Y. Larriba González, C. Rueda Sabater, M. Fernández Temprano, S. Peddada
This work is motivated by applications in chronobiology where researchers are interested in the analysis of rhythmic processes such as cell cycle or circadian clock. The contribution of this work is twofold. First, a methodology is developed around a circular signal plus error model defined using order restrictions. This mathematical formulation of rhythmicity is simple, easily interpretable and flexible because of the non-parametric signal formulation. Second, the main rhythmicity-related problems in practice are solved using Order Restricted Inference tools. Apart from the well-known rhythmicity detection problem, an interesting solution to the problem of timing estimation -that appears when, due to experimental conditions, the moment at which samples were taken is unrecorded- is also given. The proposed methodology solves real problems more efficiently than the existing alternatives and can be extended to formulate many different rhythmicity-related problems in literature.
Palabras clave / Keywords: Order Restricted Inference, rhythmicity, circular space, circadian genes, oscillatory systems
Programado
Sesión 2 Premio Ramiro Melendreras
29 de mayo de 2018 17:00
Sala Cristal