TWI823203B - Automated multi-gene assisted diagnosis of autoimmune diseases - Google Patents
Automated multi-gene assisted diagnosis of autoimmune diseases Download PDFInfo
- Publication number
- TWI823203B TWI823203B TW110145353A TW110145353A TWI823203B TW I823203 B TWI823203 B TW I823203B TW 110145353 A TW110145353 A TW 110145353A TW 110145353 A TW110145353 A TW 110145353A TW I823203 B TWI823203 B TW I823203B
- Authority
- TW
- Taiwan
- Prior art keywords
- gene
- autoimmune diseases
- automated multi
- model
- autoimmune disease
- Prior art date
Links
- 208000023275 Autoimmune disease Diseases 0.000 title claims abstract description 76
- 238000003745 diagnosis Methods 0.000 title claims abstract description 23
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 79
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000010801 machine learning Methods 0.000 claims abstract description 5
- 238000004458 analytical method Methods 0.000 claims description 16
- 230000014509 gene expression Effects 0.000 claims description 11
- 238000010219 correlation analysis Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000000546 chi-square test Methods 0.000 claims description 5
- 208000024891 symptom Diseases 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 238000000611 regression analysis Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 2
- 238000007619 statistical method Methods 0.000 abstract description 6
- 238000006243 chemical reaction Methods 0.000 abstract description 3
- 238000004393 prognosis Methods 0.000 abstract description 2
- 230000002596 correlated effect Effects 0.000 description 12
- 206010025135 lupus erythematosus Diseases 0.000 description 9
- 206010039073 rheumatoid arthritis Diseases 0.000 description 9
- 201000010099 disease Diseases 0.000 description 6
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 4
- 230000002068 genetic effect Effects 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 238000011282 treatment Methods 0.000 description 3
- 108700028369 Alleles Proteins 0.000 description 2
- 102100036242 HLA class II histocompatibility antigen, DQ alpha 2 chain Human genes 0.000 description 2
- 108010086786 HLA-DQA1 antigen Proteins 0.000 description 2
- 208000021386 Sjogren Syndrome Diseases 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000026278 immune system disease Diseases 0.000 description 2
- 206010002556 Ankylosing Spondylitis Diseases 0.000 description 1
- 101001011442 Homo sapiens Interferon regulatory factor 5 Proteins 0.000 description 1
- 101000829779 Homo sapiens Probable G-protein coupled receptor 19 Proteins 0.000 description 1
- 101000764263 Homo sapiens Tumor necrosis factor ligand superfamily member 4 Proteins 0.000 description 1
- 101000984551 Homo sapiens Tyrosine-protein kinase Blk Proteins 0.000 description 1
- 102100030131 Interferon regulatory factor 5 Human genes 0.000 description 1
- 102000014150 Interferons Human genes 0.000 description 1
- 108010050904 Interferons Proteins 0.000 description 1
- 102100023417 Probable G-protein coupled receptor 19 Human genes 0.000 description 1
- 201000001263 Psoriatic Arthritis Diseases 0.000 description 1
- 208000036824 Psoriatic arthropathy Diseases 0.000 description 1
- 108010019992 STAT4 Transcription Factor Proteins 0.000 description 1
- 102000005886 STAT4 Transcription Factor Human genes 0.000 description 1
- 102100026890 Tumor necrosis factor ligand superfamily member 4 Human genes 0.000 description 1
- 102100027053 Tyrosine-protein kinase Blk Human genes 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 229940079322 interferon Drugs 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
本發明係提供一種自動化多基因輔助診斷自體免疫疾病之方法，其係先收集大量罹患自體免疫疾病患者之基因位點資料，並且進行基因位點資料之轉換處理，而後，將轉換處理後之資料投入一處理模組中，透過統計分析或/及機器學習方式進行演算分析，進而從中得知該些基因位點對於各該自體免疫疾病之發生或預後之重要性。The present invention provides an automated multi-gene auxiliary diagnosis method for autoimmune diseases. It first collects a large number of gene locus data of patients suffering from autoimmune diseases, and performs conversion processing of the gene locus data, and then converts the converted data. The data is put into a processing module and analyzed through statistical analysis or/and machine learning methods to learn the importance of these gene loci for the occurrence or prognosis of each autoimmune disease.
Description
本發明係有關於一種人工智慧用於基因分析之方法，特別係指一種自動化多基因輔助診斷自體免疫疾病之方法。 The present invention relates to a method for using artificial intelligence for gene analysis, and in particular, to an automated multi-gene assisted diagnosis method for autoimmune diseases.
按，根據統計，於美國，罹患超過1種自體免疫疾病之人口數約有5千萬人；於台灣，自體免疫疾病為重大傷病排名之第三位，罹患自體免疫疾病之人口數逐年增加，每年約新增4000人。臨床上已知之自體免疫疾病超過70種，如類風濕性關節炎、僵直性脊椎炎、乾癬性關節炎、紅斑性狼瘡、乾燥症等，但該些自體免疫疾病之病症表現複雜且不同自體免疫疾病會具有共同臨床特徵，並且會受到患者遺傳背景、生活環境等影響而使病症重疊，導致臨床上對於自體免疫疾病之區分不容易，甚至許多自體免疫疾病患者難以被確診，進而導致無法即時提供正確治療策略。 According to statistics, in the United States, the number of people suffering from more than one autoimmune disease is approximately 50 million; in Taiwan, autoimmune diseases rank third among major injuries and illnesses, and the number of people suffering from autoimmune diseases It is increasing year by year, with about 4,000 new people added every year. There are more than 70 clinically known autoimmune diseases, such as rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, lupus erythematosus, Sjogren's disease, etc. However, the symptoms of these autoimmune diseases are complex and different. Autoimmune diseases will have common clinical characteristics, and will be affected by the patient's genetic background, living environment, etc., causing overlapping symptoms. This makes it difficult to distinguish autoimmune diseases clinically, and even many patients with autoimmune diseases are difficult to diagnose. This results in the inability to provide the correct treatment strategy immediately.
雖然目前有研究係透過單變量的統計模型來分析單一基因位點與疾病間之關聯性，但是該些研究都是針對自體免疫疾病患者與健康者間之比較，並且都僅能評估單一基因位點，不僅無法考慮到多個基因交互作用與疾病間之關聯性，更無法用於同時判斷多種自體免疫疾病或是進行自體免疫疾病之分類。舉例來說，罹患類風濕關節炎和紅斑狼瘡之患者中都可能會發生RHUPUS綜合徵 或第1型干擾素過度表現之臨床特徵，並且研究指出類風濕關節炎和紅斑狼瘡之患者具有相同之遺傳特徵，因此於臨床上極易發生誤判或無法區分之情形。 Although there are current studies that use univariate statistical models to analyze the association between a single gene locus and disease, these studies are all aimed at comparing patients with autoimmune diseases and healthy people, and they can only evaluate a single gene. Loci not only cannot take into account the correlation between multiple gene interactions and diseases, but also cannot be used to simultaneously judge multiple autoimmune diseases or classify autoimmune diseases. For example, RHUPUS syndrome may occur in patients with rheumatoid arthritis and lupus erythematosus. Or the clinical characteristics of excessive expression of type 1 interferon, and studies have shown that patients with rheumatoid arthritis and lupus erythematosus have the same genetic characteristics, so it is easy to be misdiagnosed or indistinguishable in clinical practice.
本發明之主要目的在於提供一種自動化多基因輔助診斷自體免疫疾病之方法，其係能夠用以提供臨床上區分自體免疫疾病之候選基因及其相對應之重要度資訊，以能於發病初期即能正確診斷自體免疫疾病，並且能夠提供適合之治療手段。 The main purpose of the present invention is to provide an automated multi-gene auxiliary diagnosis method for autoimmune diseases, which can be used to provide candidate genes for clinical differentiation of autoimmune diseases and their corresponding importance information, so as to diagnose autoimmune diseases in the early stages. That is, autoimmune diseases can be correctly diagnosed and suitable treatments can be provided.
本發明之另一目的在於提供一種自動化多基因輔助診斷自體免疫疾病之方法，其係將資料處理技術及機器學習演算法用於多基因位點與自體免疫疾病間關連性之分析，以綜合判斷與自體免疫疾病相關之變數，提高診斷及治療方案之準確度。 Another object of the present invention is to provide an automated multi-gene auxiliary diagnosis method for autoimmune diseases, which uses data processing technology and machine learning algorithms to analyze the correlation between multi-gene loci and autoimmune diseases, so as to Comprehensively judge variables related to autoimmune diseases to improve the accuracy of diagnosis and treatment plans.
緣是，為能達成上述目的，本發明係揭露一種自動化多基因輔助診斷自體免疫疾病之方法，其係透過分析臨床表現相近之自體免疫疾病患者及其基因位點資訊，透過關連性分析篩選出與自體免疫疾病具高度關連性之基因位點，再將所有被篩選出之基因位點作為演算變數並以一預定模型進行演算，得到一解析結果，用以顯示該些被篩選出之基因位點對於該自體免疫疾病之重要度。 Therefore, in order to achieve the above purpose, the present invention discloses an automated multi-gene auxiliary diagnosis method for autoimmune diseases, which is based on analyzing autoimmune disease patients with similar clinical manifestations and their gene locus information, and through correlation analysis Screen out gene loci that are highly correlated with autoimmune diseases, and then use all screened gene loci as calculation variables and perform calculations with a predetermined model to obtain an analysis result to display the screened out gene loci. The importance of the genetic locus for the autoimmune disease.
於本發明之實施例中，該自動化多基因輔助診斷自體免疫疾病之方法主要係包含有下列步驟： In an embodiment of the present invention, the method for automated multi-gene assisted diagnosis of autoimmune diseases mainly includes the following steps:
步驟a：取得複數自體免疫疾病患者所提供之基因位點及其資訊，其中，該些自體免疫疾病患者係以具有臨床表現相近病症者為佳。 Step a: Obtain gene loci and information provided by multiple autoimmune disease patients, preferably those with similar clinical manifestations.
步驟b：處理該些自體免疫疾病患者之基因位點資料，依據基因型之不同，將各基因位點資訊轉換為一可分析資料。 Step b: Process the gene locus data of these patients with autoimmune diseases, and convert each gene locus information into analyzable data based on different genotypes.
步驟c：將各該基因位點之該可分析資料與該些自體免疫疾病進行關聯性分析，自該些基因位點篩選出與一特定自體免疫疾病高度相關之一候選基因位點。 Step c: Perform correlation analysis on the analyzable data of each gene locus and the autoimmune diseases, and select candidate gene loci that are highly correlated with a specific autoimmune disease from the gene loci.
步驟d：將該候選基因位點以一預定模型進行演算，得到一解析結果，其中，該解析結果係為該特定自體免疫疾病所對應之該候選基因的重要性資訊，而此所指重要性資訊係為與該預定自體免疫疾病相關之該候選基因之重要度排序、重要值、影響比重值、影響程度等。 Step d: Calculate the candidate gene locus with a predetermined model to obtain an analysis result, wherein the analysis result is the importance information of the candidate gene corresponding to the specific autoimmune disease, and this refers to the importance The sexual information is the importance ranking, importance value, influence proportion value, influence degree, etc. of the candidate genes related to the predetermined autoimmune disease.
其中，該步驟b中之該可分析資料係分析各基因位點之性狀表現後之結果，而該可分析資料之轉換標準係以各基因位點及其性狀表現進行定義。 Among them, the analyzable data in step b is the result of analyzing the trait expression of each gene locus, and the conversion standard of the analyzable data is defined based on each gene locus and its trait expression.
於本發明之一實施例中，該步驟c中係透過以一統計分析方法進行關聯性分析，判斷各基因位點與該特定自體免疫疾病之關連性，當該基因位點進行關連性分析後之結果高於一預定關連值時，表示該基因位點與該特定自體免疫疾病具高度相關而被認定為該候選基因位點；其中，該預定關連值係可依據所使用之統計方法而設定之，例如p值。 In one embodiment of the present invention, in step c, a correlation analysis is performed using a statistical analysis method to determine the correlation between each gene locus and the specific autoimmune disease. When the correlation analysis is performed on the gene locus When the subsequent result is higher than a predetermined correlation value, it means that the gene locus is highly correlated with the specific autoimmune disease and is identified as a candidate gene locus; wherein, the predetermined correlation value can be based on the statistical method used. And set it, such as p value.
其中，統計分析方法係為本領域所屬技術領域且具通常知識者所周知的關連性分析法，如卡方檢定與羅吉斯迴歸分析。 Among them, the statistical analysis method is a correlation analysis method that is well known to those with ordinary knowledge in the technical field of this field, such as chi-square test and Logis regression analysis.
於本發明之又一實施例中，為能提升分析各變數對於自體免疫疾病之分類及區分準確率，於該步驟d中係採取機器學習演算法進行演算，其中，該預定模型係包含有羅吉斯迴歸模型、決策樹模型、隨機森林模型、支持向量機模型、梯度提升樹模型或極限梯度提升模型。 In another embodiment of the present invention, in order to improve the accuracy of analyzing various variables for classifying and distinguishing autoimmune diseases, a machine learning algorithm is used in step d, where the predetermined model includes Logis regression model, decision tree model, random forest model, support vector machine model, gradient boosting tree model or extreme gradient boosting model.
於本發明之另一實施例中，為能更方便醫療人員判讀解析結果，本發明所揭自動化多基因輔助診斷自體免疫疾病之方法，其更包含有一步驟e， 位於該步驟d之後，用以將步驟d中之該解析結果進行視覺化處理程序，而得以圖形、圖表、表格等方式呈現，並得以數字、顏色、幾何圖形等作為表現元素。 In another embodiment of the present invention, in order to make it easier for medical personnel to interpret the analysis results, the method of automated multi-gene auxiliary diagnosis of autoimmune diseases disclosed in the present invention further includes a step e, After step d, it is used to perform a visual processing procedure on the analysis result in step d, and present it in the form of graphics, charts, tables, etc., and use numbers, colors, geometric figures, etc. as expression elements.
圖1係以曼哈頓圖形篩選與類風濕性關節炎有高度相關之基因位點的結果。 Figure 1 shows the results of screening gene loci that are highly correlated with rheumatoid arthritis using Manhattan graphics.
圖2係以曼哈頓圖形篩選與紅斑性狼瘡有高度相關之基因位點的結果。 Figure 2 is the result of using Manhattan pattern to screen gene loci that are highly correlated with lupus erythematosus.
圖3係以曼哈頓圖形篩選與乾燥症有高度相關之基因位點的結果。 Figure 3 shows the results of screening gene loci that are highly correlated with Sjögren's syndrome using Manhattan patterns.
圖4係以梯度提升樹模型評估與紅斑性狼瘡及類風濕性關節炎有高度相關之基因位點，並以SHAP總結圖形呈現之結果。 Figure 4 is a gradient boosting tree model used to evaluate gene loci that are highly correlated with lupus erythematosus and rheumatoid arthritis, and the results are presented in a SHAP summary graphic.
圖5係以極限梯度提升模型評估與紅斑性狼瘡及類風濕性關節炎有高度相關之基因位點，並以SHAP總結圖形呈現的結果。 Figure 5 uses the extreme gradient boosting model to evaluate gene loci that are highly correlated with lupus erythematosus and rheumatoid arthritis, and summarizes the results graphically with SHAP.
本發明係提供一種自動化多基因輔助診斷自體免疫疾病之方法，其能夠建立一套以基因位點判斷或分類自體免疫疾病之標準，不僅能夠用以提供醫療人員於臨床上分析病患基因位點數據之用，亦能夠作為相關醫療器材或診斷治療系統之主要技術來源，如晶片、分析軟體等。更進一步來說，本發明所揭自動化多基因輔助診斷自體免疫疾病之方法係先收集大量罹患自體免疫疾病患者之基因位點資料，並且進行基因位點資料之轉換處理，而後，將轉換處理後之資料投入一處理模組中，透過統計分析或/及機器學習方式進行演算分析，進而從中得知該些基因位點對於各該自體免疫疾病之發生或預後之重要性。 The present invention provides an automated multi-gene auxiliary diagnosis method for autoimmune diseases, which can establish a set of standards for judging or classifying autoimmune diseases based on gene loci, and can not only provide medical personnel with the ability to clinically analyze patient genes The use of site data can also be used as the main technology source for related medical equipment or diagnostic and treatment systems, such as chips, analysis software, etc. Furthermore, the method of automated multi-gene assisted diagnosis of autoimmune diseases disclosed in the present invention first collects a large number of gene locus data of patients suffering from autoimmune diseases, and performs conversion processing of the gene locus data, and then converts the gene locus data. The processed data is put into a processing module and analyzed through statistical analysis or/and machine learning methods to learn the importance of these gene loci for the occurrence or prognosis of each autoimmune disease.
於本發明之一實施例中係揭露一種自動化多基因輔助診斷自體免疫疾病之方法，其主要包含下列步驟： In one embodiment of the present invention, a method for automated multi-gene assisted diagnosis of autoimmune diseases is disclosed, which mainly includes the following steps:
步驟a：取得複數自體免疫疾病患者所提供之基因位點及其資訊，其中，該些自體免疫疾病患者具有臨床表現相近之症狀。 Step a: Obtain gene loci and information provided by multiple autoimmune disease patients, wherein these autoimmune disease patients have similar clinical symptoms.
步驟b：分析該些自體免疫疾病患者之基因位點資料，分析各基因位點之性狀表現，依據性狀表現之不同而轉換為一可分析資料，並該可分析資料係為一數值形態之資料，舉例來說，若A為主要(顯性)的等位基因，a為次要(隱性)的等位基因，則定義AA、Aa、aa此三種基因型分別被轉換為數值：2、1、0。 Step b: Analyze the gene locus data of patients with autoimmune diseases, analyze the trait expression of each gene locus, and convert it into an analyzable data based on the different trait expressions, and the analyzable data is in the form of a numerical value Data, for example, if A is the major (dominant) allele and a is the minor (recessive) allele, then the three genotypes AA, Aa, and aa are defined and converted into numerical values: 2 ,1,0.
步驟c：將各該基因位點之該可分析資料與該些自體免疫疾病進行以統計分析方法進行關聯性分析，並依p值結果，以調整後之檢定門檻篩選出與一特定自體免疫疾病具高度相關之基因位點，而該些被篩選出之基因位點係為候選基因位點。 Step c: Perform correlation analysis between the analyzable data of each gene locus and the autoimmune diseases using statistical analysis methods, and based on the p-value results, use the adjusted test threshold to screen out genes associated with a specific autologous Immune diseases have highly correlated gene loci, and these screened gene loci are candidate gene loci.
步驟d：以一預定演算模型評估該些候選基因位點變數，而該預定演算模型係得為羅吉斯迴歸模型、決策樹模型、隨機森林模型、支持向量機模型、梯度提升樹模型、極限梯度提升模型或其他於該領域中具有代表性之演算模型進行演算，得到一解析結果，其中，該解析結果係為針對該特定自體免疫疾病找出對應候選基因及其重要性資訊，例如各該候選基因對於該特定自體免疫疾病的重要性排名、影響比重(值)或其他可以顯示其重要度之表示方式。 Step d: Use a predetermined algorithm model to evaluate the candidate gene locus variables, and the predetermined algorithm model may be a Logis regression model, a decision tree model, a random forest model, a support vector machine model, a gradient boosting tree model, or a limit model. The gradient boosting model or other representative calculation models in this field are calculated to obtain an analysis result. The analysis result is to find the corresponding candidate genes and their importance information for the specific autoimmune disease, such as each The candidate gene's importance ranking, impact proportion (value), or other expressions that can show its importance for the specific autoimmune disease.
其中，該候選基因位點變數間之交互影響係透過哈溫定律與連鎖不平衡定律來進行評估。 Among them, the interaction between variables of the candidate gene locus is evaluated through Harwin's law and the law of linkage disequilibrium.
步驟e：將該解析結果進行視覺化處理程序，意即以圖像之方式呈現該些候選基因對於該特定自體免疫疾病之重要性及貨影響度，例如直條圖、圓餅圖、SHAP(SHapley Additive exPlanations)總結圖形等。 Step e: Visualize the analysis results, which means to present the importance and impact of the candidate genes on the specific autoimmune disease in the form of images, such as bar charts, pie charts, SHAP (SHapley Additive exPlanations) summary graphics, etc.
藉由上述步驟之組成，本發明所揭自動化多基因輔助診斷自體免疫疾病之方法係先篩選出與特定自體免疫疾病高度相關之候選基因位點後，再透過模型進行演算，以評估所有候選基因位點及其彼此間之交互作用，已得到候選基因位點對於特定自體免疫疾病重要度及影響度之結果，並且進一步藉由將資料進行視覺化處理之程序，使各自體免疫疾病之候選基因位點的重要性與影響性可以圖樣方式呈現，以達到提供評估、診斷、區分自體免疫疾病之標準或參數。 Through the composition of the above steps, the method of automated multi-gene auxiliary diagnosis of autoimmune diseases disclosed in the present invention first selects candidate gene loci that are highly related to specific autoimmune diseases, and then performs calculations through the model to evaluate all Candidate gene loci and their interactions have resulted in the importance and impact of candidate gene loci on specific autoimmune diseases. Furthermore, through the process of visualizing the data, the results of each autoimmune disease have been obtained. The importance and influence of candidate gene loci can be presented graphically to provide standards or parameters for assessment, diagnosis, and differentiation of autoimmune diseases.
更進一步來說，類風濕性關節炎、紅斑性狼瘡及乾燥症皆為自體免疫疾病且具有相近臨床特徵，故收集該三種自體免疫疾病患者的基因位點及其資訊，並且以本發明所揭自動化多基因輔助診斷自體免疫疾病之方法處理該些基因位點資訊，可先篩選出與各疾病有高度相關之基因位點，其中，於本發明所揭自動化多基因輔助診斷自體免疫疾病之方法的步驟c中所使用篩選關聯性基因位點之判斷門檻係同時使用卡方測試及羅吉斯迴歸分析，並進行卡方測試時，p值設定為小於10-8，篩選後之結果如圖1至圖3所示。，由圖1可知與類風濕性關節炎具高度相關之基因位點(候選基因位點)為HLA-DQA1；由圖2可知與紅斑性狼瘡具高度相關之基因位點(候選基因位點)包含有TNFSF4、STAT4、HLA-DQA1、IRF5、BLK、GPR19、GTF21AC211433.1等；而由圖3可知未有發現高於預設門檻之基因位點存在。 Furthermore, rheumatoid arthritis, lupus erythematosus and Sjögren's disease are all autoimmune diseases and have similar clinical characteristics. Therefore, the genetic loci and information of patients with these three autoimmune diseases are collected, and based on the present invention The disclosed method of automated multi-gene assisted diagnosis of autoimmune diseases processes the gene locus information and can first screen out gene loci that are highly correlated with each disease. Among them, the automated multi-gene assisted diagnosis of autoimmune diseases disclosed in the present invention The judgment threshold for screening related gene loci used in step c of the immune disease method is to use both chi-square test and Logis regression analysis. When performing chi-square test, the p value is set to less than 10 -8 . After screening, The results are shown in Figures 1 to 3. , Figure 1 shows that the gene locus (candidate gene locus) that is highly correlated with rheumatoid arthritis is HLA-DQA1; Figure 2 shows that the gene locus that is highly correlated with lupus erythematosus (candidate gene locus) It includes TNFSF4, STAT4, HLA-DQA1, IRF5, BLK, GPR19, GTF21AC211433.1, etc.; and as shown in Figure 3, no gene loci above the preset threshold have been found.
而後根據所得到之與類風濕性關節炎及紅斑性狼瘡疾病有高度相關之基因位點後，再分別以梯度提升樹模型及極限梯度提升模型作為演算模型，評估與疾相關基因位點變數後得到對應疾病的候選基因位點及其變數間交互作用之解析結果，進而透過資料視覺化之方式，如SHAP總結圖形將解析結果呈現，如圖4及圖5所示。 Then, based on the obtained gene loci that are highly related to rheumatoid arthritis and lupus erythematosus, the gradient boosting tree model and the extreme gradient boosting model were used as calculation models to evaluate the variables of the disease-related gene loci. The analysis results of the candidate gene loci corresponding to the disease and the interactions between the variables are obtained, and then the analysis results are presented through data visualization, such as SHAP summary graphics, as shown in Figures 4 and 5.
Claims (7)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW110145353A TWI823203B (en) | 2021-12-03 | 2021-12-03 | Automated multi-gene assisted diagnosis of autoimmune diseases |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW110145353A TWI823203B (en) | 2021-12-03 | 2021-12-03 | Automated multi-gene assisted diagnosis of autoimmune diseases |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW202324447A TW202324447A (en) | 2023-06-16 |
| TWI823203B true TWI823203B (en) | 2023-11-21 |
Family
ID=87803657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW110145353A TWI823203B (en) | 2021-12-03 | 2021-12-03 | Automated multi-gene assisted diagnosis of autoimmune diseases |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI823203B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105229649A (en) * | 2013-03-15 | 2016-01-06 | 百世嘉（上海）医疗技术有限公司 | For the human genome analysis of variance of disease association and the system and method for report |
| CN105980578A (en) * | 2013-12-16 | 2016-09-28 | 考利达基因组股份有限公司 | Basecaller for DNA sequencing using machine learning |
| CN106202936A (en) * | 2016-07-13 | 2016-12-07 | 为朔医学数据科技（北京）有限公司 | A kind of disease risks Forecasting Methodology and system |
| CN104830963B (en) * | 2014-02-12 | 2018-05-25 | 大江基因医学股份有限公司 | For in vitro measuring the method for causing risk of obesity |
| CN110301899A (en) * | 2019-07-01 | 2019-10-08 | 湘南学院附属医院 | A kind of cardiovascular and cerebrovascular disease information detecting system and method |
| US20200105413A1 (en) * | 2018-09-29 | 2020-04-02 | Roche Molecular Systems, Inc. | Multimodal machine learning based clinical predictor |
| CN113674863A (en) * | 2021-08-18 | 2021-11-19 | 河海大学 | Intelligent disease prevention and control and health care recommendation method for big data gene detection |
-
2021
- 2021-12-03 TW TW110145353A patent/TWI823203B/en active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105229649A (en) * | 2013-03-15 | 2016-01-06 | 百世嘉（上海）医疗技术有限公司 | For the human genome analysis of variance of disease association and the system and method for report |
| CN105980578A (en) * | 2013-12-16 | 2016-09-28 | 考利达基因组股份有限公司 | Basecaller for DNA sequencing using machine learning |
| CN104830963B (en) * | 2014-02-12 | 2018-05-25 | 大江基因医学股份有限公司 | For in vitro measuring the method for causing risk of obesity |
| CN106202936A (en) * | 2016-07-13 | 2016-12-07 | 为朔医学数据科技（北京）有限公司 | A kind of disease risks Forecasting Methodology and system |
| US20200105413A1 (en) * | 2018-09-29 | 2020-04-02 | Roche Molecular Systems, Inc. | Multimodal machine learning based clinical predictor |
| CN110301899A (en) * | 2019-07-01 | 2019-10-08 | 湘南学院附属医院 | A kind of cardiovascular and cerebrovascular disease information detecting system and method |
| CN113674863A (en) * | 2021-08-18 | 2021-11-19 | 河海大学 | Intelligent disease prevention and control and health care recommendation method for big data gene detection |
Also Published As
| Publication number | Publication date |
|---|---|
| TW202324447A (en) | 2023-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Richter et al. | Telemedicine for retinopathy of prematurity diagnosis: evaluation and challenges | |
| Mahendran et al. | A deep learning framework with an embedded-based feature selection approach for the early detection of the Alzheimer's disease | |
| Papik et al. | Application of neural networks in medicine-a review | |
| Peng et al. | Predicting risk of late age-related macular degeneration using deep learning | |
| KR101542529B1 (en) | Examination methods of the bio-marker of allele | |
| CN1674028A (en) | Diagnostic decision support system and method of diagnostic decision support | |
| CN112884754A (en) | Multi-modal Alzheimer's disease medical image recognition and classification method and system | |
| Huang et al. | A structure-related fine-grained deep learning system with diversity data for universal glaucoma visual field grading | |
| Bao et al. | Current application of digital diagnosing systems for retinopathy of prematurity | |
| JP2022500068A (en) | Methods and systems for assessing the risk of glaucoma | |
| Zhao et al. | Eye-brain connections revealed by multimodal retinal and brain imaging genetics in the UK Biobank | |
| Gurdon et al. | Pursuit of precision medicine: Systems biology approaches in Alzheimer's disease mouse models | |
| TWI823203B (en) | Automated multi-gene assisted diagnosis of autoimmune diseases | |
| EP4031688A1 (en) | In vitro method for determining the risk of developing breast cancer in a subject | |
| CN117012390A (en) | Tubercular meningitis death risk assessment model, construction method, system and device | |
| Villaplana-Velasco et al. | Decreased retinal vascular complexity is an early biomarker of MI supported by a shared genetic control | |
| CN111816303A (en) | Machine learning-based method for predicting risk of refractory schizophrenia | |
| Kavianfar et al. | A Review of the Management of Eye Diseases Using Artificial Intelligence, Machine Learning, and Deep Learning in Conjunction with Recent Research on Eye Health Problems | |
| CN115565690A (en) | Early-stage identification method for autism spectrum disorder based on machine learning | |
| US20040143455A1 (en) | Hyper-fractionation of transmission history for medical diagnostic software | |
| Lyona et al. | Classification of retinal images in stages of diabetic retinopathy using deep learning | |
| Kumar et al. | Neural networks in medical and healthcare | |
| Stone et al. | Recommendations of the American Academy of Ophthalmology task force on genetic testing | |
| Lum et al. | Emerging role of radiogenomics in genetically triggered thoracic aortic aneurysm and dissection: a narrative review | |
| Perera et al. | Alzheimer’s disease prediction using clinical data approach |