CN117116339A - Method and device for identifying image group biological characteristics based on WGCNA - Google Patents
Method and device for identifying image group biological characteristics based on WGCNA Download PDFInfo
- Publication number
- CN117116339A CN117116339A CN202311377289.4A CN202311377289A CN117116339A CN 117116339 A CN117116339 A CN 117116339A CN 202311377289 A CN202311377289 A CN 202311377289A CN 117116339 A CN117116339 A CN 117116339A
- Authority
- CN
- China
- Prior art keywords
- image
- gene
- module
- histology
- genes
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 98
- 206010028980 Neoplasm Diseases 0.000 claims abstract description 41
- 238000004458 analytical method Methods 0.000 claims abstract description 29
- 230000014509 gene expression Effects 0.000 claims abstract description 24
- 238000004393 prognosis Methods 0.000 claims abstract description 22
- 230000004186 co-expression Effects 0.000 claims abstract description 20
- 230000004083 survival effect Effects 0.000 claims abstract description 11
- 230000003993 interaction Effects 0.000 claims abstract description 9
- 238000012216 screening Methods 0.000 claims abstract description 8
- 230000008827 biological function Effects 0.000 claims abstract description 6
- 101150044508 key gene Proteins 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 8
- JXSJBGJIGXNWCI-UHFFFAOYSA-N diethyl 2-[(dimethoxyphosphorothioyl)thio]succinate Chemical compound CCOC(=O)CC(SP(=S)(OC)OC)C(=O)OCC JXSJBGJIGXNWCI-UHFFFAOYSA-N 0.000 claims description 7
- 230000031018 biological processes and functions Effects 0.000 claims description 4
- 239000010410 layer Substances 0.000 claims description 4
- 230000004879 molecular function Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000011160 research Methods 0.000 claims description 4
- 206010030113 Oedema Diseases 0.000 claims description 3
- 238000010201 enrichment analysis Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000003068 pathway analysis Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000019491 signal transduction Effects 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 210000003850 cellular structure Anatomy 0.000 claims description 2
- 239000011229 interlayer Substances 0.000 claims description 2
- 201000010099 disease Diseases 0.000 abstract description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 abstract description 5
- 230000000694 effects Effects 0.000 description 6
- 210000002370 ICC Anatomy 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 238000010988 intraclass correlation coefficient Methods 0.000 description 4
- 208000005017 glioblastoma Diseases 0.000 description 3
- 208000024827 Alzheimer disease Diseases 0.000 description 2
- 108090000790 Enzymes Proteins 0.000 description 2
- 102000004190 Enzymes Human genes 0.000 description 2
- 102000049982 HMGA2 Human genes 0.000 description 2
- 108700039143 HMGA2 Proteins 0.000 description 2
- 101150073387 Hmga2 gene Proteins 0.000 description 2
- 208000023105 Huntington disease Diseases 0.000 description 2
- 108091026822 U6 spliceosomal RNA Proteins 0.000 description 2
- 201000011510 cancer Diseases 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 2
- 210000003169 central nervous system Anatomy 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003012 network analysis Methods 0.000 description 2
- 230000010627 oxidative phosphorylation Effects 0.000 description 2
- 230000027756 respiratory electron transport chain Effects 0.000 description 2
- 101150029062 15 gene Proteins 0.000 description 1
- -1 CYC1 Proteins 0.000 description 1
- 102100039259 Cytochrome c oxidase subunit 8A, mitochondrial Human genes 0.000 description 1
- 102100026761 Eukaryotic translation initiation factor 5A-1 Human genes 0.000 description 1
- 101000745956 Homo sapiens Cytochrome c oxidase subunit 8A, mitochondrial Proteins 0.000 description 1
- 101001054354 Homo sapiens Eukaryotic translation initiation factor 5A-1 Proteins 0.000 description 1
- 208000012902 Nervous system disease Diseases 0.000 description 1
- 102100021969 Nucleotide pyrophosphatase Human genes 0.000 description 1
- 208000018737 Parkinson disease Diseases 0.000 description 1
- 208000027089 Parkinsonian disease Diseases 0.000 description 1
- 206010034010 Parkinsonism Diseases 0.000 description 1
- 206010039966 Senile dementia Diseases 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000010195 expression analysis Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000001325 log-rank test Methods 0.000 description 1
- 238000002595 magnetic resonance imaging Methods 0.000 description 1
- 238000010197 meta-analysis Methods 0.000 description 1
- 239000002777 nucleoside Substances 0.000 description 1
- 150000003833 nucleoside derivatives Chemical class 0.000 description 1
- 239000002773 nucleotide Substances 0.000 description 1
- 108010067588 nucleotide pyrophosphatase Proteins 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000010837 poor prognosis Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000010882 preoperative diagnosis Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000001959 radiotherapy Methods 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 239000001226 triphosphate Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B25/00—ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
- G16B25/10—Gene or protein expression profiling; Expression-ratio estimation or normalisation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B50/00—ICT programming tools or database systems specially adapted for bioinformatics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biotechnology (AREA)
- Genetics & Genomics (AREA)
- Bioethics (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Public Health (AREA)
- Chemical & Material Sciences (AREA)
- Data Mining & Analysis (AREA)
- Analytical Chemistry (AREA)
- Epidemiology (AREA)
- Evolutionary Computation (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Software Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The method and the device for identifying the biological characteristics of the image group based on the WGCNA can more accurately capture the interaction relation between the biological characteristics, provide more interpretable results, can be applied to various diseases, and have high practicability and application value. The method comprises the following steps: (1) Acquiring gene expression profile data, image data and clinical data; (2) Dividing a tumor area from the image data, and extracting image histology characteristics; (3) Performing single factor Cox analysis on the magnetic resonance image characteristics, and selecting factors with P value smaller than 0.05 as image histology prognosis labels; (4) Constructing a weighted co-expression network, and identifying a gene module most relevant to image risk classification by calculating gene saliency and module members; (5) Screening key genes, selecting the gene analysis with the highest score, discussing the expression level of the genes in tumors, and carrying out survival analysis; (6) The biological functions of the image histology module genes are analyzed and studied.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a method for identifying biological characteristics of an image group based on WGCNA and a device for identifying the biological characteristics of the image group based on WGCNA.
Background
The 2016 World Health Organization (WHO) Central Nervous System (CNS) tumor classification is added with the molecular characteristics of genes for the first time, thereby greatly improving the treatment and prognosis of patients. Magnetic resonance is a non-invasive imaging technique that is critical for preoperative diagnosis and prognosis evaluation of tumors. However, whether the imaging characteristics of the tumor are related to gene expression is Shang Weike. The student's biological characteristics of image group are difficult to recognize.
Disclosure of Invention
In order to overcome the defects of the prior art, the technical problem to be solved by the invention is to provide a method for identifying image group biological characteristics based on WGCNA, which can more accurately capture the interaction relation between the biological characteristics and provide more interpretable results, can be applied to various diseases, and has high practicability and application value.
The technical scheme of the invention is as follows: the method for identifying the biological characteristics of the image group based on the WGCNA comprises the following steps:
(1) Acquiring gene expression profile data, image data and clinical data of a tumor patient;
(2) Dividing tumor areas of each patient from the image data, and extracting image histology characteristics;
(3) Carrying out single factor Cox (proportional risk) analysis on the magnetic resonance image characteristics, selecting factors with P value smaller than 0.05, and taking the factors as an image histology prognosis label;
(4) Constructing a weighted co-expression network by using a WGCNA program package in the R language, and identifying a gene module most relevant to image risk classification by calculating gene saliency and module members;
(5) Screening key genes by using Cytoscape (interaction analysis software), selecting the genes with highest scores, analyzing the genes in an Oncomine (tumor gene expression analysis) database, discussing the expression level of the genes in tumors, and carrying out survival analysis on an oncolnc website (http:// www.oncolnc.org /);
(6) The biological functions of the image histology module genes were studied by GO (gene ontology) and KEGG (kyoto gene and genome encyclopedia) analysis.
According to the invention, through obtaining the gene expression profile data and the magnetic resonance image data of a tumor patient, dividing the tumor area of each patient from the image data, extracting the image histology characteristics, constructing the image histology prognosis tag, then further using WGCNA to screen the module genes related to the image histology risk level, and analyzing and researching the biological functions of the image histology module genes through GO and KEGG, the interaction relation between the biological characteristics can be more accurately captured, a more interpretable result is provided, and the image histology module gene can be applied to various diseases, and has high practicability and application value.
There is also provided an apparatus for identifying image group biological characteristics based on WGCNA, comprising:
a gene and image data acquisition module configured to acquire gene expression profile data, image data, and clinical data of a tumor patient;
the image histology data processing module is configured to divide the tumor area of each patient from the image data and extract image histology characteristics;
constructing an image histology prognosis tag module, wherein the image histology prognosis tag module is configured to perform single-factor Cox analysis on magnetic resonance image characteristics, and select factors with P value smaller than 0.05 as image histology prognosis tags;
constructing a weighted co-expression network module, wherein the weighted co-expression network module is configured to construct a weighted co-expression network by using a WGCNA program package in R language, and identify a gene module most relevant to image risk classification by calculating gene saliency and module members;
screening a key gene module, wherein the key gene module is configured to screen a key gene by using Cytoscape, select a gene with the highest score to analyze in an Oncomine database, discuss the expression level of the gene in tumors, and perform survival analysis in an oncolnc website;
a research gene biology function module configured to study the biology function of the image histology module gene by GO and KEGG analysis.
Drawings
Fig. 1 shows a flowchart of a method for identifying image group biological characteristics based on WGCNA according to the present invention.
Detailed Description
Weighted gene co-expression network analysis (WGCNA) is a method of weighting a gene co-expression network, and by considering the weight of gene expression, the correlation between genes can be reflected more accurately. Image histology is a method of studying diseases using medical image data. The weighted gene co-expression network analysis method is applied to biological feature recognition of image group data, can find out gene modules related to biological features, can capture interaction relations among the biological features more accurately, and provides more interpretable results. Therefore, the research has important innovative significance for identifying biological characteristics of image group biology.
As shown in fig. 1, the method for identifying biological characteristics of image group based on WGCNA comprises the following steps:
(1) Acquiring gene expression profile data, image data and clinical data of a tumor patient;
(2) Dividing tumor areas of each patient from the image data, and extracting image histology characteristics;
(3) Performing single factor Cox analysis on the magnetic resonance image characteristics, and selecting factors with P value smaller than 0.05 as image histology prognosis labels;
(4) Constructing a weighted co-expression network by using a WGCNA program package in the R language, and identifying a gene module most relevant to image risk classification by calculating gene saliency and module members;
(5) Screening key genes by using Cytoscape, selecting the genes with highest scores, analyzing in an Oncomine database, discussing the expression level of the genes in tumors, and carrying out survival analysis on an oncolnc website;
(6) The biological function of the image histology module genes was studied by GO and KEGG analysis.
According to the invention, through obtaining the gene expression profile data and the magnetic resonance image data of a tumor patient, dividing the tumor area of each patient from the image data, extracting the image histology characteristics, constructing the image histology prognosis tag, then further using WGCNA to screen the module genes related to the image histology risk level, and analyzing and researching the biological functions of the image histology module genes through GO and KEGG, the interaction relation between the biological characteristics can be more accurately captured, a more interpretable result is provided, and the image histology module gene can be applied to various diseases, and has high practicability and application value.
Preferably, in the step (1), the tumor sample has transcriptome data and magnetic resonance image data.
Preferably, in the step (2), the FLAIR image of the patient is subjected to tumor three-dimensional segmentation by using ITK-SNAP software; the FLAIR scan parameters are as follows: layer thickness= 4.0~5.5mm,FLAIR TR/te=9000-12500/140-157 ms, interlayer spacing=4.0-6.5 mm, flip angle=80-90 °; the region of interest covered the entire tumor and edema area, and all image features were extracted using the pyradiomics insert in python 3.7.
Preferably, in the step (2), a plurality of patients are randomly selected, two people divide the ROI of interest, and the intra-group correlation coefficient ICC of the two ROIs is calculated; the image is preprocessed using gaussian and laplace filters, wavelet filters, and features computed by both filters include first order statistical features and statistical-based texture features.
Preferably, in the step (3), a single factor Cox analysis is performed on the magnetic resonance image characteristics, and a factor with a P value smaller than 0.05 is selected as an image histology prognosis tag; calculating a risk score for each patient by linearly combining the selected features according to respective coefficient weights; and dividing the patients into high-risk groups or low-risk groups according to the median of the risk scores, so as to judge the risk level.
Preferably, in the step (4), the genes of which the variance is 50% are screened for constructing a weighted co-expression network, pearson correlation coefficients between the genes are calculated, and the appropriate soft threshold β is selected so that the constructed network meets the standard of a scaleless network.
Preferably, in the step (5), a threshold value is set to be 0.85, a gene interaction network is exported, the network is imported into a Cytoscape to construct a sub-network, and a MCC method in a cytohubba plug-in is adopted to screen out key genes in a module network.
Preferably, in the step (6), the gene of the clusterizer package key module in the R language is used to perform GO function enrichment analysis so as to determine the biological process, molecular function and cell components involved in the gene in the module; KEGG pathway analysis was performed simultaneously to determine which signaling pathways these genes were involved in.
Preferably, the method further comprises statistical analysis, using intra-group phase relationship numbers to assess the repeatability of image features extracted by two radiologists from several patients, each feature being used for further extraction when ICC reaches 0.8; all statistical results were double-tailed, with P values less than 0.05 considered significant statistical differences.
It will be understood by those skilled in the art that all or part of the steps in implementing the above embodiment method may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, where the program when executed includes the steps of the above embodiment method, and the storage medium may be: ROM/RAM, magnetic disks, optical disks, memory cards, etc. Accordingly, the present invention also includes, corresponding to the method of the present invention, an apparatus for identifying biological features of the imaging group based on WGCNA, which is generally represented in the form of functional blocks corresponding to the steps of the method. The device comprises:
a gene and image data acquisition module configured to acquire gene expression profile data, image data, and clinical data of a tumor patient;
the image histology data processing module is configured to divide the tumor area of each patient from the image data and extract image histology characteristics;
constructing an image histology prognosis tag module, wherein the image histology prognosis tag module is configured to perform single-factor Cox analysis on magnetic resonance image characteristics, and select factors with P value smaller than 0.05 as image histology prognosis tags;
constructing a weighted co-expression network module, wherein the weighted co-expression network module is configured to construct a weighted co-expression network by using a WGCNA program package in R language, and identify a gene module most relevant to image risk classification by calculating gene saliency and module members;
screening a key gene module, wherein the key gene module is configured to screen a key gene by using Cytoscape, select a gene with the highest score to analyze in an Oncomine database, discuss the expression level of the gene in tumors, and perform survival analysis in an oncolnc website;
a research gene biology function module configured to study the biology function of the image histology module gene by GO and KEGG analysis.
The present invention is described in more detail below, including in particular:
the study collected 57 samples of glioblastoma in a public database, with complete transcriptomic information, magnetic resonance imaging information, and clinical information.
Tumor three-dimensional segmentation was performed on FLAIR images of patients using ITK-SNAP (www.itk-SNAP. Org) software. Scan parameters that require statistics include layer thickness (thickness), TR/TE, layer spacing (slice gap), flip angle (flip angle), etc. The region of interest needs to cover the entire tumor and edema area. All image histology features were extracted using the pyradiomics plugin (https:// pyradiomics. The image was preprocessed using gaussian and laplace (Laplacian of Gaussian, loG) filters, wavelet (wavelet) filters. The features computed by both filters include first order statistical features and statistical-based texture features.
And performing single-factor Cox analysis on the magnetic resonance image characteristics, and selecting factors with P value smaller than 0.05 as image histology prognosis labels. The risk score for each patient is calculated by linearly combining the selected features weighted by the respective coefficients. And dividing the patients into high-risk groups or low-risk groups according to the median of the risk scores, so as to judge the risk level.
The gene co-expression network was constructed from WGCNA package in R. Genes of which the first 50% of variance are screened for construction of a weighted co-expression network, pearson correlation coefficients between the genes are calculated, and an appropriate soft threshold β is selected so that the constructed network meets the criteria of a scaleless network. By calculating the gene saliency and the module members, the gene module most relevant to image risk classification is identified.
Setting the threshold value to be 0.85, exporting a gene interaction network, importing the network into a Cytoscape to construct a sub-network, and screening out key genes in a module network by adopting an MCC method in a cytohubba plug-in. The genes with the highest scores are selected and analyzed in an Oncomine database, the expression level of the genes in tumors is discussed, survival analysis is carried out on the oncolnc website, and the relation between the expression level and the prognosis of patients is analyzed.
GO functional enrichment analysis was performed using genes of clusterizer package key modules in the R language to clarify the biological processes (biological process, BP), molecular functions (molecular function, MF) and cellular components (cellular component, CC) involved in the genes in the modules. KEGG pathway analysis was performed simultaneously to determine which signaling pathways these genes were involved in.
The statistical analysis used R3.6.0, the R package used was as follows: the clusterifier package performs GO and KEGG analyses and the WGCNA package builds a weighted co-expression network. The reproducibility of image features extracted by two radiologists from 30 patients was evaluated using intra-group correlation numbers (ICCs), each feature being used for further extraction when ICC reached 0.8. All statistical results were double-tailed, with P values less than 0.05 considered significant statistical differences.
57 GBM samples were obtained from cancer genomic patterns (TCGA) and cancer imaging files (TCIA), wherein 33 men and 24 women, the average age of the patients was 59.4 years, and the median survival time was 1.38 years. After reading and preprocessing the expression matrix by using R software, 17801 gene expression data are obtained in total.
851 magnetic resonance image features were obtained by a pyradiomics plug-in, of which 744 features were obtained by wavelet filters, 93 features were obtained by LoG filters, and 14 features were based on shape and size. And (3) carrying out single-factor Cox survival analysis on image features, wherein 6 features with P values smaller than 0.05 (table 1) are included in a multi-factor Cox proportional risk model, and the median of image risk scores is calculated to be 1.185, 28 persons in a high-risk group and 29 persons in a low-risk group.
TABLE 1 image characteristics relating to prognosis
The total genes of the first 50% of the calculated variance were 8900. And determining a soft threshold beta to be 5 according to the standard of the non-scale network, and then obtaining a topology matrix according to the beta value. And combining modules with higher similarity of the characteristic genes (module eigengenes, ME) by using a dynamic mixing shearing method to finally obtain 15 gene modules, wherein the gray modules are genes which are not co-expressed.
Age (age), sex (sex), race (race), radiotherapy (radiation), chemotherapy (medium), time to live (time), risk score (risk_score), risk level (risk_level), status of life (status) clinical variables are included, and the Pearson correlation coefficient of the module characteristic gene and the corresponding variable represents the correlation of the module with the corresponding clinical characteristic and the image risk level. The lightcyan module correlated positively with image risk ranking (cor=0.29, P < 0.05) compared to the other modules, where there were 126 genes involved, COX8A, CYC1, EIF5A, etc.
The gene interaction network obtained in WGCNA comprises 92 nodes and 102 sides, and the 92 nodes and 102 sides are imported into cytoscape to obtain the key gene HMGA2. Meta analysis in the Oncomine database showed that HMGA2 was highly expressed in GBM (p=0.037). For the data set of this study, the gene was significantly highly expressed in tumor tissue with P <0.05. Survival analysis of the gene on the oncolnc website shows that tumor patients with high expression of the gene have poor prognosis, and the Logrank test p= 0.0396.
The genes in the Lightcyan module are analyzed by GO, and the genes are mainly involved in electron transfer activity (GO: 0009055~electron transfer activity), U6 snRNA binding (GO: 0017070~U6 snRNA binding), nucleoside triphosphatebisphosphates activity (GO: 0047429-nucleotide-triphosphate diphosphatase activity), nucleotide bisphosphates activity (GO: 0004551~nucleotide diphosphatase activity) and the like. KEGG analysis found that the above genes were mainly involved in nervous system diseases such as oxidative phosphorylation (Oxidative phosphorylation), drug metabolism-other enzymes (Drug metanolism-other enzymes), parkinsonism (Parkinson disease), senile dementia (Alzheimer disease), huntington's disease (Huntington disease), etc.
The present invention is not limited to the preferred embodiments, but can be modified in any way according to the technical principles of the present invention, and all such modifications, equivalent variations and modifications are included in the scope of the present invention.
Claims (10)
1. A method for identifying image group biological characteristics based on WGCNA, characterized by: which comprises the following steps:
(1) Acquiring gene expression profile data, image data and clinical data of a tumor patient;
(2) Dividing tumor areas of each patient from the image data, and extracting image histology characteristics;
(3) Performing single factor Cox analysis on the magnetic resonance image characteristics, and selecting factors with P value smaller than 0.05 as image histology prognosis labels;
(4) Constructing a weighted co-expression network by using a WGCNA program package in the R language, and identifying a gene module most relevant to image risk classification by calculating gene saliency and module members;
(5) Screening key genes by using Cytoscape, selecting the genes with highest scores, analyzing in an Oncomine database, discussing the expression level of the genes in tumors, and carrying out survival analysis on an oncolnc website;
(6) The biological function of the image histology module genes was studied by GO and KEGG analysis.
2. The method for identifying image group biological characteristics based on WGCNA according to claim 1, wherein: in the step (1), the tumor sample has transcriptome data and magnetic resonance image data.
3. The method for identifying image group biological characteristics based on WGCNA according to claim 2, wherein: in the step (2), the FLAIR image of the patient is subjected to tumor three-dimensional segmentation by using ITK-SNAP software; the FLAIR scan parameters are as follows: layer thickness= 4.0~5.5mm,FLAIR TR/te=9000-12500/140-157 ms, interlayer spacing=4.0-6.5 mm, flip angle=80-90 °; the region of interest covered the entire tumor and edema area, and all image features were extracted using the pyradiomics insert in python 3.7.
4. The method for identifying image group biological characteristics based on WGCNA according to claim 2, wherein: in the step (2), a plurality of patients are randomly selected, two persons divide the ROI of interest, and the intra-group correlation coefficient ICC of the two ROIs is calculated; the image is preprocessed using gaussian and laplace filters, wavelet filters, and features computed by both filters include first order statistical features and statistical-based texture features.
5. The method for identifying image group biological characteristics based on WGCNA according to claim 4, wherein: in the step (3), single factor Cox analysis is carried out on the magnetic resonance image characteristics, and factors with P value smaller than 0.05 are selected and used as image histology prognosis labels; calculating a risk score for each patient by linearly combining the selected features according to respective coefficient weights; and dividing the patients into high-risk groups or low-risk groups according to the median of the risk scores, so as to judge the risk level.
6. The method for identifying image group biological characteristics based on WGCNA according to claim 5, wherein: in the step (4), the genes with the first 50% of variance are screened for constructing a weighted coexpression network, pearson correlation coefficients among the genes are calculated, and an appropriate soft threshold beta is selected so that the constructed network meets the standard of a scaleless network.
7. The method for identifying image group biological characteristics based on WGCNA according to claim 6, wherein: in the step (5), a threshold value is set to be 0.85, a gene interaction network is exported, the network is imported into a Cytoscape to construct a sub-network, and a MCC method in a cytohubba plug-in is adopted to screen out key genes in a module network.
8. The method for identifying image group biological characteristics based on WGCNA according to claim 7, wherein: in the step (6), performing GO function enrichment analysis by using genes of a clusterif iotaler package key module in the R language to determine biological processes, molecular functions and cell components involved in the genes in the module; KEGG pathway analysis was performed simultaneously to determine which signaling pathways these genes were involved in.
9. The method for identifying image group biological characteristics based on WGCNA according to claim 8, wherein: the method further includes statistical analysis, using intra-group phase relationship numbers to evaluate repeatability of image features extracted by two radiologists from several patients, each feature being used for further extraction when ICC reaches 0.8; all statistical results were double-tailed, with P values less than 0.05 considered significant statistical differences.
10. A device for identifying image group biological characteristics based on WGCNA, characterized in that: it comprises the following steps:
a gene and image data acquisition module configured to acquire gene expression profile data, image data, and clinical data of a tumor patient;
the image histology data processing module is configured to divide the tumor area of each patient from the image data and extract image histology characteristics;
constructing an image histology prognosis tag module, wherein the image histology prognosis tag module is configured to perform single-factor Cox analysis on magnetic resonance image characteristics, and select factors with P value smaller than 0.05 as image histology prognosis tags;
constructing a weighted co-expression network module, wherein the weighted co-expression network module is configured to construct a weighted co-expression network by using a WGCNA program package in R language, and identify a gene module most relevant to image risk classification by calculating gene saliency and module members;
screening a key gene module, wherein the key gene module is configured to screen a key gene by using Cytoscape, select a gene with the highest score to analyze in an Oncomine database, discuss the expression level of the gene in tumors, and perform survival analysis in an oncolnc website;
a research gene biology function module configured to study the biology function of the image histology module gene by GO and KEGG analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311377289.4A CN117116339A (en) | 2023-10-24 | 2023-10-24 | Method and device for identifying image group biological characteristics based on WGCNA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311377289.4A CN117116339A (en) | 2023-10-24 | 2023-10-24 | Method and device for identifying image group biological characteristics based on WGCNA |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117116339A true CN117116339A (en) | 2023-11-24 |
Family
ID=88813247
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311377289.4A Pending CN117116339A (en) | 2023-10-24 | 2023-10-24 | Method and device for identifying image group biological characteristics based on WGCNA |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117116339A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022188490A1 (en) * | 2021-03-11 | 2022-09-15 | 中国科学院深圳先进技术研究院 | Survival time prediction method and system based on imaging genomics |
| CN116385441A (en) * | 2023-06-05 | 2023-07-04 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of oligodendroglioma based on MRI |
| CN116705296A (en) * | 2023-06-06 | 2023-09-05 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of GBM patient based on conventional MRI sequence |
-
2023
- 2023-10-24 CN CN202311377289.4A patent/CN117116339A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022188490A1 (en) * | 2021-03-11 | 2022-09-15 | 中国科学院深圳先进技术研究院 | Survival time prediction method and system based on imaging genomics |
| CN116385441A (en) * | 2023-06-05 | 2023-07-04 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of oligodendroglioma based on MRI |
| CN116705296A (en) * | 2023-06-06 | 2023-09-05 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of GBM patient based on conventional MRI sequence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11610307B2 (en) | Determining biomarkers from histopathology slide images | |
| US11348661B2 (en) | Predicting total nucleic acid yield and dissection boundaries for histology slides | |
| CN112086129B (en) | Method and system for predicting cfDNA of tumor tissue | |
| US8068988B2 (en) | Method for automated processing of digital images of tissue micro-arrays (TMA) | |
| US20210166380A1 (en) | Predicting total nucleic acid yield and dissection boundaries for histology slides | |
| KR20210145778A (en) | Method for Determination of Biomarkers from Histopathology Slide Images | |
| CN110245657B (en) | Pathological image similarity detection method and detection device | |
| US20220051410A1 (en) | 3d radiomic platform for imaging biomarker development | |
| Fuchs et al. | Computational pathology analysis of tissue microarrays predicts survival of renal clear cell carcinoma patients | |
| US11348240B2 (en) | Predicting total nucleic acid yield and dissection boundaries for histology slides | |
| CN107169497A (en) | A kind of tumor imaging label extracting method based on gene iconography | |
| WO2018157381A1 (en) | Method and apparatus for intelligently classifying pathological slice image | |
| CN115170857A (en) | Pancreatic cancer image identification method based on federal transfer learning | |
| Giuste et al. | Explainable synthetic image generation to improve risk assessment of rare pediatric heart transplant rejection | |
| WO2022147503A1 (en) | Predicting total nucleic acid yield and dissection boundaries for histology slides | |
| CN117116339A (en) | Method and device for identifying image group biological characteristics based on WGCNA | |
| US20040086873A1 (en) | System and method of generating and storing correlated hyperquantified tissue structure and biomolecular expression datasets | |
| Li et al. | A novel approach of computer-aided detection of focal ground-glass opacity in 2D lung CT images | |
| CN110751983A (en) | Method for screening characteristic mRNA (messenger ribonucleic acid) for diagnosing early lung cancer | |
| Shehata et al. | Algorithms of Deep Learning: Convolutional Neural Network Role with Colon Cancer Disease | |
| Wajeed et al. | A Breast Cancer Image Classification Algorithm with 2c Multiclass Support Vector Machine | |
| Verma | Impact of multiple clusters on neural classification of ROIs in digital mammograms | |
| Doranga et al. | Automated Classification of Genetic Mutations in Cancer using Machine Learning | |
| CN117133466A (en) | Survival prediction method and device based on transcriptomics and image histology | |
| Rozikin et al. | ANOVA-Artificial Bee Colony Algorithm-Driven Feature Selection for Classifying Downy Mildew Severity in Melon Leaves. |
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 |