Magni Martens 1,2, Siren R. Veflingstad  3,4,  Erik Plahte 3, Dominique Bertrand 5, Asgeir Nilsen 1, Harald Martens 1,2,3

1) Nofima Mat, Ås, Norway
2) University of Copenhagen, Faculty of Life Sciences, Denmark
3) Centre for Integrative Genetics, Norwegian University of Life Sciences, Ås,   Norway
4) Systems Biology Centre, University of Warwick, United Kingdom
5) INRA, Nantes, France

Scientists use their eyes, often without recognising it. Valuable visual information can be systemised and made operational as efficient tools in scientific measurements, modelling and understanding of complex biological systems, as shown in a variety of literature examples spanning from sensory evaluation of fMRI images of potatoes to studies of visual perception of moving pictures.

The present abstract refers to the sensory part of a study of a mathematical model of a biological cell differentiation process in a system of interacting cells in a two-dimensional lattice. The output of the model was presented as two-dimensional images showing the distribution of protein concentrations over the lattice, and the challenge is to find approaches in characterising measures for ‘what is going on’ which we can translate into data handling enhancing cognitive understanding. A sensory scientific approach is to develop a vocabulary of images revealed from the model.

Purpose of this abstract: Outline a sensory approach for revealing, systemising and interpreting image patterns being reliable and valid in communication across various scientific fields.

Methods:

• Simulation studies of a mathematical model according to a suitable experimental design, using ‘delta-notch’ parameters as example and sorting out 64 images for sensory evaluation
• Sensory evaluation of the images (expert focus group and descriptive sensory analysis)
• Mathematical, computer-derived image analysis of the same patterns
• Partial least squares regression for relating the various data sets

Results and conclusions:

• Sensory evaluation of images provided an efficient approach for characterising complex patterns, which furthermore showed clear relationships to the mathematical modelling parameters and to the computer-derived image data.
• A preliminary sensory morphological wheel was developed. This may form a future common ground for interdisciplinary discourses and may even contribute to an increased recognition of the usefulness of human senses in natural sciences in general.