the components in the various culture medium formulations and the

initial cell concentration were the input variables and the cell concentration at any instance
was the output variable of the FBN model.
2. Materials and methods
2.1. Database
Raw data from 426 experiments originally carried out for a genetic
algorithm-based search for an optimal culture medium for the marine
dinoflagellate K. veneficum were used [5]. This database comprised 6 different batches
(generations) of 71 experiments each. Each experiment
had a different composition of the culture medium and was carried
out in triplicate.
The above database was supplemented with new batch experiments
carried out using extreme formulations of culture media, that is formulations with
exceptionally high or nil amounts of certain constituents.
The responses of these extreme formulations and the other data were
used in training the FBN to enhance its predictive capability. The extreme dataset spanned
three levels of the initial cell concentration in
the culture medium (i.e. 1 × 104, 2 × 104 and 3.3 × 104 cells mL-1)
and the following in independent experiments: (1) the L1 medium
with nitrate concentration ranging from 0 to 882 μM; (2) the L1 medium without phosphate;
(3) the culture medium formulation “MAX”
that used the maximum concentrations of all the nutrients as shown
in Table 1; and (4) seawater without any added nutrients.
Inocula were grown as batches under a 12:12 h light–dark cycle at
18 ± 1 °C. Filter-sterilized (0.22 μm membrane filter) L1 medium prepared using natural
Mediterranean Sea water was used as the control
medium in all experiments. Cells in their exponential growth phase
were used for inoculation. Multiwell plates (2-mL per well) were used
as growth vessels. Cultures were illuminated from the top using Philips
TLD 36 W/54 fluorescent lamps at an average irradiance of 200 μmol
photons m-2 s-1. All experiments were carried out in triplicate.
Growth was measured by cell counts every three days

The ANN input variables consisted of a k × s matrix with s (=

1398)
columns of the initial concentrations of 25 nutrients selected from
those
commonly found in seawater-based algal culture media [5]; the
initial
cell concentration (C0); and the time of measurement (t). The
ranges
of the values for the inputs used in establishing the ANN model
and
the output data are shown in Table 1. The target output variable
was
the dimensionless cell concentration (Ct/C0) in the culture at any
time
t. Here, Ct is the cell concentration at time t. As the values of the
input
variables ranged up to eight orders of magnitude (Table 1), the
input
and target variables were normalized using the following
equations:

the components in the various culture medium formulations and the

initial cell concentration were the input variables and the cell concentration at any instance
was the output variable of the FBN model.
2. Materials and methods
2.1. Database
Raw data from 426 experiments originally carried out for a genetic
algorithm-based search for an optimal culture medium for the marine
dinoflagellate K. veneficum were used [5]. This database comprised 6 different batches
(generations) of 71 experiments each. Each experiment
had a different composition of the culture medium and was carried
out in triplicate.
The above database was supplemented with new batch experiments
carried out using extreme formulations of culture media, that is formulations with
exceptionally high or nil amounts of certain constituents.
The responses of these extreme formulations and the other data were
used in training the FBN to enhance its predictive capability. The extreme dataset spanned
three levels of the initial cell concentration in
the culture medium (i.e. 1 × 104, 2 × 104 and 3.3 × 104 cells mL-1)
and the following in independent experiments: (1) the L1 medium
with nitrate concentration ranging from 0 to 882 μM; (2) the L1 medium without phosphate;
(3) the culture medium formulation “MAX”
that used the maximum concentrations of all the nutrients as shown
in Table 1; and (4) seawater without any added nutrients.
Inocula were grown as batches under a 12:12 h light–dark cycle at
18 ± 1 °C. Filter-sterilized (0.22 μm membrane filter) L1 medium prepared using natural
Mediterranean Sea water was used as the control
medium in all experiments. Cells in their exponential growth phase
were used for inoculation. Multiwell plates (2-mL per well) were used
as growth vessels. Cultures were illuminated from the top using Philips
TLD 36 W/54 fluorescent lamps at an average irradiance of 200 μmol
photons m-2 s-1. All experiments were carried out in triplicate.
Growth was measured by cell counts every three days

The ANN input variables consisted of a k × s matrix with s (=

1398)
columns of the initial concentrations of 25 nutrients selected from
those
commonly found in seawater-based algal culture media [5]; the
initial
cell concentration (C0); and the time of measurement (t). The
ranges
of the values for the inputs used in establishing the ANN model
and
the output data are shown in Table 1. The target output variable
was
the dimensionless cell concentration (Ct/C0) in the culture at any
time
t. Here, Ct is the cell concentration at time t. As the values of the
input
variables ranged up to eight orders of magnitude (Table 1), the
input
and target variables were normalized using the following
equations:
the components in the various culture medium formulations and the
initial cell concentration were the input variables and the cell concentration at any instance
was the output variable of the FBN model.
2. Materials and methods
2.1. Database
Raw data from 426 experiments originally carried out for a genetic
algorithm-based search for an optimal culture medium for the marine
dinoflagellate K. veneficum were used [5]. This database comprised 6 different batches
(generations) of 71 experiments each. Each experiment
had a different composition of the culture medium and was carried
out in triplicate.
The above database was supplemented with new batch experiments
carried out using extreme formulations of culture media, that is formulations with
exceptionally high or nil amounts of certain constituents.
The responses of these extreme formulations and the other data were
used in training the FBN to enhance its predictive capability. The extreme dataset spanned
three levels of the initial cell concentration in
the culture medium (i.e. 1 × 104, 2 × 104 and 3.3 × 104 cells mL-1)
and the following in independent experiments: (1) the L1 medium
with nitrate concentration ranging from 0 to 882 μM; (2) the L1 medium without phosphate;
(3) the culture medium formulation “MAX”
that used the maximum concentrations of all the nutrients as shown
in Table 1; and (4) seawater without any added nutrients.
Inocula were grown as batches under a 12:12 h light–dark cycle at
18 ± 1 °C. Filter-sterilized (0.22 μm membrane filter) L1 medium prepared using natural
Mediterranean Sea water was used as the control
medium in all experiments. Cells in their exponential growth phase
were used for inoculation. Multiwell plates (2-mL per well) were used
as growth vessels. Cultures were illuminated from the top using Philips
TLD 36 W/54 fluorescent lamps at an average irradiance of 200 μmol
photons m-2 s-1. All experiments were carried out in triplicate.
Growth was measured by cell counts every three days

The ANN input variables consisted of a k × s matrix with s (=

1398)
columns of the initial concentrations of 25 nutrients selected from
those
commonly found in seawater-based algal culture media [5]; the
initial
cell concentration (C0); and the time of measurement (t). The
ranges
of the values for the inputs used in establishing the ANN model
and
the output data are shown in Table 1. The target output variable
was
the dimensionless cell concentration (Ct/C0) in the culture at any
time
t. Here, Ct is the cell concentration at time t. As the values of the
input
variables ranged up to eight orders of magnitude (Table 1), the
input
and target variables were normalized using the following
equations:
n the FBN training process, MSE, the mean squared error between
the experimental data and the corresponding predicted data, was calculated and then
propagated back through the network in each cycle. The
algorithm adjusted the weightings between the input layer, the hidden
layer and output neurons in order to reduce the error. This procedure
was repeated until the error between the experimental data and the
predicted values satisfied the typical criteria for Bayesian regularization.
The weights (Wij and Zj) and bias (b1j and b2) values in Eq. (4) were obtained after
the training phase of the FBN model had been comple
n the FBN training process, MSE, the mean squared error between
the experimental data and the corresponding predicted data, was calculated and then
propagated back through the network in each cycle. The
algorithm adjusted the weightings between the input layer, the hidden
layer and output neurons in order to reduce the error. This procedure
was repeated until the error between the experimental data and the
predicted values satisfied the typical criteria for Bayesian regularization.
The weights (Wij and Zj) and bias (b1j and b2) values in Eq. (4) were obtained after
the training phase of the FBN model had been comple
n the FBN training process, MSE, the mean squared error between
the experimental data and the corresponding predicted data, was calculated and then
propagated back through the network in each cycle. The
algorithm adjusted the weightings between the input layer, the hidden
layer and output neurons in order to reduce the error. This procedure
was repeated until the error between the experimental data and the
predicted values satisfied the typical criteria for Bayesian regularization.
The weights (Wij and Zj) and bias (b1j and b2) values in Eq. (4) were obtained after
the training phase of the FBN model had been comple
Fig. 3 compares the model predicted values of the Ct/C0 and the measured data. The
comparisons are shown for the training runs, the test
runs and the combined (training and test) runs. There was excellent
agreement between the predicted and the measured data in the training
runs (Fig. 3). In the validation test runs, the data were event distributed
around the parity line (y = x) nearly all the predicted and the measured
values agreed within about ±75% of the measured data (R = 0.9138).
Using the combined data sets of the training runs that the test runs,
the measured and the predicted values of Ct/C0 were strongly correlated
(R N 0.984) and evenly distributed around the parity line (Fig. 3). This
confirmed the ability of the FBN model to predict new data. This is clearly seen in
Fig. 4, where the predicted and the observed values of Ct/C0
versus the observation order are shown.
Fig. 3 compares the model predicted values of the Ct/C0 and the measured data. The
comparisons are shown for the training runs, the test
runs and the combined (training and test) runs. There was excellent
agreement between the predicted and the measured data in the training
runs (Fig. 3). In the validation test runs, the data were event distributed
around the parity line (y = x) nearly all the predicted and the measured
values agreed within about ±75% of the measured data (R = 0.9138).
Using the combined data sets of the training runs that the test runs,
the measured and the predicted values of Ct/C0 were strongly correlated
(R N 0.984) and evenly distributed around the parity line (Fig. 3). This
confirmed the ability of the FBN model to predict new data. This is clearly seen in
Fig. 4, where the predicted and the observed values of Ct/C0
versus the observation order are shown.
Fig. 3 compares the model predicted values of the Ct/C0 and the measured data.
The comparisons are shown for the training runs, the test
runs and the combined (training and test) runs. There was excellent
agreement between the predicted and the measured data in the training
runs (Fig. 3). In the validation test runs, the data were event distributed
around the parity line (y = x) nearly all the predicted and the measured
values agreed within about ±75% of the measured data (R = 0.9138).
Using the combined data sets of the training runs that the test runs,
the measured and the predicted values of Ct/C0 were strongly correlated
(R N 0.984) and evenly distributed around the parity line (Fig. 3). This
confirmed the ability of the FBN model to predict new data. This is clearly seen
in Fig. 4, where the predicted and the observed values of Ct/C0
versus the observation order are shown.
How the limiting nutrients and those provided in excess interact to
produce a growth outcome is a relevant consideration in the modeling
of microalgal growth dynamics. Fig. 7 compares the FBN predicted
growth profiles with the experimental data for some quite different formulations
of the culture media. The compositions of the relevant media
are specified in Table 2. The cases represented in Fig. 7A include the following:
(1) an optimal culture medium composition (N66G6) [5]; (2) a
poorly balanced medium composition (N51G4); (3) the L1 medium
that is commonly used for marine dinoflagellate microalgae; (4) the
L1 medium without nitrates; and (5) the L1 medium without phosphates. Fig.
7B provides the model predicted growth profiles and the
measured data for the optimal medium composition N66G6, but with
different initial concentrations of nitrate. For comparison, the measured
data and the predicted growth profiles are provided also for media of
extreme compositions (Fig. 7B), that is seawater without any supplements and
the extremely rich medium MAX (Table 2). In all cases in
Fig. 7, the m

Models based on artificial neural networks can successfully predict

algae growth profiles for an extremely broad range of concentrations
of the nutrients in media with a large number of components. Such
models do not require any mechanistic understanding of the algal metabolism. A knowledge
of the initial cell concentration and the initial
composition of the culture medium in a batch operation is sufficient
for a reliable prediction of the growth profiles. Unlike the conventional
growth models, the FBN based modeling approach is highly flexible and
versatile. Results suggest that micronutrients, despite having concentrations that are much
lower than those of the macronutrients, do substantially influence the microalgal growth. In
the present work, the focus
was on identifying a suitable composition of the culture medium for a
typically used set of the environmental conditions. In the future, it
may be possible to include in an FBN model the ability to describe the
effects of the environmental factors (e.g. pH, temperature, irradiance
level) as well as the nutritional factors, on algal growth