlation result analysis facilitates
A kind of research of disease incidence mechanism, diagnosis and treatment means and disease prevention etc., and in particular to disease phase based on disease ontology
Like degree calculation method.
Background technique
In recent years, more and more researchers participate in Study on Similarity, including protein in field of biomedicine
Functional similarity, gene similitude, the similarity directions such as drug similitude and disease similitude.Wherein, disease similitude
Research is intended to disclose its possible pathology principle, and provide medical treatment appropriate by the internal association for analyzing two kinds of diseases and build
View.Relevant information based on limited disease such as gene information etc. can carry out similitude point to known disease and unidentified illness
Analysis, when their similitudes are very high, indicates that two kinds of diseases at least possess the same or similar attribute in one aspect, so also
May there are similar disease Forming Mechanism and diagnosis and treatment method, provide preferable guide for the cause of disease of study of disease.
Fig. 1 is listed based on the existing main disease based on disease ontology to similarity calculating method, can be subdivided into
The method that method based on disease Semantic Similarity, the method based on function of diseases similitude and the two combine.Wherein,
Method based on disease Semantic Similarity can be further subdivided into the method based on disease ontology nodal information amount and be based on disease
The method of sick body construction information.
Nineteen ninety-five Resnik proposes a kind of algorithm based on maximum fault information common ancestor, and the algorithm is according to disease ontology
In " is_a " relationship, the similitude of disease pair is indicated with maximum fault information common ancestor node.Couto et al. was in 2004
The concept of incompatible ancestor node is proposed, and calculates the similitude of two disease nodes with this.The side of Resnik and Couto
Method is all the similarity for calculating disease from the ancestor node of two disease nodes in disease ontology.And Lin in 1998 and
In the method that Schlicker was delivered in 2006, it is believed that the similarity of disease is only calculated from one or several ancestor nodes,
And it is incomplete for having ignored disease node information content in disease ontology itself.Lin is mentioned on the basis of the method for Resnik
A kind of disease similarity calculation method combining ancestor node and own node out, then work of the Schlicker in Lin
On the basis of introduce correction factor and improved.
In the calculation method based on disease body construction information, Rada et al. is relatively early in 1989 to be proposed based on node
The disease similarity calculation method of distance, but the value of distance may be 0 or infinity in the calculation method of Rada, for this
Problem, Lee et al. proposed innovatory algorithm in 1993, and the arithmetic result of Rada has been carried out normalized, has made similarity
Value be [0,1].In the method that Wang et al. was delivered in 2007, it is believed that when calculating disease similitude, not only to utilize
The information content of disease node, but also the path structure information of disease ontology should be utilized.In this regard, Wang sharing two diseases
Information is extended to all ancestor node by common ancestor's node, effectively overcomes information caused by local nodes routing information not
Full problem, improves accuracy rate.In addition, thinking in the method that Wu was delivered at 2006 and Jiang in 1997, disease ontology
Hierarchical structure the similitude between disease can also be described well.Wu proposes a kind of calculation based on disease ontology routing information
Method, the algorithm not only allow for two routing informations between disease and ancestor node, it is also contemplated that disease and leaf node it
Between routing information, obtained preferable calculated result.The algorithm of Jiang consider localized network complexity in ontology,
Node level, while type and while intensity, and opposite side assign weight, by shortest path between disease node across side
The sum of weight as their Similarity measures result.
In the calculation method of function of diseases similitude, Mathur et al. has delivered BOG calculation 2010 and 2012 respectively
Method and PSB algorithm, he thinks the inhereditary feature of gene control biology, its expression and disease generation are closely related, therefore disease
Between there is only the incidence relations on disease ontology, there is also incidence relations on gene ontology, and the association of the latter is more
It is important.BOG algorithm mainly calculates the similitude of disease using the shared gene of two disease nodes, so BOG algorithm can be sent out
There are close associations between the existing not similar disease of disease symptoms.PSB algorithm is on the basis of BOG algorithm, it is contemplated that disease
The relevant intergenic functional cohesion of disease, is first annotated with all gene-for-gene ontologies with disease association, based on this
Calculate the information content of each gene ontology node, then calculate gene ontology node similarity matrix, finally by disease with
Incidence relation between gene ontology obtains the similitude of disease.The algorithm is more quasi- than the algorithm based on sick language disease justice similitude
Really.
Li et al. thinks that corresponding gene may also can share phase when two gene ontology nodes are much like but not identical
With bioprocess, and in the algorithm of Mathur two genes must share identical gene ontology node just and can obtain it is high similar
Spend score.In view of this, Li et al. proposed a kind of algorithm based on gene interaction network, i.e. SemFunSim in 2013
Algorithm.The algorithm combines disease Semantic Similarity and function of diseases similitude, takes in terms of Semantic Similarity calculating
The disease semantic similarity that the method for Resnik is asked；In terms of functional similarity calculating, gene interaction network is introduced
HumanNet, in the network, the gene of each Thermodynamic parameters has a similarity scores (Log Likelihood
Score, LLS) for measuring the functional cohesion between gene.Final similarity score is function similarity multiplied by semantic similar
Degree.
Summary of the invention
The invention belongs to field of bioinformatics, method of specially a kind of assessment disease to similarity.The method of the present invention
(referred to as DGS algorithm) is divided into two parts, and first part is the function of diseases Similarity measures based on gene ontology, and second part is
Disease Semantic Similarity based on disease ontology calculates.In terms of algorithm evaluation, two kinds of rate is shared using ROC curve and PTC and is commented
Estimate method, makes comparisons with the method for Resnik, the method for Wang, PSB algorithm and SemFunSim algorithm, be as a result superior to respectively
The existing disease of these types is to similarity assessment algorithm.The complicated metabolism and vital movement often and in vivo of the pathogenesis of disease
In close relations, this brings huge challenge, side of the present invention in the understanding of disease incidence mechanism and the research of diagnosis and treatment means to the mankind
Method facilitates the research of disease incidence mechanism, diagnosis and treatment means and disease prevention etc..
A kind of disease similarity calculating method based on disease ontology proposed by the present invention, the method includes assessing disease
Pair function of diseases similarity method and assessment disease pair disease Method of Semantic Similarity Analysis two parts, it is specific as follows:
(1) the function of diseases similarity method for assessing disease pair is as follows:
(1.1) gene ontology node similitude is constructed using shortest path first (the Shortest Path, SP) first
Matrix calculates the information content of gene ontology node, each element of gene ontology node similarity matrix using Resnik method
It is the numerical value between one 0~1, indicates the similitude of any two node in gene ontology, this similarity matrix note
For SimGO；
Wherein path1For t1Shortest path of the node to the public ancestor node of maximum fault information (MICA node), path2For
t2Shortest path of the node to the public ancestor node of maximum fault information (MICA node)；
(1.2) disease ontology relation medical subject headings databases, if two disease arts in medical subject headings database
Language is respectively m1And m2, correspondingly, obtaining the relevant gene molecule functional term collection of disease term in medical subject headings database
Closing is respectively T1={ t11, t12..., t1pAnd T2={ t21, t22..., t2q}；Set T1And T2Regard as to medical subject headings
Gene ontology node similarity matrix is converted into medical subject using formula (1) by the description of disease term characteristics in database
Word node similitude:
Wherein p, q are respectively and disease term m1、m2Associated molecular function term number, max function are to take maximum
Value, SimGO(t1, t2) it is two gene ontology node t1And t2Similarity score；
(1.3) the disease node in disease ontology usually all can be with the disease term in several medical subject headings databases
It is corresponding, it is assumed that disease is to for d1And d2, two disease terms in disease ontology have been corresponded, have been obtained from disease ontology
Disease is to d1And d2Associated medical subject headings database disease term is respectively M1={ m11, m12..., m1pAnd M2={ m21,
m22..., m2q}；
Wherein p, q are respectively and disease d1、d2Associated medical subject headings database disease term number, max function are
It is maximized, SimMeSH(m1, m2) it is two gene ontology node m1And m2Similarity score；
(2) the disease Method of Semantic Similarity Analysis for assessing disease pair is as follows:
The calculating of function of diseases similitude considers the mutual incidence relation of gene ontology molecular level disease, and disease
Advantage of the disease ontology in structure organization is then utilized in the calculating of sick Semantic Similarity；When calculating disease Semantic Similarity,
Need to find out the public ancestor node of maximum fault information (the most informative common of two disease nodes
Ancestor, MICA), remember that two diseases are d1、d2, remember that this ancestor node is dMICA, enable G1、G2And GMICARespectively indicate disease
d1、d2And dMICAssociated gene sets；Similitude is calculated using following equation:
Wherein | G1|、|G2| and | GMICA| respectively indicate disease d1、d2And dMICAAssociated gene ontology interstitial content；
(3) similarity score of disease pair is assessed:
According to the score of the function of diseases similitude and disease Semantic Similarity that are calculated in step (1) and step (2), value
All between 0~1, the two is multiplied to obtain the similarity score of disease pair:
SimDGS(d1, d2)=FSimDGS(d1, d2)·SSimDGS(d1, d2)
Wherein FSimDGS(d1, d2) it is disease to d1And d2Functional similarity score, SSimDGS(d1, d2) it is disease to d1
And d2Semantic Similarity score.
The beneficial effects of the present invention are: similarity algorithm is obtained more accurately as a result, for disease than original disease
The research of pathogenesis, diagnosis and treatment means and disease prevention etc. provides new reference tool.
Detailed description of the invention
Fig. 1 is existing main disease to similarity algorithm synoptic chart.
Fig. 2 is algorithm flow of the invention.
Specific embodiment
The present invention is further illustrated below by embodiment combination attached drawing.
Embodiment 1: the specific implementation process of the method for the present invention is as follows:
Step 1: calculating the information content of gene ontology node using the method for Resnik
It is obtained from GOA database (http://geneontology.org/page/download-annotations)
The related information of gene and gene ontology node, the process of each gene association to gene ontology node are known as annotating, annotate
Cheng Hou, each gene ontology node have watched several genes attentively, this number is considered as frequency, is acquired according to the formula of information content
The information content of each gene ontology node.
Step 2: constructing gene ontology node similarity matrix using SP algorithm
The nodal information amount acquired using previous step is constructed using shortest path first (the Shortest Path, SP)
Gene ontology node similarity matrix.Each element of this matrix is the numerical value between one 0~1, is indicated in gene ontology
In two nodes similitude, this similarity matrix is denoted as SimGO, calculation formula are as follows:
Wherein path1For t1Shortest path of the node to the public ancestor node of maximum fault information (MICA node), path2For
t2Shortest path of the node to the public ancestor node of maximum fault information (MICA node).
Step 3: calculating disease to function similarity
Due to the various medical data bases of disease ontology relation, cause the selection of its data varied, traditional Chinese medicine master
Epigraph (Medical Subject Headings, MeSH) database has more than 100 years usage histories, can preferably describe
Disease, authority with higher, and medical subject headings have a wide range of applications in each large database concept, so we are counting
Medical subject headings are introduced into calculating process when calculating the similitude of disease.Assuming that two disease terms point in medical subject headings
It Wei not m1And m2, it is respectively T that the relevant gene molecule functional term set of disease term is obtained in medical subject headings database1
={ t11, t12..., t1pAnd T2={ t21, t22..., t2q}.Set T1And T2It can be regarded as to disease in medical subject headings
Gene ontology node similitude is converted into medical subject headings node similitude using following equation by the description of term characteristics:
Disease ontology is to arrange a kind of body construction summed up after multiple medical data bases, so the disease in disease ontology
Sick node usually all can be corresponding with several medical subject vocabulary, and the disease function of disease pair is next calculated with similar method
It can similitude.Assuming that disease is to for d1And d2, they have corresponded two disease terms in disease ontology, from disease ontology
In obtain d1And d2Associated medical subject headings term is respectively M1={ m11, m23..., m1pAnd M2={ m21, m22...,
m2q}。
Step 4: calculating disease to semantic similarity
The calculating of function of diseases similitude considers the mutual incidence relation of gene ontology molecular level disease, and disease
Advantage of the disease ontology in structure organization is then utilized in the calculating of sick Semantic Similarity.When calculating disease Semantic Similarity,
Need to find out the public ancestor node of maximum fault information (the most informative common of two disease nodes
Ancestor, MICA), remember that two diseases are d1、d2, remember that this ancestor node is dMICA, enable G1、G2And GMICARespectively indicate disease
d1、d2And dMICAssociated gene sets.Similitude is calculated using following equation:
Step 5: calculating disease to similarity final score
Above-mentioned two step has calculated the score of function of diseases similitude and disease Semantic Similarity, value all 0~1 it
Between, the two is multiplied to obtain the similarity score of disease pair:
SimDGS(d1, d2)=FSimDGS(d1, d2)·SSimDGS(d1, d2)
The method of the present invention performance evaluation:
According to disease to the appraisal procedure of the correlative theses of similarity calculation, two kinds of rate is shared using ROC curve and PTC and is commented
Estimate method, makees ratio with the method for the relatively good Resnik of effect, the method for Wang, PSB algorithm and SemFunSim algorithm respectively
Compared with.
The assessment of 1 disease similitude ROC curve
Test set is combined by two collection, and one is standard set, and one is random set.Standard set is from Li et al. 2013
Year paper in the data set announced, the disease pair of totally 70 pairs of higher similitudes；Random set by 100 pairs from disease ontology it is random
The disease selected is to composition.It is verified for the generality to algorithm, we generate 10 random sets at random, they are distinguished
It is combined with standard set, becomes test set, then this ten groups of test sets are calculated using five kinds of algorithms, obtain calculating knot
Fruit, and assessment is compared to calculated result.
Disease in test set to calculating respectively using five kinds of algorithms, is obtained the meter of five groups of disease similitudes by us
It calculates as a result, be then ranked up the correlation result of disease pair from high to low, by the disease of standard set to being set as positive sample, with
The disease of machine collection draws ROC curve, further calculates out AUC value, obtained assessment result to negative sample is set as.
In experiment herein, for the validity for further increasing ROC curve assessment algorithm effect, we are by random disease pair
The condition of generation generates three group data sets, the first group data set S plus constraint1It is to generate disease at random in disease ontology
It is right；Second group data set S2It is to constrain the range of choice of disease, it only could when the associated gene number of disease is greater than 5
It is selected into random disease pair；Third group data set S3It is that random disease is added in disease when the gene number of disease association is greater than 10
Pair selection.
Following table is comparing result, and the method for the present invention effect is best as the result is shown, is secondly successively SemFunSim algorithm, PSB
Algorithm, the method for Wang, the method for Resnik.
2 disease similitude PTC share rate assessment
The relevant gene sheet of disease is provided not only in CTD database (http://ctdbase.org/downloads/)
Body information additionally provides the potential treatment compound (Potential Therapeutic Compounds, PTC) of disease, PTC
The ingredient that disease therapeuticing medicine can effectively be measured, has great importance to medicament research and development.
For the validity for further studying algorithm, we obtain disease and the incidence relation of PTC from CTD database,
Carry out effect of the verification algorithm in terms of disease medicament with this.
We generate ten groups of diseases to test set at random, wherein including 1000 pairs of diseases pair in each test set, respectively
Five kinds of Similarity Algorithms are taken to calculate the disease of each test set to similarity.Then, to the Similarity measures knot of test set
Fruit is ranked up, and takes before similarity 100 disease to composition set T100, pass through analysis set T later100The PTC of middle disease pair
Shared rate carrys out measure algorithm effect.
Following table is comparing result:
| Method | Comprehensive PTC shares rate | Standard PTC shares rate | The disease logarithm of shared PTC |
| The method of Resnik | 6.26% | 7.63% | 82.1 |
| The method of Wang | 7.34% | 8.37% | 87.8 |
| PSB algorithm | 11.13% | 12.10% | 92.2 |
| SemFunSim algorithm | 13.03% | 13.54% | 96.3 |
| DGS algorithm | 15.25% | 15.28% | 99.8 |
There are two the index that PTC shares rate in table, Plays PTC shares rate and refers in set T100In all diseases pair
PTC shares the average of rate, i.e., all diseases share the sum of rate divided by the disease logarithm for sharing PTC in set to PTC；It is comprehensive
PTC shares rate and refers to that all diseases share the sum of rate divided by set T to PTC100Disease logarithm (100).Known by defining us
Road, comprehensive PTC, which shares rate, can more reflect overall effect of the algorithm on test set.But when the disease pair for including in test set
When quantity further expansion, it is possible that set T100In disease to all there is the case where shared PTC, standard PTC shares rate
Rough assessment can be carried out to data in this case.
From table it can be seen that, it is shared that rate, the disease logarithm of shared PTC or standard PTC either are shared in comprehensive PTC
In terms of rate, the performance of the method for the present invention is all best.
In two kinds of appraisal procedures, the method for the present invention all achieves best effect.The experimental results showed that the method for the present invention
The mutual association of our study of disease is not only assisted in, provides preferable guide for our study of disease, but also provide
A kind of effective way of study of disease drug.
Claims (1)
1. a kind of disease similarity calculating method based on disease ontology, it is characterised in that: the method includes assessing disease pair
Function of diseases similarity method and assessment disease pair disease Method of Semantic Similarity Analysis two parts, it is specific as follows:
(1) the function of diseases similarity method for assessing disease pair is as follows:
(1.1) gene ontology node similitude square is constructed using shortest path first (the Shortest Path, SP) first
Battle array, the information content of gene ontology node is calculated using Resnik method, and each element of gene ontology node similarity matrix is
Numerical value between one 0~1, indicates the similitude of any two node in gene ontology, this similarity matrix is denoted as
SimGO；
Wherein path1For t1Shortest path of the node to the public ancestor node of maximum fault information (MICA node), path2For t2Node
To the shortest path of the public ancestor node of maximum fault information (MICA node)；
(1.2) disease ontology relation medical subject headings databases, if two disease terms point in medical subject headings database
It Wei not m1And m2, correspondingly, obtaining the relevant gene molecule functional term set point of disease term in medical subject headings database
It Wei not T1={ t11, t12..., t1pAnd T2={ t21, t22..., t2q}；Set T1And T2Regard as to medical subject headings data
Gene ontology node similarity matrix is converted into medical subject headings section using formula (1) by the description of disease term characteristics in library
Point similitude:
Wherein p, q are respectively and disease term m1、m2Associated molecular function term number, max function are to be maximized,
SimGO(t1, t2) it is two gene ontology node t1And t2Similarity score；
(1.3) the disease node in disease ontology usually all can be opposite with the disease term in several medical subject headings databases
It answers, it is assumed that disease is to for d1And d2, two disease terms in disease ontology have been corresponded, have obtained disease from disease ontology
To d1And d2Associated medical subject headings database disease term is respectively M1={ m11, m12..., m1pAnd M2={ m21,
m22..., m2q}；
Wherein p, q are respectively and disease d1、d2Associated medical subject headings database disease term number, max function is to take most
Big value, SimMeSH(m1, m2) it is two gene ontology node m1And m2Similarity score；
(2) the disease Method of Semantic Similarity Analysis for assessing disease pair is as follows:
The calculating of function of diseases similitude considers the mutual incidence relation of gene ontology molecular level disease, and disease language
Advantage of the disease ontology in structure organization is then utilized in the calculating of adopted similitude；When calculating disease Semantic Similarity, need
Find out two disease nodes the public ancestor node of maximum fault information (most informative common ancestor,
MICA), remember that two diseases are d1、d2, remember that this ancestor node is dMICA, enable G1、G2And GMICARespectively indicate disease d1、d2With
dMICANode set in associated gene ontology；Similitude is calculated using following equation:
Wherein | G1|、|G2| and | GMICA| respectively indicate disease d1、d2And dMICAAssociated gene ontology interstitial content；
(3) similarity score of disease pair is assessed:
According to the score of the function of diseases similitude and disease Semantic Similarity that calculate in step (1) and step (2), value all exists
Between 0~1, the two is multiplied to obtain the similarity score of disease pair:
SimDGS(d1, d2)=FSimDGS(d1, d2)·SSimDGS(d1, d2)
Wherein FSimDGS(d1, d2) it is disease to d1And d2Functional similarity score, SSimDGS(d1, d2) it is disease to d1And d2
Semantic Similarity score.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811255993.1A CN109558493B (en) | 2018-10-26 | 2018-10-26 | Disease similarity calculation method based on disease ontology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811255993.1A CN109558493B (en) | 2018-10-26 | 2018-10-26 | Disease similarity calculation method based on disease ontology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109558493A true CN109558493A (en) | 2019-04-02 |
| CN109558493B CN109558493B (en) | 2023-02-10 |
Family
ID=65865401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811255993.1A Active CN109558493B (en) | 2018-10-26 | 2018-10-26 | Disease similarity calculation method based on disease ontology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109558493B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401980A (en) * | 2020-02-19 | 2020-07-10 | 北京值得买科技股份有限公司 | Method and device for improving sample sequencing diversity |
| CN112151184A (en) * | 2020-09-27 | 2020-12-29 | 东北林业大学 | System for calculating disease similarity based on network representation learning |
| CN113160996A (en) * | 2021-01-19 | 2021-07-23 | 北京安智因生物技术有限公司 | Entity-based cardiovascular disease data integration method |
| CN115312182A (en) * | 2022-07-26 | 2022-11-08 | 哈尔滨工业大学 | Model for predicting risk of adverse reaction converted into serious adverse reaction after vaccination |
| CN116343913A (en) * | 2023-03-15 | 2023-06-27 | 昆明市延安医院 | Analysis method for predicting potential pathogenic mechanism of single-gene genetic disease based on phenotype semantic association gene cluster regulation network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120185172A1 (en) * | 2011-01-18 | 2012-07-19 | Barash Joseph | Method, system and apparatus for data processing |
| CN106874706A (en) * | 2017-01-18 | 2017-06-20 | 湖南大学 | Disease association factor identification method and system based on functional module |
| US20170242959A1 (en) * | 2016-02-24 | 2017-08-24 | Ucb Biopharma Sprl | Method and system for quantifying the likelihood that a gene is casually linked to a disease |
| CN107506608A (en) * | 2017-09-29 | 2017-12-22 | 杭州电子科技大学 | A kind of improved miRNA disease association Forecasting Methodologies based on collaborative filtering |
| CN107862179A (en) * | 2017-11-06 | 2018-03-30 | 中南大学 | A kind of miRNA disease association Relationship Prediction methods decomposed based on similitude and logic matrix |
-
2018
- 2018-10-26 CN CN201811255993.1A patent/CN109558493B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120185172A1 (en) * | 2011-01-18 | 2012-07-19 | Barash Joseph | Method, system and apparatus for data processing |
| US20170242959A1 (en) * | 2016-02-24 | 2017-08-24 | Ucb Biopharma Sprl | Method and system for quantifying the likelihood that a gene is casually linked to a disease |
| CN106874706A (en) * | 2017-01-18 | 2017-06-20 | 湖南大学 | Disease association factor identification method and system based on functional module |
| CN107506608A (en) * | 2017-09-29 | 2017-12-22 | 杭州电子科技大学 | A kind of improved miRNA disease association Forecasting Methodologies based on collaborative filtering |
| CN107862179A (en) * | 2017-11-06 | 2018-03-30 | 中南大学 | A kind of miRNA disease association Relationship Prediction methods decomposed based on similitude and logic matrix |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401980A (en) * | 2020-02-19 | 2020-07-10 | 北京值得买科技股份有限公司 | Method and device for improving sample sequencing diversity |
| CN112151184A (en) * | 2020-09-27 | 2020-12-29 | 东北林业大学 | System for calculating disease similarity based on network representation learning |
| CN113160996A (en) * | 2021-01-19 | 2021-07-23 | 北京安智因生物技术有限公司 | Entity-based cardiovascular disease data integration method |
| CN115312182A (en) * | 2022-07-26 | 2022-11-08 | 哈尔滨工业大学 | Model for predicting risk of adverse reaction converted into serious adverse reaction after vaccination |
| CN116343913A (en) * | 2023-03-15 | 2023-06-27 | 昆明市延安医院 | Analysis method for predicting potential pathogenic mechanism of single-gene genetic disease based on phenotype semantic association gene cluster regulation network |
| CN116343913B (en) * | 2023-03-15 | 2023-11-14 | 昆明市延安医院 | Analysis method for predicting potential pathogenic mechanism of single-gene genetic disease based on phenotype semantic association gene cluster regulation network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109558493B (en) | 2023-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109558493A (en) | A kind of disease similarity calculating method based on disease ontology | |
| Li et al. | Bridgenet: A continuity-aware probabilistic network for age estimation | |
| Muhammad et al. | COVID-19 and non-COVID-19 classification using multi-layers fusion from lung ultrasound images | |
| CN104915561B (en) | Genius morbi intelligent Matching method | |
| Khalil et al. | A new expert system in prediction of lung cancer disease based on fuzzy soft sets | |
| Campbell et al. | Unemployment rates: an alternative to the Jarman index? | |
| Rao et al. | A new intelligence-based approach for computer-aided diagnosis of dengue fever | |
| Luo et al. | Improving biomedical information retrieval with neural retrievers | |
| CN105706097A (en) | Pathway analysis for identification of diagnostic tests | |
| Dimitri et al. | Unsupervised stratification in neuroimaging through deep latent embeddings | |
| Torres-Espín et al. | Topological network analysis of patient similarity for precision management of acute blood pressure in spinal cord injury | |
| Fathima et al. | Comparison of classification techniques-SVM and naives bayes to predict the Arboviral disease-Dengue | |
| Li et al. | Subgroup identification via homogeneity pursuit for dense longitudinal/spatial data | |
| Zambrana et al. | Network neighbors of viral targets and differentially expressed genes in COVID-19 are drug target candidates | |
| CN115423052B (en) | Traditional Chinese medicine syndrome type classification method based on multi-graph attention | |
| Kaur et al. | A novel framework for drug synergy prediction using differential evolution based multinomial random forest | |
| Li et al. | Tree-structured subgroup analysis of receiver operating characteristic curves for diagnostic tests | |
| CN115691751A (en) | Traditional Chinese medicine prescription screening method and system based on diagnosis and treatment experience and intelligent learning | |
| Lee et al. | Comparative evaluation of bibliometric content networks by tomographic content analysis: An application to Parkinson's disease | |
| Lassila | Use of electronic health records and machine learning for sleep apnea diagnosis and treatment cost prediction | |
| Simarmata et al. | Grouping diabetes diagnosis based on age range with k-means algorithm | |
| Zhao et al. | Enhancing Biomedical ReQA With Adversarial Hard In-Batch Negative Samples | |
| Moor | Machine Learning on Clinical Time Series: Classification and Representation Learning | |
| Shen | Additive Mixed Modeling of HIV Patient Outcomes Across Multiple Studies | |
| Shukla et al. | Random forest v/s scaled conjugate gradient to predict diabetes mellitus |
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 |