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

Abstract

Background: 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.

Methods and findings: 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.

Conclusions: Interactions between pairs of sIgE components are associated with increased risk of asthma and may provide the basis for designing diagnostic tools for asthma.

Fig 1. Patterns of sensitisations stratified by asthma status.

Participants are represented in columns and sIgE components in rows. A black square indicates that a participant has a sIgE>0.30 to a particular allergen component. sIgE, specific immunoglobulin E.

Fig 2. Component-specific IgE network and hierarchical cluster reveal connectivity structure in sIgE.

The network consists of a set of nodes, joined in pairs by lines or edges. Colours represent cluster memberships and node diameter is proportional to the scaled connectivity of each sIgE, while edge colour and width represent the strength of connection between pairs of sIgE components. IgE, immunoglobulin E; sIgE, specific immunoglobulin E.

Fig 3. Patterns of IgE responses to allergen components for individual participants.

Rows represent sIgEs, while columns indicate children. Colours represent sensitisation clusters’ membership. Squares are coloured if and only if a child has a positive response, <0.30 to a particular sIgE. IgE, immunoglobulin E; sIgE, specific immunoglobulin E.

Fig 4. Bipartite network to uncover the relationship between sensitisation clusters and asthma, and the connectivity with component-specific IgEs and component clusters.

In the bipartite network, nodes represent one or more types of entities, and edges between the nodes represent a specific relationship between the entities. Here, pie charts represent individuals aggregated according to sensitisation cluster membership and asthma status. Red indicates children with asthma, while blue indicates no asthma. Squares represent sIgE allergens and colours represent cluster membership. Edges show whether a subject has a positive response to a particular c-sIgE. MDS layout was used to infer the network. HDM, house dust mite; IgE, immunoglobulin E; MDS, multidimensional scaling; sIgE, specific immunoglobulin E.

Fig 5. ROC curves for JDINAC and penalised logistic regression.

The curves were obtained through the prediction averaging procedure on 50 independent repetitions combined with of 10-fold cross validation. JDINAC, joint density-based nonparametric differential interaction network analysis and classification; ROC, receiver operating characteristic.

Fig 6. Differential pairwise component-specific IgE interactions in asthma estimated by JDINAC.

The presence of an edge presented in the differential network means that the dependency of corresponding pair sIgEs is different between those who have asthma and those who do not have asthma. The edge colour indicates the direction of association. Red: interaction linked to asthma presence; green: interaction linked to reduced risk of asthma. Edge width is proportional to differential weight. Only pairs of sIgEs that were significantly associated to the risk of asthma in 25% of the validation runs were included in the network. IgE, immunoglobulin E; JDINAC, joint density-based nonparametric differential interaction network analysis and classification; sIgE, specific immunoglobulin E.

Fig 7. Differential pairwise component-specific IgE interactions in asthma estimated by JDINAC on the 8-year-old children data set.

IgE, immunoglobulin E; JDINAC, joint density-based nonparametric differential interaction network analysis and classification; sIgE, specific immunoglobulin E.