Human phenotype ontology annotation and cluster analysis to unravel genetic defects in 707 cases with unexplained bleeding and platelet disorders

Abstract

Background: Heritable bleeding and platelet disorders (BPD) are heterogeneous and frequently have an unknown genetic basis. The BRIDGE-BPD study aims to discover new causal genes for BPD by high throughput sequencing using cluster analyses based on improved and standardised deep, multi-system phenotyping of cases.

Methods: We report a new approach in which the clinical and laboratory characteristics of BPD cases are annotated with adapted Human Phenotype Ontology (HPO) terms. Cluster analyses are then used to characterise groups of cases with similar HPO terms and variants in the same genes.

Results: We show that 60% of index cases with heritable BPD enrolled at 10 European or US centres were annotated with HPO terms indicating abnormalities in organ systems other than blood or blood-forming tissues, particularly the nervous system. Cases within pedigrees clustered closely together on the bases of their HPO-coded phenotypes, as did cases sharing several clinically suspected syndromic disorders. Cases subsequently found to harbour variants in ACTN1 also clustered closely, even though diagnosis of this recently described disorder was not possible using only the clinical and laboratory data available to the enrolling clinician.

Conclusions: These findings validate our novel HPO-based phenotype clustering methodology for known BPD, thus providing a new discovery tool for BPD of unknown genetic basis. This approach will also be relevant for other rare diseases with significant genetic heterogeneity.

Figure 1

Recruitment to the BRIDGE-BPD study by enrolment centre. Gender stratified recruitment to the BRIDGE-BPD Consortium study is shown according to enrolment centre. (A) University of Leuven, Leuven, Belgium; (B) Royal Free NHS Trust, London, UK; (C) Centre de Référence des Pathologies Plaquettaires, Pessac, France; (D) Imperial College Healthcare NHS Trust, London, UK; (E) Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; (F) University Hospitals Bristol NHS Foundation Trust, Bristol, UK; (G) Haematological Laboratory, Trousseau Children’s Hospital and INsermU1009, Paris, France; (H) Children’s Hospital of Philadelphia, Philadelphia, USA; (I) Great Ormond Street Hospital For Children NHS Trust, London, UK; (J) Charité Universitätsmedizin, Berlin, Germany.

Figure 2

Platelet characteristics of the BRIDGE-BPD study index cases. (A, B) The relationship between PLT and MPV in femtolitre (fL) in male and female index cases. The crosses indicate index cases with greater than four (males) or greater than five (females) bleeding symptoms. The circles indicate index cases with fewer bleeding symptoms. NA: not available - cases in which large platelets were identified by the enrolling clinician but the MPV was not recorded. (C) Heat map showing the results of light transmission aggregation tests with the indicated activating agonists, classified as normal or abnormal by the enrolling clinician. The numbers in brackets indicate the number of index cases tested for each agonist. Cases are ordered by the number of abnormal or not done results across all six agonists. The bar above the heat map is colour-coded to indicate the enrolment centre for each case.

Figure 3

Bleeding symptoms of BRIDGE-BPD study index cases. Heat map showing the bleeding symptoms recorded for the BRIDGE-BPD index cases. The numbers in brackets indicate the number of index cases reporting each bleeding symptom. Cases are ordered by the number of bleeding symptoms. The bar above heat map is colour-coded to indicate the enrolment centre for each case.

Figure 4

HPO terms coded in BRIDGE-BPD study index cases. (A) Bar plot indicating the number of human phenotype ontology (HPO) terms annotated to the index cases. (B) Heat map showing the relative frequencies of HPO terms pertinent to abnormalities in platelets and bleeding symptoms and in other organ or disease areas in the BRIDGE-BPD index cases. The numbers in the brackets and the barplots indicate the number of index cases with at least one HPO term pertinent to abnormality in the organ or disease area after removal of overlapping terms. *indicates that the distribution of terms pertinent to enrolment for a particular column is significantly different compared to the sum (along rows) of all other columns (P value <0.05 after Bonferroni correction by chi-squared test). The columns are ordered by the number of cases having a term in each leading class.

Figure 5

Phenotype clusters 18 and 29. Illustrative subgraphs of the HPO showing terms for the phenotype clusters 18 (15 cases) and 29 (16 cases). Arrows indicate direct (solid) or indirect (dashed) is a relations between terms in the ontology. DMPV: decreased mean platelet volume; PA: phenotypic abnormality; Plt-agg: platelet aggregation abnormality.

Figure 6

Phenotype similarity clustering of cases according to HPO terms. Heat map showing pairwise phenotypic similarity among affected members of pedigrees, cases with classical syndromes and cases with variants in ACTN1. The groups are ordered through complete-linkage hierarchical clustering within each class and P values of phenotypic similarity are shown in a scatterplot superimposed over a histogram showing the distribution of P values.