Sammendrag
Astrocytes play an important part in the central nervous system, but their role and structure are not yet fully understood. Previous studies of cultured astrocytes and astrocytes in brain slices have concluded that the sizes of Ca2+ events, the main signalling pathway of astrocytes, are power-law distributed. This thesis argues that a more comprehensive modeling approach is necessary and develops a variant of a gamma process model fitted on in-vivo measurements of Ca2+. Though the model type fits well, it is difficult to estimate the importance of this modeling approach, both due to the idiosyncratic nature of Ca2+ measurement methods and the lack of any comparable efforts in statistical modeling of Ca2+ events. Doing inferecence within our framework we find no evidence of what we call a relative refractory period after Ca2+ events in astrocytes, though we do find a robust positive relation between time since the last Ca2+ event and the size of the next. We also find weak evidence that supports a negative relation between the size of a Ca2+ event and the size of the following Ca2+ event.