Dataset Information

Advances in computer-assisted syndrome recognition by the example of inborn errors of metabolism.

ABSTRACT: Significant improvements in automated image analysis have been achieved in recent years and tools are now increasingly being used in computer-assisted syndromology. However, the ability to recognize a syndromic facial gestalt might depend on the syndrome and may also be confounded by severity of phenotype, size of available training sets, ethnicity, age, and sex. Therefore, benchmarking and comparing the performance of deep-learned classification processes is inherently difficult. For a systematic analysis of these influencing factors we chose the lysosomal storage diseases mucolipidosis as well as mucopolysaccharidosis type I and II that are known for their wide and overlapping phenotypic spectra. For a dysmorphic comparison we used Smith-Lemli-Opitz syndrome as another inborn error of metabolism and Nicolaides-Baraitser syndrome as another disorder that is also characterized by coarse facies. A classifier that was trained on these five cohorts, comprising 289 patients in total, achieved a mean accuracy of 62%. We also developed a simulation framework to analyze the effect of potential confounders, such as cohort size, age, sex, or ethnic background on the distinguishability of phenotypes. We found that the true positive rate increases for all analyzed disorders for growing cohorts (n = [10...40]) while ethnicity and sex have no significant influence. The dynamics of the accuracies strongly suggest that the maximum distinguishability is a phenotype-specific value, which has not been reached yet for any of the studied disorders. This should also be a motivation to further intensify data sharing efforts, as computer-assisted syndrome classification can still be improved by enlarging the available training sets.

SUBMITTER: Pantel JT

PROVIDER: S-EPMC5959962 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Glutamate is involved in a variety of metabolic pathways. We reviewed the literature on genetic defects of enzymes that directly metabolise glutamate, leading to inborn errors of glutamate metabolism. Seventeen genetic defects of glutamate metabolising enzymes have been reported, of which three were only recently identified. These 17 defects affect the inter-conversion of glutamine and glutamate, amino acid metabolism, ammonia detoxification, and glutathione metabolism. We provide an overview of the clinical and biochemical phenotypes of these rare defects in an effort to ease their recognition. By categorising these by biochemical pathway, we aim to create insight into the contributing role of deviant glutamate and glutamine levels to the pathophysiology. For those disorders involving the inter-conversion of glutamine and glutamate, these deviant levels are postulated to play a pivotal pathophysiologic role. For the other IEM however-with the exception of urea cycle defects-abnormal glutamate and glutamine concentrations were rarely reported. To create insight into the clinical consequences of disturbed glutamate metabolism-rather than individual glutamate and glutamine levels-the prevalence of phenotypic abnormalities within the 17 IEM was compared to their prevalence within all Mendelian disorders and subsequently all disorders with metabolic abnormalities notated in the Human Phenotype Ontology (HPO) database. For this, a hierarchical database of all phenotypic abnormalities of the 17 defects in glutamate metabolism based on HPO was created. A neurologic phenotypic spectrum of developmental delay, ataxia, seizures, and hypotonia are common in the inborn errors of enzymes in glutamate metabolism. Additionally, ophthalmologic and skin abnormalities are often present, suggesting that disturbed glutamate homeostasis affects tissues of ectodermal origin: brain, eye, and skin. Reporting glutamate and glutamine concentrations in patients with inborn errors of glutamate metabolism would provide additional insight into the pathophysiology.

Project description:Glutathione is a tripeptide composed of glutamate, cysteine and glycine. Glutathione is present in millimolar concentrations in most mammalian cells and it is involved in several fundamental biological functions, including free radical scavenging, detoxification of xenobiotics and carcinogens, redox reactions, biosynthesis of DNA, proteins and leukotrienes, as well as neurotransmission/neuromodulation. Glutathione is metabolised via the gamma-glutamyl cycle, which is catalyzed by six enzymes. In man, hereditary deficiencies have been found in five of the six enzymes. Glutathione synthetase deficiency is the most frequently recognized disorder and, in its severe form, it is associated with hemolytic anemia, metabolic acidosis, 5-oxoprolinuria, central nervous system (CNS) damage and recurrent bacterial infections. Gamma-glutamylcysteine synthetase deficiency is also associated with hemolytic anemia, and some patients with this disorder show defects of neuromuscular function and generalized aminoaciduria. Gamma-glutamyl transpeptidase deficiency has been found in patients with CNS involvement and glutathionuria. 5-Oxoprolinase deficiency is associated with 5-oxoprolinuria but without a clear association with other symptoms. Dipeptidase deficiency has been described in one patient. All disorders are very rare and inherited in an autosomal recessive manner. Most of the mutations are leaky so that many patients have residual enzyme activity. Diagnosis is made by measuring the concentration of different metabolites in the gamma-glutamyl cycle, enzyme activity and in glutathione synthetase and gamma-glutamylcysteine synthetase deficiency, also by mutation analysis. Prenatal diagnosis has been preformed in glutathione synthetase deficiency. The prognosis is difficult to predict, as few patients are known, but seems to vary significantly between different patients. The aims of the treatment of glutathione synthesis defects are to avoid hemolytic crises and to increase the defense against reactive oxygen species. No treatment has been recommended for gamma-glutamyl transpeptidase, 5-oxoprolinase and dipeptidase deficiency.

Dataset Information

Advances in computer-assisted syndrome recognition by the example of inborn errors of metabolism.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets