Page 313 - Data Science Algorithms in a Week

294 Jose M. Prieto

In this chapter, we present work showing the potential of ANNs as a tool to
accomplish the prediction of bioactivities for very complex chemical entities such as
natural products, and suggest strategies on the selection of inputs and conditions for the
in silico experiments. We highlight the limitations of the scientific data so far available -
that suffer from little standardization of the experimental conditions and disparate choice
of reference drugs - as well as the shortfalls of some popular assay methods which limit
the accuracy of the ANNs prediction.
From the number and range of scientific output published, we cannot see that this
tool has been used to its full potential in the pharmaceutical, cosmetic or food industry.
There is a need to form multidisciplinary groups to generate high quality experimental
data and process them to exploit the full potential offered by ANNs. The author foresees
a future where omics technology and systems biology will feed data in real time cloud-
based ANNs to build increasingly accurate predictions and classifications of biochemical
activities of complex natural products facilitating their rational clinical use to improve
healthcare and food safety worldwide.

REFERENCES

Agatonovic-Kustrin, S., Beresford, R. (2000). Basic concepts of artificial neural network
(ANN) modeling and its application in pharmaceutical research. J Pharm Biomed
Anal, 22, 717-727.
Agatonovic-Kustrin, S. & Loescher, C. (2013). Qualitative and quantitative high
performance thin layer chromatography analysis of Calendula officinalis using high
resolution plate imaging and artificial neural network data modelling. Anal Chim
Acta, 798, 103-108.
Asnaashari, E., Asnaashari, M., Ehtiati, A., & Farahmandfar, R. (2015). Comparison of
adaptive neuro-fuzzy inference system and artificial neural networks (MLP and RBF)
for estimation of oxidation parameters of soybean oil added with curcumin. J Food
Meas Char, 9, 215-224.
Asnaashari, M., Farhoosh, R., & Farahmandfar, R. (2016), Prediction of oxidation
parameters of purified Kilka fish oil including gallic acid and methyl gallate by
adaptive neuro-fuzzy inference system (ANFIS) and artificial neural network. J Sci
Food Agr, 96, 4594-4602.
Batchelor, B. (1993). Automated inspection of bread and loaves. Int Soc Opt Eng (USA),
2064, 124-134.
Bhotmange, M. & Shastri, P. (2011). Application of Artificial Neural Networks to Food
and Fermentation Technology. In: Suzuki K Artificial Neural Networks - Industrial
and Control Engineering Applications, Shanghai, InTech, 2011; 201-222.