Franzi’s and Theo’s abstracts have been accepted at the GCB 2019 and they will give talks on Deep Learning in Chaos Game Representation for Proteins and On the semantics of unknown DNA motifs – a machine learning approach respectively. Looking forward to GCB 2019 in Heidelberg!
Dominik gave a talk on Biomedical Big Data Science at the Department of Microbiology at Rutgers University.
We recently developed an R package kaos, which can be used to build fractals from sequences of symbols. These fractals are based on the Chaos Game Representation (CGR), which is a recurrent iterative function system. While CGR has been used mainly for genome sequence encoding and classification in the context of bioinformatics, we modified it…
Our paper has been accepted in PLoS One: Canbay A, Kälsch J, Neumann U, Rau M, Hohenester S, Strucksberg K-H, Baba HA, Rust C, Bantel H, Geier A, Heider D, Sowa J-P: Non-invasive assessment of NAFLD as systemic disease – a machine learning perspective. PLoS One 2019, 14(3):e0214436. Abstract Background & aims Current non-invasive scores…
Dominik has been elected to the executive board of the GfKl – Data Science Society (http://www.gfkl.org/welcome-2/ ). He will be part of the executive board for the next three years. The Gesellschaft für Klassifikation (GfKl) Data Science Society was founded in 1977 to develop research activities related to the classification and ordering of data, to the…
Our paper has been accepted in BioData Mining: Spänig S, Heider D: Encodings and models for antimicrobial peptide classification for multi-resistant pathogens. BioData Mining 2019, 12:7. (Link) Abstract Antimicrobial peptides (AMPs) are part of the inherent immune system. In fact, they occur in almost all organisms including, e.g., plants, animals, and humans. Remarkably, they show…
Our research project application Die Anwendung von modernen Informationstechnologien in der Neurorehabilitation von Patienten mit erworbener Hirnschädigung (project no. 191) with a total volume of EUR 791,573 has just been granted for the next 3 years. The project was submitted to the european INTERREG V-A call. The lead is done by University of West Bohemia,…
Marta’s, Sebastian’s and Theo’s abstracts have been accepted at the ECDA 2019 and they will give talks on A novel privacy-preserving software framework for federated machine learning in clinical settings and A comparative study on peptide encodings for biomedical classification and Illuminating interactions in microbial lake ecology using General Learning Chain Ensembles respectively.
Our new article has been accepted in BMC Bioinformatics: Sperlea T, Füser S, Boenigk J, Heider D: SEDE-GPS: socio-economic data enrichment based on GPS information. BMC Bioinformatics 2018, 19(Suppl 15):440. (Link) Abstract Background Microbes are essentail components of all ecosystems because they drive many biochemical processes and act as primary producers. In freshwater ecosystems, the…
Our new article has been accepted in Bioinformatics: Schwarz J and Heider D: GUESS: Projecting Machine Learning Scores to Well-Calibrated Probability Estimates for Clinical Decision Making. Bioinformatics 2018, in press. Abstract Motivation: Clinical decision-support-systems (CDSS) have been applied in numerous fields, ranging from cancer survival towards drug resistance prediction. Nevertheless, CDSS typically have a caveat:…
