Machine learning to identify pairwise interactions between specific IgE antibodies and their association with asthma: A cross-sectional analysis within a population-based birth cohort.
Machine learning to identify pairwise interactions between specific IgE antibodies and their association with asthma: A cross-sectional analysis within a population-based birth cohort.
复制标题
DOI：
10.1371/journal.pmed.1002691
复制
发表时间：
2018-11
期刊：
影响因子：
15.8
通讯作者：
Custovic A
中科院分区：
文献类型：
--
作者：
Fontanella S;Frainay C;Murray CS;Simpson A;Custovic A
过敏反应和哮喘之间的关系是复杂的；关于这种联系的强度的数据是相互矛盾的。我们认为，差异的产生部分是因为过敏反应可能不是一个单一的实体(如传统上所认为的)，而是几种不同类别的过敏反应的集合。我们假设，与哮喘风险增加相关的是针对个别过敏性分子(成分)的免疫球蛋白E(IgE)抗体之间的配对，而不是对“信息性”分子的IgE反应。在一项横断面分析中，461名11岁的儿童参与了基于人群的出生队列，我们使用多重阵列(免疫固相过敏芯片[ISAC])测量了112种变应原成分的血清特异性IgE反应。在213名(46.2%)对这44种活性成分中的至少一种敏感的参与者中，我们表征了对这44种活性成分(至少5%的儿童中有0.30单位的特异性免疫球蛋白E)的敏感性。我们采用了几种机器学习方法，为研究高度复杂的sIgE与哮喘的关系提供了一个强大的框架。首先，我们应用网络分析和层次聚类(HC)来探索组件特定的IGE的连通性结构，并识别组件特定敏感度的簇(组件簇)。在模型中包含的44个成分中，33个分成7个集群(C.sIgE-1-7)，其余11个组成单一集群。簇成员关系与蛋白质和/或其生物来源的结构同源性密切相关。病程相关(PR)-10蛋白簇(C.sIgE-5)中的组分是该网络的中心，并介导草(C.sIgE-4)、树(C.sIgE-6)和Profilin簇(C.sIgE-7)中的组分与尘螨(C.sIgE-1)、脂蛋白(C.sIgE-3)和花生簇(C.sIgE-2)中的组分之间的联系。然后，我们使用HC在研究参与者中确定了四个常见的“敏感性簇”：(1)多重敏感性(对所有七个组分和单一组分的多种成分的sIgE)，(2)主要的尘螨敏感性(主要对C.sIgE-1的成分有反应)，(3)主要的草和树敏感性(跨C.sIgE-4-7的多组分的sIgE)，以及(4)较低级别的敏感性。我们使用二部网络来探索成分簇、致敏簇和哮喘之间的关系，并使用基于密度的联合非参数差异相互作用网络分析和分类(JDINAC)来测试成分特异性IgE的成对相互作用是否与哮喘相关。成对交互作用的JDINAC在敏感性(0.84)和特异性(0.87)之间提供了良好的平衡，并且在预测哮喘方面优于单独使用sIgE成分的惩罚Logistic回归，曲线下面积(AUC)为0.94，而不是0.73。然后，我们推断了成对成分特异性IgE相互作用的差异网络，这表明18对成分预测哮喘。这些发现在一个独立的8岁儿童样本中得到了证实，这些儿童参加了相同的出生队列，但在11岁时没有成分解析诊断(CRD)数据。我们研究的主要局限是排除了由ISAC芯片分辨率以及过滤步骤引起的潜在重要过敏原。如果有其他组件可用，集群和网络分析可能会提供不同的解决方案。SIgE组分之间的相互作用与哮喘风险增加相关，并可能为设计哮喘诊断工具提供基础。阿德南·库斯托维奇和他的同事证明，儿童过敏原敏感度的组合可以用来预测同时诊断哮喘。过敏反应和哮喘之间的关系是复杂的。基于对全部变应原提取物的IgE反应的哮喘预测模型表现出相对较差的性能。这项研究考察了成分解析诊断(CRD)中对多种变应原成分的IgE反应之间的关系及其与哮喘的关系。在以人群为基础的出生队列中，使用多重阵列测量了儿童对112种过敏原成分的血清特异性IgE反应。研究人员应用网络分析和层次聚类(HC)来探索特定成分的IgE的连接性结构，并确定了七个特定成分敏化的簇。簇成员关系与蛋白质和/或其生物来源的结构同源性密切相关。研究小组在研究参与者中发现了四个“敏感化群”。使用二部网络探讨了成分簇、致敏簇和哮喘之间的关系。结果表明，18对变应原成分之间的相互作用预测哮喘的敏感性和特异性达到了良好的平衡。例如，对狗和猫的不同过敏原蛋白或马和屋尘螨都有IgE抗体的儿童患哮喘的风险更高。免疫球蛋白E对多种致敏蛋白的反应在功能上是相互协调和共同调节的。这个复杂网络中的成对交互作用可以预测临床表型。SIgE组分之间的相互作用与哮喘风险的增加有关，并为设计哮喘诊断工具提供了基础。
The relationship between allergic sensitisation and asthma is complex; the data about the strength of this association are conflicting. We propose that the discrepancies arise in part because allergic sensitisation may not be a single entity (as considered conventionally) but a collection of several different classes of sensitisation. We hypothesise that pairings between immunoglobulin E (IgE) antibodies to individual allergenic molecules (components), rather than IgE responses to ‘informative’ molecules, are associated with increased risk of asthma. In a cross-sectional analysis among 461 children aged 11 years participating in a population-based birth cohort, we measured serum-specific IgE responses to 112 allergen components using a multiplex array (ImmunoCAP Immuno‑Solid phase Allergy Chip [ISAC]). We characterised sensitivity to 44 active components (specific immunoglobulin E [sIgE] > 0.30 units in at least 5% of children) among the 213 (46.2%) participants sensitised to at least one of these 44 components. We adopted several machine learning methodologies that offer a powerful framework to investigate the highly complex sIgE–asthma relationship. Firstly, we applied network analysis and hierarchical clustering (HC) to explore the connectivity structure of component-specific IgEs and identify clusters of component-specific sensitisation (‘component clusters’). Of the 44 components included in the model, 33 grouped in seven clusters (C.sIgE-1–7), and the remaining 11 formed singleton clusters. Cluster membership mapped closely to the structural homology of proteins and/or their biological source. Components in the pathogenesis-related (PR)-10 proteins cluster (C.sIgE-5) were central to the network and mediated connections between components from grass (C.sIgE-4), trees (C.sIgE-6), and profilin clusters (C.sIgE-7) with those in mite (C.sIgE-1), lipocalins (C.sIgE-3), and peanut clusters (C.sIgE-2). We then used HC to identify four common ‘sensitisation clusters’ among study participants: (1) multiple sensitisation (sIgE to multiple components across all seven component clusters and singleton components), (2) predominantly dust mite sensitisation (IgE responses mainly to components from C.sIgE-1), (3) predominantly grass and tree sensitisation (sIgE to multiple components across C.sIgE-4–7), and (4) lower-grade sensitisation. We used a bipartite network to explore the relationship between component clusters, sensitisation clusters, and asthma, and the joint density-based nonparametric differential interaction network analysis and classification (JDINAC) to test whether pairwise interactions of component-specific IgEs are associated with asthma. JDINAC with pairwise interactions provided a good balance between sensitivity (0.84) and specificity (0.87), and outperformed penalised logistic regression with individual sIgE components in predicting asthma, with an area under the curve (AUC) of 0.94, compared with 0.73. We then inferred the differential network of pairwise component-specific IgE interactions, which demonstrated that 18 pairs of components predicted asthma. These findings were confirmed in an independent sample of children aged 8 years who participated in the same birth cohort but did not have component-resolved diagnostics (CRD) data at age 11 years. The main limitation of our study was the exclusion of potentially important allergens caused by both the ISAC chip resolution as well as the filtering step. Clustering and the network analyses might have provided different solutions if additional components had been available. Interactions between pairs of sIgE components are associated with increased risk of asthma and may provide the basis for designing diagnostic tools for asthma. Adnan Custovic and colleagues demonstrate that combinations of allergen sensitivities in children can be used to predict concurrent asthma diagnosis. The relationship between allergic sensitisation and asthma is complex. Asthma prediction models based on the IgE responses to the whole allergen extracts exhibit relatively poor performance. This study examines the relationship between IgE responses to multiple allergen components in component-resolved diagnostics (CRD) and their associations with asthma. Serum-specific IgE responses to 112 allergen components were measured using a multiplex array among children in a population-based birth cohort. Researchers applied network analysis and hierarchical clustering (HC) to explore the connectivity structure of component-specific IgEs and identified seven clusters of component-specific sensitisation. Cluster membership mapped closely to the structural homology of proteins and/or their biological source. HC identified four ‘sensitisation clusters’ among study participants. The relationship between component clusters, sensitisation clusters, and asthma was explored using a bipartite network. The differential network of pairwise component-specific IgE interactions was inferred, which demonstrated that interactions among 18 pairs of allergen components predicted asthma with a good balance between sensitivity and specificity. For example, children with IgE antibodies to different allergenic proteins from both dog and cat, or horse and house dust mite, are at higher risk of developing asthma. IgE responses to multiple allergenic proteins are functionally coordinated and co-regulated. Pairwise interactions within this complex network predict clinical phenotypes. Interactions between pairs of sIgE components are associated with increased risk of asthma and provide the basis for designing diagnostic tools for asthma.
登录
查看更多内容
影响因子：
3.7
作者：
Lødrup Carlsen KC;Roll S;Carlsen KH;Mowinckel P;Wijga AH;Brunekreef B;Torrent M;Roberts G;Arshad SH;Kull I;Krämer U;von Berg A;Eller E;Høst A;Kuehni C;Spycher B;Sunyer J;Chen CM;Reich A;Asarnoj A;Puig C;Herbarth O;Mahachie John JM;Van Steen K;Willich SN;Wahn U;Lau S;Keil T;GALEN WP 1.5 ‘Birth Cohorts’ working group
通讯作者：
GALEN WP 1.5 ‘Birth Cohorts’ working group
DOI：
10.1016/s2352-4642(17)30008-1
发表时间：
2017-10
期刊：
The Lancet. Child & adolescent health
影响因子：
--
作者：
Murray C;Foden P;Lowe L;Durrington H;Custovic A;Simpson A
通讯作者：
Simpson A
影响因子：
4.4
作者：
Bjerg, Anders;Winberg, Anna;Ronmark, Eva
通讯作者：
Ronmark, Eva
影响因子：
39.2
作者：
Collins, Gary S.;Reitsma, Johannes B.;Moons, Karel G. M.
通讯作者：
Moons, Karel G. M.
影响因子：
12.4
作者：
Lazic, N.;Roberts, G.;Simpson, A.
通讯作者：
Simpson, A.