Maturation and application of phenome-wide association studies.

In the past 10 years since its introduction, phenome-wide association studies (PheWAS) have uncovered novel genotype-phenotype relationships. Along the way, PheWAS have evolved in many aspects as a study design with the expanded availability of large data repositories with genome-wide data linked to detailed phenotypic data. Advancement in methods, including algorithms, software, and publicly available integrated resources, makes it feasible to more fully realize the potential of PheWAS, overcoming the previous computational and analytical limitations. We review here the most recent improvements and notable applications of PheWAS since the second half of the decade from its inception. We also note the challenges that remain embedded along the entire PheWAS analytical pipeline that necessitate further development of tools and resources to further advance the understanding of the complex genetic architecture underlying human diseases and traits.

Network enrichment significance testing in brain-phenotype association studies.

Functional networks often guide our interpretation of spatial maps of brain-phenotype associations. However, methods for assessing enrichment of associations within networks of interest have varied in terms of both scientific rigor and underlying assumptions. While some approaches have relied on subjective interpretations, others have made unrealistic assumptions about spatial properties of imaging data, leading to inflated false positive rates. We seek to address this gap in existing methodology by borrowing insight from a method widely used in genetics research for testing enrichment of associations between a set of genes and a phenotype of interest. We propose network enrichment significance testing (NEST), a flexible framework for testing the specificity of brain-phenotype associations to functional networks or other sub-regions of the brain. We apply NEST to study enrichment of associations with structural and functional brain imaging data from a large-scale neurodevelopmental cohort study.

How structural variants shape avian phenotypes: Lessons from model systems.

Despite receiving significant recent attention, the relevance of structural variation (SV) in driving phenotypic diversity remains understudied, although recent advances in long-read sequencing, bioinformatics and pangenomic approaches have enhanced SV detection. We review the role of SVs in shaping phenotypes in avian model systems, and identify some general patterns in SV type, length and their associated traits. We found that most of the avian SVs so far identified are short indels in chickens, which are frequently associated with changes in body weight and plumage colouration. Overall, we found that relatively short SVs are more frequently detected, likely due to a combination of their prevalence compared to large SVs, and a detection bias, stemming primarily from the widespread use of short-read sequencing and associated analytical methods. SVs most commonly involve non-coding regions, especially introns, and when patterns of inheritance were reported, SVs associated primarily with dominant discrete traits. We summarise several examples of phenotypic convergence across different species, mediated by different SVs in the same or different genes and different types of changes in the same gene that can lead to various phenotypes. Complex rearrangements and supergenes, which can simultaneously affect and link several genes, tend to have pleiotropic phenotypic effects. Additionally, SVs commonly co-occur with single-nucleotide polymorphisms, highlighting the need to consider all types of genetic changes to understand the basis of phenotypic traits. We end by summarising expectations for when long-read technologies become commonly implemented in non-model birds, likely leading to an increase in SV discovery and characterisation. The growing interest in this subject suggests an increase in our understanding of the phenotypic effects of SVs in upcoming years.

Clinical and molecular characteristics of hemophilia A affected individuals and carriers: A 24 years experience from three centers.

Hemophilia A is a rare bleeding disorder with variable expressivity and allelic heterogeneity. Despite the advancement of prenatal diagnostics and molecular studies, the number of studies reviewing the reproductive choices of hemophilia A carriers and affected individuals remains limited. Through this retrospective review, we hope to gain a deeper understanding of hemophilia A-affected individuals' clinical and molecular characteristics, as well as the reproductive choices of the at-risk couples. A total of 122 individuals harboring likely causative F8 gene alterations from 64 apparently unrelated families attending three centers between 3/2000 and 3/2023 were included in this study. Their clinical and molecular findings as well as reproductive choices were gathered in a clinical setting and verified through the electronic medical record database of the public health system. Forty-seven affected males and 75 female heterozygous carriers were included in the analysis. Among 64 apparently unrelated families, 36 distinct pathogenic/likely pathogenic variants were identified, of which 30.6% (11/36) of variants were novel. While the majority of clinical findings and genotype-phenotype correlations appear to be in accordance with existing literature, female carriers who had no fertility intention were significantly more likely to have affected sons than those who had fertility intention (5/19 vs. 4/5; p = 0.047). Through this retrospective review, we summarized the clinical and molecular characteristics of 122 individuals harboring pathogenic/likely pathogenic F8 variants, as well as their fertility intentions and reproductive outcomes. Further studies are required to look into the considerations involved in reproductive decision-making.

Genotype-phenotype associations in 1018 individuals with SCN1A-related epilepsies.

OBJECTIVE: SCN1A variants are associated with epilepsy syndromes ranging from mild genetic epilepsy with febrile seizures plus (GEFS+) to severe Dravet syndrome (DS). Many variants are de novo, making early phenotype prediction difficult, and genotype-phenotype associations remain poorly understood. METHODS: We assessed data from a retrospective cohort of 1018 individuals with SCN1A-related epilepsies. We explored relationships between variant characteristics (position, in silico prediction scores: Combined Annotation Dependent Depletion (CADD), Rare Exome Variant Ensemble Learner (REVEL), SCN1A genetic score), seizure characteristics, and epilepsy phenotype. RESULTS: DS had earlier seizure onset than other GEFS+ phenotypes (5.3 vs. 12.0 months, p < .001). In silico variant scores were higher in DS versus GEFS+ (p < .001). Patients with missense variants in functionally important regions (conserved N-terminus, S4-S6) exhibited earlier seizure onset (6.0 vs. 7.0 months, p = .003) and were more likely to have DS (280/340); those with missense variants in nonconserved regions had later onset (10.0 vs. 7.0 months, p = .036) and were more likely to have GEFS+ (15/29, χ2 = 19.16, p < .001). A minority of protein-truncating variants were associated with GEFS+ (10/393) and more likely to be located in the proximal first and last exon coding regions than elsewhere in the gene (9.7% vs. 1.0%, p < .001). Carriers of the same missense variant exhibited less variability in age at seizure onset compared with carriers of different missense variants for both DS (1.9 vs. 2.9 months, p = .001) and GEFS+ (8.0 vs. 11.0 months, p = .043). Status epilepticus as presenting seizure type is a highly specific (95.2%) but nonsensitive (32.7%) feature of DS. SIGNIFICANCE: Understanding genotype-phenotype associations in SCN1A-related epilepsies is critical for early diagnosis and management. We demonstrate an earlier disease onset in patients with missense variants in important functional regions, the occurrence of GEFS+ truncating variants, and the value of in silico prediction scores. Status epilepticus as initial seizure type is a highly specific, but not sensitive, early feature of DS.

Functional genotype-phenotype associations in recessive dystrophic epidermolysis bullosa.

BACKGROUND: Genotype-phenotype associations in recessive dystrophic epidermolysis bullosa (RDEB) have been difficult to elucidate. OBJECTIVE: To investigate RDEB genotype-phenotype associations and explore a functional approach to genotype classification. METHODS: Clinical examination and genetic testing of RDEB subjects, including assessment of clinical disease by RDEB subtype and extent of blistering. Genotypes were evaluated according to each variant's effect on type VII collagen function per updated literature and subsequently categorized by degree of impact on VII collagen function as low-impact (splice/missense, missense/missense), medium-impact (premature termination codon [PTC]/missense, splice/splice), and high-impact (PTC/PTC, PTC/splice). Genotype-phenotype associations were investigated using Kruskal-Wallis and Fisher's exact tests, and age-adjusted regressions. RESULTS: Eighty-three participants were included. High-impact variants were associated with worse RDEB subtype and clinical disease, including increased prevalence of generalized blistering (55.6% for low-impact vs 72.7% medium-impact vs 90.4% high-impact variants, P = .002). In age-adjusted regressions, participants with high-impact variants had 40.8-fold greater odds of squamous cell carcinoma compared to low-impact variants (P = .02), and 5.7-fold greater odds of death compared to medium-impact variants (P = .05). LIMITATIONS: Cross-sectional design. CONCLUSION: Functional genotype categories may stratify RDEB severity; high-impact variants correlated with worse clinical outcomes. Further validation in larger cohorts is needed.

Multimarker omnibus tests by leveraging individual marker summary statistics from large biobanks.

As biobanks become increasingly popular, access to genotypic and phenotypic data continues to increase in the form of precomputed summary statistics (PCSS). Widespread accessibility of PCSS alleviates many issues related to biobank data, including that of data privacy and confidentiality, as well as high computational costs. However, questions remain about how to maximally leverage PCSS for downstream statistical analyses. Here we present a novel method for testing the association of an arbitrary number of single nucleotide variants (SNVs) on a linear combination of phenotypes after adjusting for covariates for common multimarker tests (e.g., SKAT, SKAT-O) without access to individual patient-level data (IPD). We validate exact formulas for each method, and demonstrate their accuracy through simulation studies and an application to fatty acid phenotypic data from the Framingham Heart Study.

The structure of associations: Method insights from analyzing 28 clinical isolates of Cryptococcus neoformans.

Clinical isolates of a fungal pathogen from a single region or country often exhibit structural clonality or phylogenetic clustering at the sequence or MLST level; such population structure can persist also in larger samples. In efforts to improve causal understanding of pathogenesis at the molecular level, genome-wide association screening methods initially designed for other kingdoms have been applied to fungi. The example of a Colombian dataset of 28 clinical Cryptococcus neoformans VNI isolates indicates where the output from standard pipelines may need to be analyzed in new ways in order to efficiently extract hypotheses for experiments from fungal genotype-phenotype data.

Collections of clinical isolates of a human fungal pathogen can consist of clusters of genetically similar isolates. Such clustering complicates the screening for genetic associations with clinically relevant traits. We propose new methods, illustrating them for the fungus causing cryptococcosis.

Genome Alert!: A standardized procedure for genomic variant reinterpretation and automated gene-phenotype reassessment in clinical routine.

PURPOSE: Retrospective interpretation of sequenced data in light of the current literature is a major concern of the field. Such reinterpretation is manual and both human resources and variable operating procedures are the main bottlenecks. METHODS: Genome Alert! method automatically reports changes with potential clinical significance in variant classification between releases of the ClinVar database. Using ClinVar submissions across time, this method assigns validity category to gene-disease associations. RESULTS: Between July 2017 and December 2019, the retrospective analysis of ClinVar submissions revealed a monthly median of 1247 changes in variant classification with potential clinical significance and 23 new gene-disease associations. Re-examination of 4929 targeted sequencing files highlighted 45 changes in variant classification, and of these classifications, 89% were expert validated, leading to 4 additional diagnoses. Genome Alert! gene-disease association catalog provided 75 high-confidence associations not available in the OMIM morbid list; of which, 20% became available in OMIM morbid list For more than 356 negative exome sequencing data that were reannotated for variants in these 75 genes, this elective approach led to a new diagnosis. CONCLUSION: Genome Alert! (https://genomealert.univ-grenoble-alpes.fr/) enables systematic and reproducible reinterpretation of acquired sequencing data in a clinical routine with limited human resource effect.

Molecular characterization and investigation of the role of genetic variation in phenotypic variability and response to treatment in a large pediatric Marfan syndrome cohort.

PURPOSE: In a large cohort of 373 pediatric patients with Marfan syndrome (MFS) with a severe cardiovascular phenotype, we explored the proportion of patients with MFS with a pathogenic FBN1 variant and analyzed whether the type/location of FBN1 variants was associated with specific clinical characteristics and response to treatment. Patients were recruited on the basis of the following criteria: aortic root z-score > 3, age 6 months to 25 years, no prior or planned surgery, and aortic root diameter < 5 cm. METHODS: Targeted resequencing and deletion/duplication testing of FBN1 and related genes were performed. RESULTS: We identified (likely) pathogenic FBN1 variants in 91% of patients. Ectopia lentis was more frequent in patients with dominant-negative (DN) variants (61%) than in those with haploinsufficient variants (27%). For DN FBN1 variants, the prevalence of ectopia lentis was highest in the N-terminal region (84%) and lowest in the C-terminal region (17%). The association with a more severe cardiovascular phenotype was not restricted to DN variants in the neonatal FBN1 region (exon 25-33) but was also seen in the variants in exons 26 to 49. No difference in the therapeutic response was detected between genotypes. CONCLUSION: Important novel genotype-phenotype associations involving both cardiovascular and extra-cardiovascular manifestations were identified, and existing ones were confirmed. These findings have implications for prognostic counseling of families with MFS.

Genetic basis of growth reaction to drought stress differs in contrasting high-latitude treeline ecotones of a widespread conifer.

Climate change is increasing the frequency and intensity of drought events in many boreal forests. Trees are sessile organisms with a long generation time, which makes them vulnerable to fast climate change and hinders fast adaptations. Therefore, it is important to know how forests cope with drought stress and to explore the genetic basis of these reactions. We investigated three natural populations of white spruce (Picea glauca) in Alaska, located at one drought-limited and two cold-limited treelines with a paired plot design of one forest and one treeline plot. We obtained individual increment cores from 458 trees and climate data to assess dendrophenotypes, in particular the growth reaction to drought stress. To explore the genetic basis of these dendrophenotypes, we genotyped the individual trees at 3000 single nucleotide polymorphisms in candidate genes and performed genotype-phenotype association analysis using linear mixed models and Bayesian sparse linear mixed models. Growth reaction to drought stress differed in contrasting treeline populations. Therefore, the populations are likely to be unevenly affected by climate change. We identified 40 genes associated with dendrophenotypic traits that differed among the treeline populations. Most genes were identified in the drought-limited site, indicating comparatively strong selection pressure of drought-tolerant phenotypes. Contrasting patterns of drought-associated genes among sampled sites and in comparison to Canadian populations in a previous study suggest that drought adaptation acts on a local scale. Our results highlight genes that are associated with wood traits which in turn are critical for the establishment and persistence of future forests under climate change.

Genes underlying the evolution of tetrapod testes size.

BACKGROUND: Testes vary widely in mass relative to body mass across species, but we know very little about which genes underlie and contribute to such variation. This is partly because evidence for which genes are implicated in testis size variation tends to come from investigations involving just one or a few species. Contemporary comparative phylogenetic methods provide an opportunity to test candidate genes for their role in phenotypic change at a macro-evolutionary scale-across species and over millions of years. Previous attempts to detect genotype-phenotype associations across species have been limited in that they can only detect where genes have driven directional selection (e.g. brain size increase). RESULTS: Here, we introduce an approach that uses rates of evolutionary change to overcome this limitation to test whether any of twelve candidate genes have driven testis size evolution across tetrapod vertebrates-regardless of directionality. We do this by seeking a relationship between the rates of genetic and phenotypic evolution. Our results reveal five genes (Alkbh5, Dmrtb1, Pld6, Nlrp3, Sp4) that each have played unique and complex roles in tetrapod testis size diversity. In all five genes, we find strong significant associations between the rate of protein-coding substitutions and the rate of testis size evolution. Such an association has never, to our knowledge, been tested before for any gene or phenotype. CONCLUSIONS: We describe a new approach to tackle one of the most fundamental questions in biology: how do individual genes give rise to biological diversity? The ability to detect genotype-phenotype associations that have acted across species has the potential to build a picture of how natural selection has sculpted phenotypic change over millions of years.

Endotyping asthma related to 3 different work exposures.

BACKGROUND: Work exposures play a significant role in adult-onset asthma, but the mechanisms of work-related asthma are not fully elucidated. OBJECTIVE: We aimed to reveal the molecular mechanisms of work-related asthma associated with exposure to flour (flour asthma), isocyanate (isocyanate asthma), or welding fumes (welding asthma) and identify potential biomarkers that distinguish these groups from each other. METHODS: We used a combination of clinical tests, transcriptomic analysis, and associated pathway analyses to investigate the underlying disease mechanisms of the blood immune cells and the airway epithelium of 61 men. RESULTS: Compared with the healthy controls, the welding asthma patients had more differentially expressed genes than the flour asthma and isocyanate asthma patients, both in the airway epithelia and in the blood immune cells. In the airway epithelia, active inflammation was detected only in welding asthma patients. In contrast, many differentially expressed genes were detected in blood cells in all 3 asthma groups. Disease-related immune functions in blood cells, including leukocyte migration and inflammatory responses, and decreased expression of upstream cytokines such as TNF and IFN-γ were suppressed in all the asthma groups. In transcriptome-phenotype correlations, hyperresponsiveness (R ∼ |0.6|) had the highest clinical relevance and was associated with a set of exposure group-specific genes. Finally, biomarker subsets of only 5 genes specifically distinguished each of the asthma exposure groups. CONCLUSIONS: This study provides novel data on the molecular mechanisms underlying work-related asthma. We identified a set of 5 promising biomarkers in asthma related to flour, isocyanate, and welding fume exposure to be tested and clinically validated in future studies.

Discovery and Mediation Analysis of Cross-Phenotype Associations Between Asthma and Body Mass Index in 12q13.2.

Twin studies suggest that shared genetics contributes to the comorbidity of asthma and obesity, but candidate-gene studies provide limited evidence of pleiotropy. We conducted genome-wide association analyses of asthma and body mass index (BMI; weight (kg)/height (m)2)) among 305,945 White British subjects recruited into the UK Biobank in 2006-2010. We searched for overlapping signals and conducted mediation analyses on genome-wide-significant cross-phenotype associations, assessing moderation by sex and age at asthma diagnosis, and adjusting for confounders of the asthma-BMI relationship. We identified a genome-wide-significant cross-phenotype association at rs705708 (asthma odds ratio = 1.05, 95% confidence interval: 1.03, 1.07; P = 7.20 × 10-9; and BMI ß = -0.065, 95% confidence interval: -0.087, -0.042; P = 1.30 × 10-8). rs705708 resides on 12q13.2, which harbors 9 other asthma- and BMI-associated variants (all P < 5 × 10-5 for asthma; all but one P < 5 × 10-5 for BMI). Follow-up analyses of rs705708 show that most of the BMI association occurred independently of asthma, with consistent magnitude between men and women and persons with and without asthma, irrespective of age at diagnosis; the asthma association was stronger for childhood versus adult asthma; and both associations remained after confounder adjustment. This suggests that 12q13.2 displays pleiotropy for asthma and BMI. Upon further characterization, 12q13.2 might provide a target for interventions that simultaneously prevent or treat asthma and obesity.

Rapid repeatable phenotypic and genomic adaptation following multiple introductions.

Uncovering the genomic basis of repeated adaption can provide important insights into the constraints and biases that limit the diversity of genetic responses. Demographic processes such as admixture or bottlenecks affect genetic variation underlying traits experiencing selection. The impact of these processes on the genetic basis of adaptation remains, however, largely unexamined empirically. We here test repeatability in phenotypes and genotypes along parallel climatic clines within the native North American and introduced European and Australian Ambrosia artemisiifolia ranges. To do this, we combined multiple lines of evidence from phenotype-environment associations, FST -like outlier tests, genotype-environment associations and genotype-phenotype associations. We used 853 individuals grown in common garden from 84 sampling locations, targeting 19 phenotypes, >83 k SNPs and 22 environmental variables. We found that 17%-26% of loci with adaptive signatures were repeated among ranges, despite alternative demographic histories shaping genetic variation and genetic associations. Our results suggest major adaptive changes can occur on short timescales, with seemingly minimum impacts due to demographic changes linked to introduction. These patterns reveal some predictability of evolutionary change during range expansion, key in a world facing ongoing climate change, and rapid invasive spread.

Genomic footprints of adaptation in a cooperatively breeding tropical bird across a vegetation gradient.

Identifying the genetic basis of phenotypic variation and its relationship with the environment is key to understanding how local adaptations evolve. Such patterns are especially interesting among populations distributed across habitat gradients, where genetic structure can be driven by isolation by distance (IBD) and/or isolation by environment (IBE). Here, we used variation in ~1,600 high-quality SNPs derived from paired-end sequencing of double-digest restriction site-associated DNA (ddRAD-Seq) to test hypotheses related to IBD and IBE in the Yucatan jay (Cyanocorax yucatanicus), a tropical bird endemic to the Yucatán Peninsula. This peninsula is characterized by a precipitation and vegetation gradient-from dry to evergreen tropical forests-that is associated with morphological variation in this species. We found a moderate level of nucleotide diversity (π = .008) and little evidence for genetic differentiation among vegetation types. Analyses of neutral and putatively adaptive SNPs (identified by complementary genome-scan approaches) indicate that IBD is the most reliable explanation to account for frequency distribution of the former, while IBE has to be invoked to explain those of the later. These results suggest that selective factors acting along a vegetation gradient can promote local adaptation in the presence of gene flow in a vagile, nonmigratory and geographically restricted species. The putative candidate SNPs identified here are located within or linked to a variety of genes that represent ideal targets for future genomic surveys.

Different tradeoffs result from alternate genetic adaptations to a common environment.

Fitness tradeoffs are often assumed by evolutionary theory, yet little is known about the frequency of fitness tradeoffs during stress adaptation. Even less is known about the genetic factors that confer these tradeoffs and whether alternative adaptive mutations yield contrasting tradeoff dynamics. We addressed these issues using 114 clones of Escherichia coli that were evolved independently for 2,000 generations under thermal stress (42.2 °C). For each clone, we measured their fitness relative to the ancestral clone at 37 °C and 20 °C. Tradeoffs were common at 37 °C but more prevalent at 20 °C, where 56% of clones were outperformed by the ancestor. We also characterized the upper and lower thermal boundaries of each clone. All clones shifted their upper boundary to at least 45 °C; roughly half increased their lower niche boundary concomitantly, representing a shift of thermal niche. The remaining clones expanded their thermal niche by increasing their upper limit without a commensurate increase of lower limit. We associated these niche dynamics with genotypes and confirmed associations by engineering single mutations in the rpoB gene, which encodes the beta subunit of RNA polymerase, and the rho gene, which encodes a termination factor. Single mutations in the rpoB gene exhibit antagonistic pleiotropy, with fitness tradeoffs at 18 °C and fitness benefits at 42.2 °C. In contrast, a mutation within the rho transcriptional terminator, which defines an alternative adaptive pathway from that of rpoB, had no demonstrable effect on fitness at 18 °C. This study suggests that two different genetic pathways toward high-temperature adaptation have contrasting effects with respect to thermal tradeoffs.

Comparative genomics of the mimicry switch in Papilio dardanus.

The African Mocker Swallowtail, Papilio dardanus, is a textbook example in evolutionary genetics. Classical breeding experiments have shown that wing pattern variation in this polymorphic Batesian mimic is determined by the polyallelic H locus that controls a set of distinct mimetic phenotypes. Using bacterial artificial chromosome (BAC) sequencing, recombination analyses and comparative genomics, we show that H co-segregates with an interval of less than 500 kb that is collinear with two other Lepidoptera genomes and contains 24 genes, including the transcription factor genes engrailed (en) and invected (inv). H is located in a region of conserved gene order, which argues against any role for genomic translocations in the evolution of a hypothesized multi-gene mimicry locus. Natural populations of P. dardanus show significant associations of specific morphs with single nucleotide polymorphisms (SNPs), centred on en. In addition, SNP variation in the H region reveals evidence of non-neutral molecular evolution in the en gene alone. We find evidence for a duplication potentially driving physical constraints on recombination in the lamborni morph. Absence of perfect linkage disequilibrium between different genes in the other morphs suggests that H is limited to nucleotide positions in the regulatory and coding regions of en. Our results therefore support the hypothesis that a single gene underlies wing pattern variation in P. dardanus.

Genotype-phenotype spectrum and correlation of PHARC Syndrome due to pathogenic ABHD12 variants.

BACKGROUND: A comprehensive understanding of the genetic basis of rare diseases and their regulatory mechanisms is essential for human molecular genetics. However, the genetic mutant spectrum of pathogenic genes within the Chinese population remains underrepresented. Here, we reported previously unreported functional ABHD12 variants in two Chinese families and explored the correlation between genetic polymorphisms and phenotypes linked to PHARC syndrome. METHODS: Participants with biallelic pathogenic ABHD12 variants were recruited from the Chinese Deafness Genetics Cohort. These participants underwent whole-genome sequencing. Subsequently, a comprehensive literature review was conducted. RESULTS: Two Han Chinese families were identified, one with a compound heterozygous variant and the other with a novel homozygous variant in ABHD12. Among 65 PHARC patients, including 62 from the literature and 3 from this study, approximately 90% (57 out of 63) exhibited hearing loss, 82% (50 out of 61) had cataracts, 82% (46 out of 56) presented with retinitis pigmentosa, 79% (42 out of 53) experienced polyneuropathy, and 63% (36 out of 57) displayed ataxia. Seventeen different patterns were observed in the five main phenotypes of PHARC syndrome. A total of 33 pathogenic variants were identified in the ABHD12. Compared with other genotypes, individuals with biallelic truncating variants showed a higher incidence of polyneuropathy (p = 0.006), but no statistically significant differences were observed in the incidence of hearing loss, ataxia, retinitis pigmentosa and cataracts. CONCLUSIONS: The diagnosis of PHARC syndrome is challenging because of its genetic heterogeneity. Therefore, exploring novel variants and establishing genotype-phenotype correlations can significantly enhance gene diagnosis and genetic counseling for this complex disease.

Improving the classification of cardinality phenotypes using collections.

MOTIVATION: Phenotypes are observable characteristics of an organism and they can be highly variable. Information about phenotypes is collected in a clinical context to characterize disease, and is also collected in model organisms and stored in model organism databases where they are used to understand gene functions. Phenotype data is also used in computational data analysis and machine learning methods to provide novel insights into disease mechanisms and support personalized diagnosis of disease. For mammalian organisms and in a clinical context, ontologies such as the Human Phenotype Ontology and the Mammalian Phenotype Ontology are widely used to formally and precisely describe phenotypes. We specifically analyze axioms pertaining to phenotypes of collections of entities within a body, and we find that some of the axioms in phenotype ontologies lead to inferences that may not accurately reflect the underlying biological phenomena. RESULTS: We reformulate the phenotypes of collections of entities using an ontological theory of collections. By reformulating phenotypes of collections in phenotypes ontologies, we avoid potentially incorrect inferences pertaining to the cardinality of these collections. We apply our method to two phenotype ontologies and show that the reformulation not only removes some problematic inferences but also quantitatively improves biological data analysis.