MICA: Multiple interval-based curve alignment

Abstract

MICA enables the automatic synchronization of discrete data curves. To this end, characteristic points of the curves’ shapes are identified. These landmarks are used within a heuristic curve registration approach to align profile pairs by mapping similar characteristics onto each other. In combination with a progressive alignment scheme, this enables the computation of multiple curve alignments.Multiple curve alignments are needed to derive meaningful representative consensus data of measured time or data series. MICA was already successfully applied to generate representative profiles of tree growth data based on intra-annual wood density profiles or cell formation data.The MICA package provides a command-line and graphical user interface. The R interface enables the direct embedding of multiple curve alignment computation into larger analyses pipelines. Source code, binaries and documentation are freely available at https://github.com/BackofenLab/MICA Keywords: Curve alignment, Landmark registration, Global alignment, Progressive alignment