You are here: Home


Translate NAMSE

software kick-off meeting


GPI anchor

GPI biosynthesis defects

a rare monogenic disorder



PEDIA scoring approach

contact us if you need support


Welcome to the IGSB

The IGSB just launched. We are recruiting!

Welcome to the newly founded Institute for Genomic Statistics and Bioinformatics. We are convinced that precision medicine can only be achieved when human and artificial intelligence join efforts. That's why we are currently building up a strong research team. Please have a look at some of our ongoing projects and our open positions!

Apr 10, 2018


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.

Aug 12, 2017

Neues Kompetenznetzwerk für Seltene Erkrankungen geht an den Start

Künftig soll bundesweit die Versorgung von Menschen mit unklaren Diagnosen und seltenen Erkrankungen verbessert werden. Dazu wird ein Verbund von Universitätskliniken - darunter auch das Universitätsklinikum Bonn -, der Patientenorganisation Achse und gesetzlichen Krankenkassen den von 2009 bis 2013 erarbeiteten Nationalen Aktionsplan für Menschen mit seltenen Erkrankungen („NAMSE“) umsetzen. „TRANSLATE-NAMSE“ wird mit ca. 13,4 Millionen Euro aus dem Innovationsfonds beim Gemeinsamen Bundesausschuss gefördert.

Sep 25, 2017

IGSB, IHG, and FDNA team up in Diagnostics for Rare Genetic Disorders

The Institute for Genome Statistics and Bioinformatics, IGSB, and the Institute of Human Genetics, IHG, Bonn, announce a new initiative to further precision medicine through integration with FDNA’s ( Face2Gene suite of applications. This integration is expected to dramatically increase and transform the diagnostic power of genetic testing in rare diseases, and is the first of its kind in Germany. In most instances, genetic testing of rare disease patients yields a diagnosis in only 25% of cases. This collaboration will augment genetic testing with facial analysis and artificial intelligence technologies that is expected to increase the diagnostic yield of molecular testing for thousands of rare diseases.

Jan 08, 2018

Computergestützte Gesichtsanalyse hilft der Diagnose

Bei seltenen Erkrankungen kann die computergestützte Bildauswertung von Patientenporträts die Diagnose erleichtern und deutlich verbessern. Das stellt ein internationales Wissenschaftlerteam unter Federführung des Universitätsklinikums Bonn und der Charité Universitätsmedizin Berlin anhand der so genannten GPI-Ankerstörungen unter Beweis. Die Forscher verwendeten Methoden der künstlichen Intelligenz um aus Daten zum Erbgut, der Oberflächenbeschaffenheit der Zellen und typischen Gesichtsmerkmalen Modelle der Erkrankungen zu simulieren. Die Ergebnisse können auch für andere Erkrankungen wegweisend sein und werden nun im Fachjournal „Genome Medicine“ vorgestellt.

Document Actions