The software KeSTE is an
exploratory tool developed to investigate the possible consequences of
different sampling strategies on
the structure and properties of bulk
samples, when contaminants are not randomly distributed within kernel
lots. KeSTE allows the evaluation of sampling
strategies on both simulated lots, with user-defined characteristics
(exploration of hypothetical scenarios), and on real lots, through the
analyses of data collected from a maximum of 100 increments (direct reference
to the pan-EU project KeLDA Kernel Lot Distribution
Assessment - presently on-going within ENGL
European Network of GMOs Laboratories).
This first version of KeSTE
should be considered a preliminary one, meaning that some part of the
code must still be optimized (e.g. the population
graph), and other must be still added. Furthermore, KeSTE cannot be applied
to estimate either optimal sample
numerosity, or the complete characteristics of real lots. This is not
due only to the state of development of the software, but also to the
state of development of the conceptual models behind KeSTE, which are
continuously evolving as a consequence of the use of the software by the
KeSTE Developing Team.
The use of modelling techniques
in simulating lot populations is providing
many information which are continuously leading to new insights on many
developments of the conceptual models to optimize the effectiveness of sampling. Making available KeSTE at this stage
of development has the meaning of making available now,
even if in a preliminary version, what we feel can be a powerful tool
to approach sampling optimization. The current
version of KeSTE is based on MS Excel to provide maximum flexibility to
researchers working with it. The side effect is that its user friendliness
is not optimal. As a result, the software target users are researchers
with some experience in sampling theory and
practice. When the conceptual models will be consolidated, we can foresee
the development of a derived, user-friendly version targeted to a more
applied rather than explorative evaluation of sampling.
The KeSTE Development Team
is open to suggestions and criticisms. Also,
we hope that a fruitful collaboration with research groups can start to
further develop both the software and, more importantly, the conceptual
models.
The KeSTE Development Team