Berlin 2018 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

DY: Fachverband Dynamik und Statistische Physik

DY 63: Statistical Physics of Biological Systems II (joint session BP/DY)

DY 63.4: Vortrag

Donnerstag, 15. März 2018, 15:45–16:00, H 2013

Asymmetric Link detection via a generalized ESABO approach — •Jens Christian Claussen — Computational Systems Biology, Jacobs University Bremen, Germany

Mutualisms in biological populations are widespread from bacteria to mammals. Mutualistic interactions can be positive (synergistic) or negative. Often even in microbial data the number of available samples is marginally sufficient to allow for detection of interactions, especially for the low-abundance species that may carry important information in clinical context. The recently introduced ESABO method (PloS Comp Biol 13: e1005361 (2017)) utilizes an information-theoretic approach to evaluate binarized abundances and was demonstrated to detect interaction links that were not apparent in the classical correlation analyses. ESABO provides high (resp. low) scores if joint occurence is higher (resp. lower that in surrogate data. As so far, ESABO concludes on negative interactions when co-occurrence is lower than expected. However, this can be due to asymmetric (unidirectional parasitic) interaction in any of two directions, or due to symmetric interactions. Here we generalize the ESABO method to analyze co-abundance data resolving for asymmetry between the interactions.