Integrating deep phenotyping with genetic analysis: a comprehensive workflow for diagnosis and management of rare bone diseases.

Phenotypes play a fundamental role in medical genetics, serving as external manifestations of underlying genotypes. Deep phenotyping, a cornerstone of precision medicine, involves precise multi-system phenotype assessments, facilitating disease subtyping and genetic understanding. Despite their significance, the field lacks standardized protocols for accurate phenotype evaluation, hindering clinical comprehension and research comparability. We present a comprehensive workflow of deep phenotyping for rare bone diseases from the Genetics Clinic of Skeletal Deformity at Peking Union Medical College Hospital. Our workflow integrates referral, informed consent, and detailed phenotype evaluation through HPO standards, capturing nuanced phenotypic characteristics using clinical examinations, questionnaires, and multimedia documentation. Genetic testing and counseling follow, based on deep phenotyping results, ensuring personalized interventions. Multidisciplinary team consultations facilitate comprehensive patient care and clinical guideline development. Regular follow-up visits emphasize dynamic phenotype reassessment, ensuring treatment strategies remain responsive to evolving patient needs. In conclusion, this study highlights the importance of deep phenotyping in rare bone diseases, offering a standardized framework for phenotype evaluation, genetic analysis, and multidisciplinary intervention. By enhancing clinical care and research outcomes, this approach contributes to the advancement of precision medicine in the field of medical genetics.

Advances in big data and omics: Paving the way for discovery in childhood epilepsies.

The insights gained from big data and omics approaches have transformed the field of childhood genetic epilepsy. With an increasing number of individuals receiving genetic testing for seizures, we are provided with an opportunity to identify clinically relevant subgroups and extract meaningful observations from this large-scale clinical data. However, the volume of data from electronic medical records and omics (e.g., genomics, transcriptomics) is so vast that standardized methods, such as the Human Phenotype Ontology, are necessary for reliable and comprehensive characterization. Here, we explore the integration of clinical and omics data, highlighting how these approaches pave the way for discovery in childhood epilepsies.

Objectivizing issues in the diagnosis of complex rare diseases: lessons learned from testing existing diagnosis support systems on ciliopathies.

BACKGROUND: There are approximately 8,000 different rare diseases that affect roughly 400 million people worldwide. Many of them suffer from delayed diagnosis. Ciliopathies are rare monogenic disorders characterized by a significant phenotypic and genetic heterogeneity that raises an important challenge for clinical diagnosis. Diagnosis support systems (DSS) applied to electronic health record (EHR) data may help identify undiagnosed patients, which is of paramount importance to improve patients' care. Our objective was to evaluate three online-accessible rare disease DSSs using phenotypes derived from EHRs for the diagnosis of ciliopathies. METHODS: Two datasets of ciliopathy cases, either proven or suspected, and two datasets of controls were used to evaluate the DSSs. Patient phenotypes were automatically extracted from their EHRs and converted to Human Phenotype Ontology terms. We tested the ability of the DSSs to diagnose cases in contrast to controls based on Orphanet ontology. RESULTS: A total of 79 cases and 38 controls were selected. Performances of the DSSs on ciliopathy real world data (best DSS with area under the ROC curve = 0.72) were not as good as published performances on the test set used in the DSS development phase. None of these systems obtained results which could be described as "expert-level". Patients with multisystemic symptoms were generally easier to diagnose than patients with isolated symptoms. Diseases easily confused with ciliopathy generally affected multiple organs and had overlapping phenotypes. Four challenges need to be considered to improve the performances: to make the DSSs interoperable with EHR systems, to validate the performances in real-life settings, to deal with data quality, and to leverage methods and resources for rare and complex diseases. CONCLUSION: Our study provides insights into the complexities of diagnosing highly heterogenous rare diseases and offers lessons derived from evaluation existing DSSs in real-world settings. These insights are not only beneficial for ciliopathy diagnosis but also hold relevance for the enhancement of DSS for various complex rare disorders, by guiding the development of more clinically relevant rare disease DSSs, that could support early diagnosis and finally make more patients eligible for treatment.

GestaltMML: Enhancing Rare Genetic Disease Diagnosis through Multimodal Machine Learning Combining Facial Images and Clinical Texts.

Individuals with suspected rare genetic disorders often undergo multiple clinical evaluations, imaging studies, laboratory tests and genetic tests, to find a possible answer over a prolonged period of time. Addressing this "diagnostic odyssey" thus has substantial clinical, psychosocial, and economic benefits. Many rare genetic diseases have distinctive facial features, which can be used by artificial intelligence algorithms to facilitate clinical diagnosis, in prioritizing candidate diseases to be further examined by lab tests or genetic assays, or in helping the phenotype-driven reinterpretation of genome/exome sequencing data. Existing methods using frontal facial photos were built on conventional Convolutional Neural Networks (CNNs), rely exclusively on facial images, and cannot capture non-facial phenotypic traits and demographic information essential for guiding accurate diagnoses. Here we introduce GestaltMML, a multimodal machine learning (MML) approach solely based on the Transformer architecture. It integrates facial images, demographic information (age, sex, ethnicity), and clinical notes (optionally, a list of Human Phenotype Ontology terms) to improve prediction accuracy. Furthermore, we also evaluated GestaltMML on a diverse range of datasets, including 528 diseases from the GestaltMatcher Database, several in-house datasets of Beckwith-Wiedemann syndrome (BWS, over-growth syndrome with distinct facial features), Sotos syndrome (overgrowth syndrome with overlapping features with BWS), NAA10-related neurodevelopmental syndrome, Cornelia de Lange syndrome (multiple malformation syndrome), and KBG syndrome (multiple malformation syndrome). Our results suggest that GestaltMML effectively incorporates multiple modalities of data, greatly narrowing candidate genetic diagnoses of rare diseases and may facilitate the reinterpretation of genome/exome sequencing data.

The clinical and genetic spectrum of paediatric speech and language disorders in 52,143 individuals.

Speech and language disorders are known to have a substantial genetic contribution. Although frequently examined as components of other conditions, research on the genetic basis of linguistic differences as separate phenotypic subgroups has been limited so far. Here, we performed an in-depth characterization of speech and language disorders in 52,143 individuals, reconstructing clinical histories using a large-scale data mining approach of the Electronic Medical Records (EMR) from an entire large paediatric healthcare network. The reported frequency of these disorders was the highest between 2 and 5 years old and spanned a spectrum of twenty-six broad speech and language diagnoses. We used Natural Language Processing to assess to which degree clinical diagnosis in full-text notes were reflected in ICD-10 diagnosis codes. We found that aphasia and speech apraxia could be easily retrieved through ICD-10 diagnosis codes, while stuttering as a speech phenotype was only coded in 12% of individuals through appropriate ICD-10 codes. We found significant comorbidity of speech and language disorders in neurodevelopmental conditions (30.31%) and to a lesser degree with epilepsies (6.07%) and movement disorders (2.05%). The most common genetic disorders retrievable in our EMR analysis were STXBP1 (n=21), PTEN (n=20), and CACNA1A (n=18). When assessing associations of genetic diagnoses with specific linguistic phenotypes, we observed associations of STXBP1 and aphasia (P=8.57 × 10-7, CI=18.62-130.39) and MYO7A with speech and language development delay due to hearing loss (P=1.24 × 10-5, CI=17.46-Inf). Finally, in a sub-cohort of 726 individuals with whole exome sequencing data, we identified an enrichment of rare variants in synaptic protein and neuronal receptor pathways and associations of UQCRC1 with expressive aphasia and WASHC4 with abnormality of speech or vocalization. In summary, our study outlines the landscape of paediatric speech and language disorders, confirming the phenotypic complexity of linguistic traits and novel genotype-phenotype associations. Subgroups of paediatric speech and language disorders differ significantly with respect to the composition of monogenic aetiologies.

Rare Genetic Developmental Disabilities: Mabry Syndrome (MIM 239300) Index Cases and Glycophosphatidylinositol (GPI) Disorders.

The case report by Mabry et al. (1970) of a family with four children with elevated tissue non-specific alkaline phosphatase, seizures and profound developmental disability, became the basis for phenotyping children with the features that became known as Mabry syndrome. Aside from improvements in the services available to patients and families, however, the diagnosis and treatment of this, and many other developmental disabilities, did not change significantly until the advent of massively parallel sequencing. As more patients with features of the Mabry syndrome were identified, exome and genome sequencing were used to identify the glycophosphatidylinositol (GPI) biosynthesis disorders (GPIBDs) as a group of congenital disorders of glycosylation (CDG). Biallelic variants of the phosphatidylinositol glycan (PIG) biosynthesis, type V (PIGV) gene identified in Mabry syndrome became evidence of the first in a phenotypic series that is numbered HPMRS1-6 in the order of discovery. HPMRS1 [MIM: 239300] is the phenotype resulting from inheritance of biallelic PIGV variants. Similarly, HPMRS2 (MIM 614749), HPMRS5 (MIM 616025) and HPMRS6 (MIM 616809) result from disruption of the PIGO, PIGW and PIGY genes expressed in the endoplasmic reticulum. By contrast, HPMRS3 (MIM 614207) and HPMRS4 (MIM 615716) result from disruption of post attachment to proteins PGAP2 (HPMRS3) and PGAP3 (HPMRS4). The GPI biosynthesis disorders (GPIBDs) are currently numbered GPIBD1-21. Working with Dr. Mabry, in 2020, we were able to use improved laboratory diagnostics to complete the molecular diagnosis of patients he had originally described in 1970. We identified biallelic variants of the PGAP2 gene in the first reported HPMRS patients. We discuss the longevity of the Mabry syndrome index patients in the context of the utility of pyridoxine treatment of seizures and evidence for putative glycolipid storage in patients with HPMRS3. From the perspective of the laboratory innovations made that enabled the identification of the HPMRS phenotype in Dr. Mabry's patients, the need for treatment innovations that will benefit patients and families affected by developmental disabilities is clear.

Multilayer concept of autoimmune mechanisms and manifestations in inborn errors of immunity: Relevance for precision therapy.

Autoimmunity in inborn errors of immunity (IEIs) has a multifactorial pathogenesis and develops subsequent to a genetic predisposition in conjunction with gene regulation, environmental modifiers, and infectious triggers. On the basis of incremental data availability owing to upfront application of omics technologies, a more granular and dynamic view of mechanisms and manifestations is warranted. Here, we present a comprehensive novel concept of autoimmunity in IEIs that considers multiple layers of interdependent elements and connects 101 causative genes or deletions according to the quality of the allelic variants with 47 molecular pathways and 22 immune effector mechanisms. Furthermore, we list 50 resulting manifestations together with the corresponding Human Phenotype Ontology terms and review the types and frequencies of the most relevant clinical presentations. When all of its elements are taken together, this concept (1) extends the historical anatomic view of central versus peripheral tolerance toward multiple interdependent mechanisms of immune tolerance, (2) delineates the mechanisms underlying the protean clinical manifestations, and thereby, (3) points toward the most suitable precision therapy for autoimmunity in IEIs. The multilayer concept of autoimmune mechanisms and manifestations in IEIs will facilitate research design and provide clinical guidance on the use of precision medicine irrespective of the data depth available in each health care scenario.

Enhancing phenotype recognition in clinical notes using large language models: PhenoBCBERT and PhenoGPT.

To enhance phenotype recognition in clinical notes of genetic diseases, we developed two models-PhenoBCBERT and PhenoGPT-for expanding the vocabularies of Human Phenotype Ontology (HPO) terms. While HPO offers a standardized vocabulary for phenotypes, existing tools often fail to capture the full scope of phenotypes due to limitations from traditional heuristic or rule-based approaches. Our models leverage large language models to automate the detection of phenotype terms, including those not in the current HPO. We compare these models with PhenoTagger, another HPO recognition tool, and found that our models identify a wider range of phenotype concepts, including previously uncharacterized ones. Our models also show strong performance in case studies on biomedical literature. We evaluate the strengths and weaknesses of BERT- and GPT-based models in aspects such as architecture and accuracy. Overall, our models enhance automated phenotype detection from clinical texts, improving downstream analyses on human diseases.

MSeqDR Quick-Mitome (QM): Combining Phenotype-Guided Variant Interpretation and Machine Learning Classifiers to Aid Primary Mitochondrial Disease Genetic Diagnosis.

The international Mitochondrial Disease Sequence Data Resource Consortium (MSeqDR) Quick-Mitome (QM) is a web-based platform enabling automated variant interpretation of whole-exome sequencing (WES) datasets for the genetic diagnosis of primary mitochondrial diseases (PMD). Designed specifically to address the unique dual genome nature of PMD etiologies, QM includes features for both nuclear and mitochondrial DNA (mtDNA) genome analysis. QM requires VCF variant lists, HPO ID clinical phenotypes, and pedigree files for multiple-sample VCF inputs. QM maps phenotypes to HPO terms before analysis. QM analysis requires 2 to 20 min for 100,000 variants on an 8-vCPU AWS server using Exomiser's "PASS_ONLY" mode for nuclear variants. QM ranks variants based on allele frequency, phenotype-gene association, functional impact, and inheritance mode. Variants are further annotated with multiple data sources such as OMIM, ClinVar, dbNSFP, gnoMAD, MITOMAP, and MSeqDR. In addition to standard Exomiser results, QM generates an Analysis Report and QM Integrated Report with add-on mtDNA-specific analyses, including haplogroup prediction with Phy-Mer, heteroplasmy calculation, and mvTool annotations. We developed the Mitochondrial Disease Variant (MDV) classifier using XGBoost to predict variant pathogenicity for PMD. The MDV classifier was trained on >120 features and performance benchmarking showed that it correctly classified >98% of nuclear gene variants as being pathogenic or benign, and predicted PMD-causing variants with 94% precision. The MSeqDR QM server is an open-access resource for phenotype-driven dual-genome analyses for PMD diagnosis by the global mitochondrial disease community. It is publicly available for non-commercial, non-clinical research use at https://mseqdr.org/quickmitome.php. © 2024 Wiley Periodicals LLC. Basic Protocol 1: Standardizing clinical phenotypes into human phenotype ontology (HPO) terms as the phenotype input for Quick-Mitome (QM) Basic Protocol 2: Prepare the pedigree input for multiple-sample VCF Basic Protocol 3: Quick-Mitome (QM) analysis Basic Protocol 4: Reviewing and understanding the QM Integrated Report and Analysis Report.

An evaluation of GPT models for phenotype concept recognition.

OBJECTIVE: Clinical deep phenotyping and phenotype annotation play a critical role in both the diagnosis of patients with rare disorders as well as in building computationally-tractable knowledge in the rare disorders field. These processes rely on using ontology concepts, often from the Human Phenotype Ontology, in conjunction with a phenotype concept recognition task (supported usually by machine learning methods) to curate patient profiles or existing scientific literature. With the significant shift in the use of large language models (LLMs) for most NLP tasks, we examine the performance of the latest Generative Pre-trained Transformer (GPT) models underpinning ChatGPT as a foundation for the tasks of clinical phenotyping and phenotype annotation. MATERIALS AND METHODS: The experimental setup of the study included seven prompts of various levels of specificity, two GPT models (gpt-3.5-turbo and gpt-4.0) and two established gold standard corpora for phenotype recognition, one consisting of publication abstracts and the other clinical observations. RESULTS: The best run, using in-context learning, achieved 0.58 document-level F1 score on publication abstracts and 0.75 document-level F1 score on clinical observations, as well as a mention-level F1 score of 0.7, which surpasses the current best in class tool. Without in-context learning, however, performance is significantly below the existing approaches. CONCLUSION: Our experiments show that gpt-4.0 surpasses the state of the art performance if the task is constrained to a subset of the target ontology where there is prior knowledge of the terms that are expected to be matched. While the results are promising, the non-deterministic nature of the outcomes, the high cost and the lack of concordance between different runs using the same prompt and input make the use of these LLMs challenging for this particular task.

RDmaster: A novel phenotype-oriented dialogue system supporting differential diagnosis of rare disease.

BACKGROUND: Clinicians often lack the necessary expertise to differentially diagnose multiple underlying rare diseases (RDs) due to their complex and overlapping clinical features, leading to misdiagnoses and delayed treatments. The aim of this study is to develop a novel electronic differential diagnostic support system for RDs. METHOD: Through integrating two Bayesian diagnostic methods, a candidate list was generated with enhance clinical interpretability for the further Q&A based differential diagnosis (DDX). To achieve an efficient Q&A dialogue strategy, we introduce a novel metric named the adaptive information gain and Gini index (AIGGI) to evaluate the expected gain of interrogated phenotypes within real-time diagnostic states. RESULTS: This DDX tool called RDmaster has been implemented as a web-based platform (http://rdmaster.nbscn.org/). A diagnostic trial involving 238 published RD patients revealed that RDmaster outperformed existing RD diagnostic tools, as well as ChatGPT, and was shown to enhance the diagnostic accuracy through its Q&A system. CONCLUSIONS: The RDmaster offers an effective multi-omics differential diagnostic technique and outperforms existing tools and popular large language models, particularly enhancing differential diagnosis in collecting diagnostically beneficial phenotypes.

Enhancing Phenotype Recognition in Clinical Notes Using Large Language Models: PhenoBCBERT and PhenoGPT.

To enhance phenotype recognition in clinical notes of genetic diseases, we developed two models - PhenoBCBERT and PhenoGPT - for expanding the vocabularies of Human Phenotype Ontology (HPO) terms. While HPO offers a standardized vocabulary for phenotypes, existing tools often fail to capture the full scope of phenotypes, due to limitations from traditional heuristic or rule-based approaches. Our models leverage large language models (LLMs) to automate the detection of phenotype terms, including those not in the current HPO. We compared these models to PhenoTagger, another HPO recognition tool, and found that our models identify a wider range of phenotype concepts, including previously uncharacterized ones. Our models also showed strong performance in case studies on biomedical literature. We evaluated the strengths and weaknesses of BERT-based and GPT-based models in aspects such as architecture and accuracy. Overall, our models enhance automated phenotype detection from clinical texts, improving downstream analyses on human diseases.

Curation and expansion of the Human Phenotype Ontology for systemic autoinflammatory diseases improves phenotype-driven disease-matching.

Introduction: Accurate and standardized phenotypic descriptions are essential in diagnosing rare diseases and discovering new diseases, and the Human Phenotype Ontology (HPO) system was developed to provide a rich collection of hierarchical phenotypic descriptions. However, although the HPO terms for inborn errors of immunity have been improved and curated, it has not been investigated whether this curation improves the diagnosis of systemic autoinflammatory disease (SAID) patients. Here, we aimed to study if improved HPO annotation for SAIDs enhanced SAID identification and to demonstrate the potential of phenotype-driven genome diagnostics using curated HPO terms for SAIDs. Methods: We collected HPO terms from 98 genetically confirmed SAID patients across eight different European SAID expertise centers and used the LIRICAL (Likelihood Ratio Interpretation of Clinical Abnormalities) computational algorithm to estimate the effect of HPO curation on the prioritization of the correct SAID for each patient. Results: Our results show that the percentage of correct diagnoses increased from 66% to 86% and that the number of diagnoses with the highest ranking increased from 38 to 45. In a further pilot study, curation also improved HPO-based whole-exome sequencing (WES) analysis, diagnosing 10/12 patients before and 12/12 after curation. In addition, the average number of candidate diseases that needed to be interpreted decreased from 35 to 2. Discussion: This study demonstrates that curation of HPO terms can increase identification of the correct diagnosis, emphasizing the high potential of HPO-based genome diagnostics for SAIDs.

Identification and validation of subclusters of papillary thyroid carcinoma based on Human Phenotype Ontology.

Background: The increase in the diagnosis of papillary thyroid carcinoma (PTC) has prompted researchers to establish a diagnostic model and identify functional subclusters. The Human Phenotype Ontology (HPO) platform is widely available for differential diagnostics and phenotype-driven investigations based on next-generation sequence-variation data. However, a systematic and comprehensive study to identify and validate PTC subclusters based on HPO is lacking. Methods: We first used the HPO platform to identify the PTC subclusters. An enrichment analysis was then conducted to examine the key biological processes and pathways associated with the subclusters, and a gene mutation analysis of the subclusters was conducted. For each subcluster, the differentially expressed genes (DEGs) were selected and validated. Finally, a single-cell RNA-sequencing data set was used to verify the DEGs. Results: In our study, 489 PTC patients from The Cancer Genome Atlas (TCGA) were included. Our analysis demonstrated that distinct subclusters of PTC are associated with different survival times and have different functional enrichment, and that C-C motif chemokine ligand 21 (CCL21) and zinc finger CCHC-type containing 12 (ZCCHC12) were the common down- and upregulated genes, respectively, in the 4 subclusters. Additionally, 20 characteristic genes were identified in the 4 subclusters, some of which have previously been reported to have roles in PTC. Further, we found that these characteristic genes were mainly expressed in thyrocytes, endothelial cells, and fibroblasts, and were rarely expressed in immune cells. Conclusions: We first identified subclusters in PTC based on HPO and found that patients with distinct subclusters have different prognoses. We then identified and validated the characteristic genes in the 4 subclusters. These findings are expected to serve as a crucial reference that will improve our understanding of PTC heterogeneity and the use of novel targets.

Rare variant enrichment analysis supports GREB1L as a contributory driver gene in the etiology of Mayer-Rokitansky-Küster-Hauser syndrome.

Mayer-Rokitansky-Küster-Hauser (MRKH) syndrome is characterized by aplasia of the female reproductive tract; the syndrome can include renal anomalies, absence or dysgenesis, and skeletal anomalies. While functional models have elucidated several candidate genes, only WNT4 (MIM: 603490) variants have been definitively associated with a subtype of MRKH with hyperandrogenism (MIM: 158330). DNA from 148 clinically diagnosed MRKH probands across 144 unrelated families and available family members from North America, Europe, and South America were exome sequenced (ES) and by family-based genomics analyzed for rare likely deleterious variants. A replication cohort consisting of 442 Han Chinese individuals with MRKH was used to further reproduce GREB1L findings in diverse genetic backgrounds. Proband and OMIM phenotypes annotated using the Human Phenotype Ontology were analyzed to quantitatively delineate the phenotypic spectrum associated with GREB1L variant alleles found in our MRKH cohort and those previously published. This study reports 18 novel GREB1L variant alleles, 16 within a multiethnic MRKH cohort and two within a congenital scoliosis cohort. Cohort-wide analyses for a burden of rare variants within a single gene identified likely damaging variants in GREB1L (MIM: 617782), a known disease gene for renal hypoplasia and uterine abnormalities (MIM: 617805), in 16 of 590 MRKH probands. GREB1L variant alleles, including a CNV null allele, were found in 8 MRKH type 1 probands and 8 MRKH type II probands. This study used quantitative phenotypic analyses in a worldwide multiethnic cohort to identify and strengthen the association of GREB1L to isolated uterine agenesis (MRKH type I) and syndromic MRKH type II.

Phen2Disease: a phenotype-driven model for disease and gene prioritization by bidirectional maximum matching semantic similarities.

Human Phenotype Ontology (HPO)-based approaches have gained popularity in recent times as a tool for genomic diagnostics of rare diseases. However, these approaches do not make full use of the available information on disease and patient phenotypes. We present a new method called Phen2Disease, which utilizes the bidirectional maximum matching semantic similarity between two phenotype sets of patients and diseases to prioritize diseases and genes. Our comprehensive experiments have been conducted on six real data cohorts with 2051 cases (Cohort 1, n = 384; Cohort 2, n = 281; Cohort 3, n = 185; Cohort 4, n = 784; Cohort 5, n = 208; and Cohort 6, n = 209) and two simulated data cohorts with 1000 cases. The results of the experiments showed that Phen2Disease outperforms the three state-of-the-art methods when only phenotype information and HPO knowledge base are used, particularly in cohorts with fewer average numbers of HPO terms. We also observed that patients with higher information content scores have more specific information, leading to more accurate predictions. Moreover, Phen2Disease provides high interpretability with ranked diseases and patient HPO terms presented. Our method provides a novel approach to utilizing phenotype data for genomic diagnostics of rare diseases, with potential for clinical impact. Phen2Disease is freely available on GitHub at https://github.com/ZhuLab-Fudan/Phen2Disease.

Mapping Exposome Derived Phenotypes into SNOMED Codes.

Human phenotypes define the healthy or diseased status of an individual and they arise from the complex interactions between environmental and genetic factors. The whole set of human exposures constitute the human exposome. These exposures have multiple sources including physical and socioeconomic factors. In this manuscript we have used text mining techniques to retrieve 1295 and 1903 Human Phenotype Ontology terms associated with these exposome factors and we have subsequently mapped 83% and 90% of the HPO terms respectively) into SNOMED as a clinically actionable code. We have developed a proof-of-concept approach to facilitate the integration of exposomic and clinical data.

ChatGPT: a pioneering approach to complex prenatal differential diagnosis.

This commentary examines how ChatGPT can assist healthcare teams in the prenatal diagnosis of rare and complex cases by creating a differential diagnoses based on deidentified clinical findings, while also acknowledging its limitations.

Enriching representation learning using 53 million patient notes through human phenotype ontology embedding.

The Human Phenotype Ontology (HPO) is a dictionary of >15,000 clinical phenotypic terms with defined semantic relationships, developed to standardize phenotypic analysis. Over the last decade, the HPO has been used to accelerate the implementation of precision medicine into clinical practice. In addition, recent research in representation learning, specifically in graph embedding, has led to notable progress in automated prediction via learned features. Here, we present a novel approach to phenotype representation by incorporating phenotypic frequencies based on 53 million full-text health care notes from >1.5 million individuals. We demonstrate the efficacy of our proposed phenotype embedding technique by comparing our work to existing phenotypic similarity-measuring methods. Using phenotype frequencies in our embedding technique, we are able to identify phenotypic similarities that surpass current computational models. Furthermore, our embedding technique exhibits a high degree of agreement with domain experts' judgment. By transforming complex and multidimensional phenotypes from the HPO format into vectors, our proposed method enables efficient representation of these phenotypes for downstream tasks that require deep phenotyping. This is demonstrated in a patient similarity analysis and can further be applied to disease trajectory and risk prediction.

GA4GH Phenopackets: A Practical Introduction.

The Global Alliance for Genomics and Health (GA4GH) is developing a suite of coordinated standards for genomics for healthcare. The Phenopacket is a new GA4GH standard for sharing disease and phenotype information that characterizes an individual person, linking that individual to detailed phenotypic descriptions, genetic information, diagnoses, and treatments. A detailed example is presented that illustrates how to use the schema to represent the clinical course of a patient with retinoblastoma, including demographic information, the clinical diagnosis, phenotypic features and clinical measurements, an examination of the extirpated tumor, therapies, and the results of genomic analysis. The Phenopacket Schema, together with other GA4GH data and technical standards, will enable data exchange and provide a foundation for the computational analysis of disease and phenotype information to improve our ability to diagnose and conduct research on all types of disorders, including cancer and rare diseases.