Automatically pre-screening patients for the rare disease aromatic l-amino acid decarboxylase deficiency using knowledge engineering, natural language processing, and machine learning on a large EHR population.

OBJECTIVES: Electronic health record (EHR) data may facilitate the identification of rare diseases in patients, such as aromatic l-amino acid decarboxylase deficiency (AADCd), an autosomal recessive disease caused by pathogenic variants in the dopa decarboxylase gene. Deficiency of the AADC enzyme results in combined severe reductions in monoamine neurotransmitters: dopamine, serotonin, epinephrine, and norepinephrine. This leads to widespread neurological complications affecting motor, behavioral, and autonomic function. The goal of this study was to use EHR data to identify previously undiagnosed patients who may have AADCd without available training cases for the disease. MATERIALS AND METHODS: A multiple symptom and related disease annotated dataset was created and used to train individual concept classifiers on annotated sentence data. A multistep algorithm was then used to combine concept predictions into a single patient rank value. RESULTS: Using an 8000-patient dataset that the algorithms had not seen before ranking, the top and bottom 200 ranked patients were manually reviewed for clinical indications of performing an AADCd diagnostic screening test. The top-ranked patients were 22.5% positively assessed for diagnostic screening, with 0% for the bottom-ranked patients. This result is statistically significant at P < .0001. CONCLUSION: This work validates the approach that large-scale rare-disease screening can be accomplished by combining predictions for relevant individual symptoms and related conditions which are much more common and for which training data is easier to create.

Spectrum of DDC variants causing aromatic l-amino acid decarboxylase (AADC) deficiency and pathogenicity interpretation using ACMG-AMP/ACGS recommendations.

Pathogenic variants in dopa decarboxylase (DDC), the gene encoding the aromatic l-amino acid decarboxylase (AADC) enzyme, lead to a severe deficiency of neurotransmitters, resulting in neurological, neuromuscular, and behavioral manifestations clinically characterized by developmental delays, oculogyric crises, dystonia, and severe neurologic dysfunction in infancy. Historically, therapy has been aimed at compensating for neurotransmitter abnormalities, but response to pharmacologic therapy varies, and in most cases, the therapy shows little or no benefit. A novel human DDC gene therapy was recently approved in the European Union that targets the underlying genetic cause of the disorder, providing a new treatment option for patients with AADC deficiency. However, the applicability of human DDC gene therapy depends on the ability of laboratories and clinicians to interpret the results of genetic testing accurately enough to diagnose the patient. An accurate interpretation of genetic variants depends in turn on expert-guided curation of locus-specific databases. The purpose of this research was to identify previously uncharacterized DDC variants that are of pathologic significance in AADC deficiency as well as characterize and curate variants of unknown significance (VUSs) to further advance the diagnostic accuracy of genetic testing for this condition. DDC variants were identified using existing databases and the literature. The pathogenicity of the variants was classified using modified American College of Medical Genetics and Genomics/Association for Molecular Pathology/Association for Clinical Genomic Science (ACMG-AMP/ACGS) criteria. To improve the current variant interpretation recommendations, in silico variant interpretation tools were combined with structural 3D modeling of protein variants and applied comparative analysis to predict the impact of the variant on protein function. A total of 422 variants were identified (http://biopku.org/home/pnddb.asp). Variants were identified on nearly all introns and exons of the DDC gene, as well as the 3' and 5' untranslated regions. The largest percentage of the identified variants (48%) were classified as missense variants. The molecular effects of these missense variants were then predicted, and the pathogenicity of each was classified using a number of variant effect predictors. Using ACMG-AMP/ACGS criteria, 7% of variants were classified as pathogenic, 32% as likely pathogenic, 58% as VUSs of varying subclassifications, 1% as likely benign, and 1% as benign. For 101 out of 108 reported genotypes, at least one allele was classified as pathogenic or likely pathogenic. In silico variant pathogenicity interpretation tools, combined with structural 3D modeling of variant proteins and applied comparative analysis, have improved the current DDC variant interpretation recommendations, particularly of VUSs.

Where are the males? Gender differences in undergraduates' interest in and perceptions of the genetic counseling profession.

Genetic counseling is a female-dominated field, with women comprising about 95% of the profession (Smith et al. 2009). Greater patient choice and satisfaction may be achieved by increasing the number of male counselors, but empirical evidence about the reasons for this gender imbalance is limited. In this study 190 undergraduates (110 females, 79 males, 1 unknown) in upper division bioscience courses completed a survey assessing their knowledge and perceptions of and interest in genetic counseling as a career. There were only two significant gender differences. Females indicated significantly greater interest than males in pursuing a genetic counseling career, and they rated interpersonal skills as more integral to genetic counseling than males. Multiple regression analyses of knowledge and perceptions as possible predictors of male and female interest in pursuing a genetic counseling career yielded no significant predictors of male interest. For females, there were four significant predictors: estimated salary, career characteristics, perceptions of genetic counseling as interpersonally focused, and whether they had already chosen a career. Implications for recruiting males to the profession, and research recommendations are presented.