Information Technology Reference
In-Depth Information
of all the epithelial cells fully fills the well's area and that the surface area of
each cell is roughly circular, such that
A
cell
=
π
(
Δx/
2)
2
, we can compute the
number of epithelial cells in the experimental well
A
well
π
(
Δx/
2)
2
N
cells
=
(6)
113 mm
2
π
(11
=
(7)
m
/
2)
2
µ
∼
1
,
200
,
000 cells
.
(8)
For our ABM, we found that setting the well radius of
R
well
= 160 cells, which
corresponds to about 93,000 simulated cells, is sucient to accurately capture
the behaviour of a full scale simulation.
Initial Number of Virions per Epithelial Cell (
V
0
):
At time
t
=2hpost-
harvest, the time at which we begin the simulation, 635
273 virions were found
on the monolayer. Hence, we can compute the number of virions per epithelial
cell present on the monolayer at time
t
= 2 h post-harvest,
±
635 virions
N
cells
V
0
=
(9)
10
−
4
virions
/
cell
,
∼
5
.
3
×
(10)
which corresponds to the number of virions per cell at initialization time.
Fraction of Cells Initially Infected (
C
0
):
The parameter
C
0
gives the frac-
tion of cells which are initially set to the containing state. Those are the cells that
were infected during incubation with the inoculum. Staining the ALI monolayer
with viral antigen at
t
= 8 h post-harvest revealed that approximately 1.8% of
the cells contained influenza protein, i.e. were producing virions. Hence, we set
C
0
=0
.
018 in the ABM such that 1.8% of cells are set to the containing stage
at initialization time.
4
Preliminary Results
In its current implementation, the ABM has 11 parameters shown in Table 1. A
screenshot of the simulation grid is presented in Fig. 4, and Fig. 5 presents the
dynamics of the various cell states and viral titer as a function of time against
preliminary experimental data. We can see that the ABM provides a reasonable
fit to the experimental data.
