Current state and prospects of artificial intelligence in allergy.

The field of medicine is witnessing an exponential growth of interest in artificial intelligence (AI), which enables new research questions and the analysis of larger and new types of data. Nevertheless, applications that go beyond proof of concepts and deliver clinical value remain rare, especially in the field of allergy. This narrative review provides a fundamental understanding of the core concepts of AI and critically discusses its limitations and open challenges, such as data availability and bias, along with potential directions to surmount them. We provide a conceptual framework to structure AI applications within this field and discuss forefront case examples. Most of these applications of AI and machine learning in allergy concern supervised learning and unsupervised clustering, with a strong emphasis on diagnosis and subtyping. A perspective is shared on guidelines for good AI practice to guide readers in applying it effectively and safely, along with prospects of field advancement and initiatives to increase clinical impact. We anticipate that AI can further deepen our knowledge of disease mechanisms and contribute to precision medicine in allergy.

Growth disrupting mutations in epigenetic regulatory molecules are associated with abnormalities of epigenetic aging.

Germline mutations in fundamental epigenetic regulatory molecules including DNA methyltransferase 3 alpha (DNMT3A) are commonly associated with growth disorders, whereas somatic mutations are often associated with malignancy. We profiled genome-wide DNA methylation patterns in DNMT3A c.2312G > A; p.(Arg771Gln) carriers in a large Amish sibship with Tatton-Brown-Rahman syndrome (TBRS), their mosaic father, and 15 TBRS patients with distinct pathogenic de novo DNMT3A variants. This defined widespread DNA hypomethylation at specific genomic sites enriched at locations annotated as genes involved in morphogenesis, development, differentiation, and malignancy predisposition pathways. TBRS patients also displayed highly accelerated DNA methylation aging. These findings were most marked in a carrier of the AML-associated driver mutation p.Arg882Cys. Our studies additionally defined phenotype-related accelerated and decelerated epigenetic aging in two histone methyltransferase disorders: NSD1 Sotos syndrome overgrowth disorder and KMT2D Kabuki syndrome growth impairment. Together, our findings provide fundamental new insights into aberrant epigenetic mechanisms, the role of epigenetic machinery maintenance, and determinants of biological aging in these growth disorders.

Preconceptional smoking alters spermatozoal miRNAs of murine fathers and affects offspring's body weight.

BACKGROUND: Active smoking has been reported among 7% of teenagers worldwide, with ages ranging from 13 to 15 years. An epidemiological study suggested that preconceptional paternal smoking is associated with adolescent obesity in boys. We developed a murine adolescent smoking model before conception to investigate the paternal molecular causes of changes in offspring's phenotype. METHOD: Male and female C57BL/6J mice were exposed to increasing doses of mainstream cigarette smoke (CS) from onset of puberty for 6 weeks and mated with room air (RA) controls. RESULTS: Thirteen miRNAs were upregulated and 32 downregulated in the spermatozoa of CS-exposed fathers, while there were no significant differences in the count and morphological integrity of spermatozoa, as well as the proliferation of spermatogonia between CS- and RA-exposed fathers. Offspring from preconceptional CS-exposed mothers had lower body weights (p = 0.007). Moreover, data from offspring from CS-exposed fathers suggested a potential increase in body weight (p = 0.062). CONCLUSION: We showed that preconceptional paternal CS exposure regulates spermatozoal miRNAs, and possibly influences the body weight of F1 progeny in early life. The regulated miRNAs may modulate transmittable epigenetic changes to offspring, thus influence the development of respiratory- and metabolic-related diseases such as obesity, a mechanism that warrants further studies for elaborate explanations.

Gaussian Bayesian network comparisons with graph ordering unknown.

A Bayesian approach is proposed that unifies Gaussian Bayesian network constructions and comparisons between two networks (identical or differential) for data with graph ordering unknown. When sampling graph ordering, to escape from local maximums, an adjusted single queue equi-energy algorithm is applied. The conditional posterior probability mass function for network differentiation is derived and its asymptotic proposition is theoretically assessed. Simulations are used to demonstrate the approach and compare with existing methods. Based on epigenetic data at a set of DNA methylation sites (CpG sites), the proposed approach is further examined on its ability to detect network differentiations. Findings from theoretical assessment, simulations, and real data applications support the efficacy and efficiency of the proposed method for network comparisons.

Changes of DNA methylation are associated with changes in lung function during adolescence.

BACKGROUND: Adolescence is a significant period for the gender-dependent development of lung function. Prior studies have shown that DNA methylation (DNA-M) is associated with lung function and DNA-M at some cytosine-phosphate-guanine dinucleotide sites (CpGs) changes over time. This study examined whether changes of DNA-M at lung-function-related CpGs are associated with changes in lung function during adolescence for each gender, and if so, the biological significance of the detected CpGs. METHODS: Genome-scale DNA-M was measured in peripheral blood samples at ages 10 (n = 330) and 18 years (n = 476) from the Isle of Wight (IOW) birth cohort in United Kingdom, using Illumina Infinium arrays (450 K and EPIC). Spirometry was conducted at both ages. A training and testing method was used to screen 402,714 CpGs for their potential associations with lung function. Linear regressions were applied to assess the association of changes in lung function with changes of DNA-M at those CpGs potentially related to lung function. Adolescence-related and personal and family-related confounders were included in the model. The analyses were stratified by gender. Multiple testing was adjusted by controlling false discovery rate of 0.05. Findings were further examined in two independent birth cohorts, the Avon Longitudinal Study of Children and Parents (ALSPAC) and the Children, Allergy, Milieu, Stockholm, Epidemiology (BAMSE) cohort. Pathway analyses were performed on genes to which the identified CpGs were mapped. RESULTS: For females, 42 CpGs showed statistically significant associations with change in FEV1/FVC, but none for change in FEV1 or FVC. No CpGs were identified for males. In replication analyses, 16 and 21 of the 42 CpGs showed the same direction of associations among the females in the ALSPAC and BAMSE cohorts, respectively, with 11 CpGs overlapping across all the three cohorts. Through pathway analyses, significant biological processes were identified that have previously been related to lung function development. CONCLUSIONS: The detected 11 CpGs in all three cohorts have the potential to serve as the candidate epigenetic markers for changes in lung function during adolescence in females.

Prediction models for childhood asthma: A systematic review.

BACKGROUND: The inability to objectively diagnose childhood asthma before age five often results in both under-treatment and over-treatment of asthma in preschool children. Prediction tools for estimating a child's risk of developing asthma by school-age could assist physicians in early asthma care for preschool children. This review aimed to systematically identify and critically appraise studies which either developed novel or updated existing prediction models for predicting school-age asthma. METHODS: Three databases (MEDLINE, Embase and Web of Science Core Collection) were searched up to July 2019 to identify studies utilizing information from children ≤5 years of age to predict asthma in school-age children (6-13 years). Validation studies were evaluated as a secondary objective. RESULTS: Twenty-four studies describing the development of 26 predictive models published between 2000 and 2019 were identified. Models were either regression-based (n = 21) or utilized machine learning approaches (n = 5). Nine studies conducted validations of six regression-based models. Fifteen (out of 21) models required additional clinical tests. Overall model performance, assessed by area under the receiver operating curve (AUC), ranged between 0.66 and 0.87. Models demonstrated moderate ability to either rule in or rule out asthma development, but not both. Where external validation was performed, models demonstrated modest generalizability (AUC range: 0.62-0.83). CONCLUSION: Existing prediction models demonstrated moderate predictive performance, often with modest generalizability when independently validated. Limitations of traditional methods have shown to impair predictive accuracy and resolution. Exploration of novel methods such as machine learning approaches may address these limitations for future school-age asthma prediction.

Role of DNA methylation in the association of lung function with body mass index: a two-step epigenetic Mendelian randomisation study.

BACKGROUND: Low lung function has been associated with increased body mass index (BMI). The aim of this study was to investigate whether the effect of BMI on lung function is mediated by DNA methylation. METHODS: We used individual data from 285,495 participants in four population-based cohorts: the European Community Respiratory Health Survey, the Northern Finland Birth Cohort 1966, the Swiss Study on Air Pollution and Lung Disease in Adults, and the UK Biobank. We carried out Mendelian randomisation (MR) analyses in two steps using a two-sample approach with SNPs as instrumental variables (IVs) in each step. In step 1 MR, we estimated the causal effect of BMI on peripheral blood DNA methylation (measured at genome-wide level) using 95 BMI-associated SNPs as IVs. In step 2 MR, we estimated the causal effect of DNA methylation on FEV1, FVC, and FEV1/FVC using two SNPs acting as methQTLs occurring close (in cis) to CpGs identified in the first step. These analyses were conducted after exclusion of weak IVs (F statistic < 10) and MR estimates were derived using the Wald ratio, with standard error from the delta method. Individuals whose data were used in step 1 were not included in step 2. RESULTS: In step 1, we found that BMI might have a small causal effect on DNA methylation levels (less than 1% change in methylation per 1 kg/m2 increase in BMI) at two CpGs (cg09046979 and cg12580248). In step 2, we found no evidence of a causal effect of DNA methylation at cg09046979 on lung function. We could not estimate the causal effect of DNA methylation at cg12580248 on lung function as we could not find publicly available data on the association of this CpG with SNPs. CONCLUSIONS: To our knowledge, this is the first paper to report the use of a two-step MR approach to assess the role of DNA methylation in mediating the effect of a non-genetic factor on lung function. Our findings do not support a mediating effect of DNA methylation in the association of lung function with BMI.

Epigenome-wide meta-analysis of DNA methylation and childhood asthma.

BACKGROUND: Epigenetic mechanisms, including methylation, can contribute to childhood asthma. Identifying DNA methylation profiles in asthmatic patients can inform disease pathogenesis. OBJECTIVE: We sought to identify differential DNA methylation in newborns and children related to childhood asthma. METHODS: Within the Pregnancy And Childhood Epigenetics consortium, we performed epigenome-wide meta-analyses of school-age asthma in relation to CpG methylation (Illumina450K) in blood measured either in newborns, in prospective analyses, or cross-sectionally in school-aged children. We also identified differentially methylated regions. RESULTS: In newborns (8 cohorts, 668 cases), 9 CpGs (and 35 regions) were differentially methylated (epigenome-wide significance, false discovery rate < 0.05) in relation to asthma development. In a cross-sectional meta-analysis of asthma and methylation in children (9 cohorts, 631 cases), we identified 179 CpGs (false discovery rate < 0.05) and 36 differentially methylated regions. In replication studies of methylation in other tissues, most of the 179 CpGs discovered in blood replicated, despite smaller sample sizes, in studies of nasal respiratory epithelium or eosinophils. Pathway analyses highlighted enrichment for asthma-relevant immune processes and overlap in pathways enriched both in newborns and children. Gene expression correlated with methylation at most loci. Functional annotation supports a regulatory effect on gene expression at many asthma-associated CpGs. Several implicated genes are targets for approved or experimental drugs, including IL5RA and KCNH2. CONCLUSION: Novel loci differentially methylated in newborns represent potential biomarkers of risk of asthma by school age. Cross-sectional associations in children can reflect both risk for and effects of disease. Asthma-related differential methylation in blood in children was substantially replicated in eosinophils and respiratory epithelium.

Maternal BMI at the start of pregnancy and offspring epigenome-wide DNA methylation: findings from the pregnancy and childhood epigenetics (PACE) consortium.

Pre-pregnancy maternal obesity is associated with adverse offspring outcomes at birth and later in life. Individual studies have shown that epigenetic modifications such as DNA methylation could contribute. Within the Pregnancy and Childhood Epigenetics (PACE) Consortium, we meta-analysed the association between pre-pregnancy maternal BMI and methylation at over 450,000 sites in newborn blood DNA, across 19 cohorts (9,340 mother-newborn pairs). We attempted to infer causality by comparing the effects of maternal versus paternal BMI and incorporating genetic variation. In four additional cohorts (1,817 mother-child pairs), we meta-analysed the association between maternal BMI at the start of pregnancy and blood methylation in adolescents. In newborns, maternal BMI was associated with small (<0.2% per BMI unit (1 kg/m2), P < 1.06 × 10-7) methylation variation at 9,044 sites throughout the genome. Adjustment for estimated cell proportions greatly attenuated the number of significant CpGs to 104, including 86 sites common to the unadjusted model. At 72/86 sites, the direction of the association was the same in newborns and adolescents, suggesting persistence of signals. However, we found evidence for acausal intrauterine effect of maternal BMI on newborn methylation at just 8/86 sites. In conclusion, this well-powered analysis identified robust associations between maternal adiposity and variations in newborn blood DNA methylation, but these small effects may be better explained by genetic or lifestyle factors than a causal intrauterine mechanism. This highlights the need for large-scale collaborative approaches and the application of causal inference techniques in epigenetic epidemiology.

Epigenome-wide association study of lung function level and its change.

Previous reports link differential DNA methylation (DNAme) to environmental exposures that are associated with lung function. Direct evidence on lung function DNAme is, however, limited. We undertook an agnostic epigenome-wide association study (EWAS) on pre-bronchodilation lung function and its change in adults.In a discovery-replication EWAS design, DNAme in blood and spirometry were measured twice, 6-15 years apart, in the same participants of three adult population-based discovery cohorts (n=2043). Associated DNAme markers (p<5×10-7) were tested in seven replication cohorts (adult: n=3327; childhood: n=420). Technical bias-adjusted residuals of a regression of the normalised absolute ß-values on control probe-derived principle components were regressed on level and change of forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC) and their ratio (FEV1/FVC) in the covariate-adjusted discovery EWAS. Inverse-variance-weighted meta-analyses were performed on results from discovery and replication samples in all participants and never-smokers.EWAS signals were enriched for smoking-related DNAme. We replicated 57 lung function DNAme markers in adult, but not childhood samples, all previously associated with smoking. Markers not previously associated with smoking failed replication. cg05575921 (AHRR (aryl hydrocarbon receptor repressor)) showed the statistically most significant association with cross-sectional lung function (FEV1/FVC: pdiscovery=3.96×10-21 and pcombined=7.22×10-50). A score combining 10 DNAme markers previously reported to mediate the effect of smoking on lung function was associated with lung function (FEV1/FVC: p=2.65×10-20).Our results reveal that lung function-associated methylation signals in adults are predominantly smoking related, and possibly of clinical utility in identifying poor lung function and accelerated decline. Larger studies with more repeat time-points are needed to identify lung function DNAme in never-smokers and in children.

DNA methylation and allergic sensitizations: A genome-scale longitudinal study during adolescence.

BACKGROUND: The presence of allergic sensitization has a major influence on the development and course of common childhood conditions such as asthma and rhinitis. The etiology of allergic sensitization is poorly understood, and its underlying biological mechanisms are not well established. Several studies showed that DNA methylation (DNAm) at some CpGs is associated with allergic sensitization. However, no studies have focused on the critical adolescence period. METHODS: We assessed the association of pre- and postadolescence genome-wide DNAm with allergic sensitization against indoor, outdoor and food allergens, using linear mixed models. We hypothesized that DNAm is associated with sensitization in general, and with poly-sensitization status, and these associations are age- and gender-specific. We tested these hypotheses in the IoW cohort (n = 376) and examined the findings in the BAMSE cohort (n = 267). RESULTS: Via linear mixed models, we identified 35 CpGs in IoW associated with allergic sensitization (at false discovery rate of 0.05), of which 33 were available in BAMSE and replicated with respect to the direction of associations with allergic sensitization. At the 35 CpGs except for cg19210306 on C13orf27, a reduction in methylation among atopic subjects was observed, most notably for cg21220721 and cg11699125 (ACOT7). DNAm at cg10159529 was strongly correlated with expression of IL5RA in peripheral blood (P-value = 6.76 × 10-20 ). Three CpGs (cg14121142, cg23842695, and cg26496795) were identified in IoW with age-specific association between DNAm and allergic sensitization. CONCLUSION: In adolescence, the status of allergic sensitization was associated with DNAm differentiation and at some CpGs the association is likely to be age-specific.

Maternal variants in NLRP and other maternal effect proteins are associated with multilocus imprinting disturbance in offspring.

BACKGROUND: Genomic imprinting results from the resistance of germline epigenetic marks to reprogramming in the early embryo for a small number of mammalian genes. Genetic, epigenetic or environmental insults that prevent imprints from evading reprogramming may result in imprinting disorders, which impact growth, development, behaviour and metabolism. We aimed to identify genetic defects causing imprinting disorders by whole-exome sequencing in families with one or more members affected by multilocus imprinting disturbance. METHODS: Whole-exome sequencing was performed in 38 pedigrees where probands had multilocus imprinting disturbance, in five of whom maternal variants in NLRP5 have previously been found. RESULTS: We now report 15 further pedigrees in which offspring had disturbance of imprinting, while their mothers had rare, predicted-deleterious variants in maternal effect genes, including NLRP2, NLRP7 and PADI6. As well as clinical features of well-recognised imprinting disorders, some offspring had additional features including developmental delay, behavioural problems and discordant monozygotic twinning, while some mothers had reproductive problems including pregnancy loss. CONCLUSION: The identification of 20 putative maternal effect variants in 38 families affected by multilocus imprinting disorders adds to the evidence that maternal genetic factors affect oocyte fitness and thus offspring development. Testing for maternal-effect genetic variants should be considered in families affected by atypical imprinting disorders.

Transgenerational and intergenerational epigenetic inheritance in allergic diseases.

It has become clear that early life (including in utero exposures) is a key window of vulnerability during which environmental exposures can alter developmental trajectories and initiate allergic disease development. However, recent evidence suggests that there might be additional windows of vulnerability to environmental exposures in the parental generation before conception or even in previous generations. There is evidence suggesting that information of prior exposures can be transferred across generations, and experimental animal models suggest that such transmission can be conveyed through epigenetic mechanisms. Although the molecular mechanisms of intergenerational and transgenerationational epigenetic transmission have yet to be determined, the realization that environment before conception can alter the risks of allergic diseases has profound implications for the development of public health interventions to prevent disease. Future research in both experimental models and in multigenerational human cohorts is needed to better understand the role of intergenerational and transgenerational effects in patients with asthma and allergic disease. This will provide the knowledge basis for a new approach to efficient intervention strategies aimed at reducing the major public health challenge of these conditions.

SERPINA1 methylation and lung function in tobacco-smoke exposed European children and adults: a meta-analysis of ALEC population-based cohorts.

BACKGROUND: The pathophysiological role of SERPINA1 in respiratory health may be more strongly determined by the regulation of its expression than by common genetic variants. A family based study of predominantly smoking adults found methylation at two Cytosine-phosphate-Guanine sites (CpGs) in SERPINA1 gene to be associated with chronic obstructive pulmonary disease risk. The objective of this study was to confirm the association of lung function with SERPINA1 methylation in general population samples by testing a comprehensive set of CpGs in the SERPINA gene cluster. We considered lung function level and decline in adult smokers from three European population-based cohorts and lung function level and growth in tobacco-smoke exposed children from a birth cohort. METHODS: DNA methylation using Illumina Infinium Human Methylation 450 k and EPIC beadchips and lung function were measured at two time points in 1076 SAPALDIA, ECRHS and NFBC adult cohort participants and 259 ALSPAC children. Associations of methylation at 119 CpG sites in the SERPINA gene cluster (PP4R4-SERPINA13P) with lung functions and circulating alpha-1-antitripsin (AAT) were assessed using multivariable cross-sectional and longitudinal regression models. RESULTS: Methylation at cg08257009 in the SERPINA gene cluster, located 32 kb downstream of SERPINA1, not annotated to a gene, was associated with FEV1/FVC at the Bonferroni corrected level in adults, but not in children. None of the methylation signals in the SERPINA1 gene showed associations with lung function after correcting for multiple testing. CONCLUSIONS: The results do not support a role of SERPINA1 gene methylation as determinant of lung function across the life course in the tobacco smoke exposed general population exposed.

Systematic review of lung function and COPD with peripheral blood DNA methylation in population based studies.

BACKGROUND: Epigenetic variations in peripheral blood have potential as biomarkers for disease. This systematic review assesses the association of lung function and chronic obstructive pulmonary disease (COPD) with DNA methylation profiles in peripheral blood from population-based studies. METHODS: Online databases Medline, Embase, and Web of Science were searched. Google Scholar was searched to identify grey literature. After removing duplicate articles, 1155 articles were independently screened by two investigators. Peer reviewed reports on population-based studies that examined peripheral blood DNA methylation in participants with measured lung function (FEV1, FEV1/FVC ratio) or known COPD status were selected for full-text review. Six articles were suitable for inclusion. Information regarding study characteristics, designs, methodologies and conclusions was extracted. A narrative synthesis was performed based on published results. RESULTS: Three of the six articles assessed the association of COPD with DNA methylation, and two of these also included associations with lung function. Overall, five reports examined the association of lung function with DNA methylation profiles. Five of the six articles reported 'significant' results. However, no consistent CpG sites were identified across studies for COPD status or lung function values. CONCLUSIONS: DNA methylation patterns in peripheral blood from individuals with reduced lung function or COPD may be different to those in people with normal lung function. However, this systematic review did not find any consistent associations of lung function or COPD with differentially methylated CpG sites. Large studies with a longitudinal design to address reverse causality may prove a more fruitful area of research. TRIAL REGISTRATION: PROSPERO 2016: CRD42016037352 .

Congenital imprinting disorders: Application of multilocus and high throughput methods to decipher new pathomechanisms and improve their management.

Imprinting disorders (IDs) are a group of congenital diseases affecting growth, development and metabolism. They are caused by changes in the allele-specific regulation ("epigenetic mutation") or in the genomic sequence ("genetic mutation") of imprinted genes. Currently molecular tests in ID patients are generally restricted to single loci classically associated with the disease, but this approach limits diagnostic yield, because of the molecular and clinical heterogeneity between IDs. From the technical point of view, these limitations are aggravated by the lack of standardization in testing methodology, in the DNA sequences tested, and in clinical inclusion criteria prompting testing. However, an increasing number of new studies show that these problems can be addressed by the use of new tests targeting multiple loci and/or a total exome and genome analysis. The rapid development of efficient and high-throughput molecular techniques and their applications in research and diagnostics in the last decade have led to an impressive increase of knowledge on IDs and their basic pathomechanisms. In combination with the improvement of data recording and documentation, the diagnostic strategies are increasingly based on standardized protocols, and thereby provide the backbone for directed counseling, more personalized management, and new therapeutic approaches.

Genome-wide DNA methylation analysis of patients with imprinting disorders identifies differentially methylated regions associated with novel candidate imprinted genes.

BACKGROUND: Genomic imprinting is allelic restriction of gene expression potential depending on parent of origin, maintained by epigenetic mechanisms including parent of origin-specific DNA methylation. Among approximately 70 known imprinted genes are some causing disorders affecting growth, metabolism and cancer predisposition. Some imprinting disorder patients have hypomethylation of several imprinted loci (HIL) throughout the genome and may have atypically severe clinical features. Here we used array analysis in HIL patients to define patterns of aberrant methylation throughout the genome. DESIGN: We developed a novel informatic pipeline capable of small sample number analysis, and profiled 10 HIL patients with two clinical presentations (Beckwith-Wiedemann syndrome and neonatal diabetes) using the Illumina Infinium Human Methylation450 BeadChip array to identify candidate imprinted regions. We used robust statistical criteria to quantify DNA methylation. RESULTS: We detected hypomethylation at known imprinted loci, and 25 further candidate imprinted regions (nine shared between patient groups) including one in the Down syndrome critical region (WRB) and another previously associated with bipolar disorder (PPIEL). Targeted analysis of three candidate regions (NHP2L1, WRB and PPIEL) showed allelic expression, methylation patterns consistent with allelic maternal methylation and frequent hypomethylation among an additional cohort of HIL patients, including six with Silver-Russell syndrome presentations and one with pseudohypoparathyroidism 1B. CONCLUSIONS: This study identified novel candidate imprinted genes, revealed remarkable epigenetic convergence among clinically divergent patients, and highlights the potential of epigenomic profiling to expand our understanding of the normal methylome and its disruption in human disease.