Statistical Study of Cellular Aging

Julien  Guyon, Ariane  Bize, Grégory  Paul, Eric  Stewart, Jean-François
Delmas and Francois Taddéi.

Whereas  aging  is  obvious  in  macroscopic organisms,  it  is  not  in
single-celled ones,  such as the model  organism E. Coli,  where one has
the  best chances of  describing and  quantifying the  molecular process
involved.  To determine  if E.  Coli  experiences aging  related to  the
inheritance  of  the  old pole,  E.  J.  Stewart  and al.   followed  94
individual exponentially  growing cells through up  to nine generations,
determining  the   complete  lineage  and   the  growth  rate   of  each
cell.  Averaging over  these 94  experiments  leads the  authors to  the
conclusion that the two  supposedly identical cells produced during cell
division are functionally asymmetric.

This  preliminary study,  avoiding inter-experiment  averaging,  aims at
reasoning  experiment-wise and  describing the  structure  of dependency
within a  lineage of E. Coli.  We propose a  single-experiment model for
the dynamics of  such a lineage of  growth rates. We fit it  to the data
and perform experiment-wise  tests which lead us to  accept that the old
pole progeny  cell experiences  slowed growth rate  and hence  should be
considered   an   aging    parent   repeatedly   producing   rejuvenated
offspring.  Eventually, as  a first  step towards  answering statistical
questions about markovian models on  binary trees, we run simulations of
the model and conjecture results. At this preliminary stage, we have not
used the results  on bifurcative autoregressive models which  seem to be
an appropriate background for the gaussian model.