Augmented Structure Preserving Neural Networks for cell biomechanics

Cell biomechanics involve a great number of complex phenomena that are fundamental to the evolution of life itself and other associated processes, ranging from the very early stages of embryo-genesis to the maintenance of damaged structures or the growth of tumors. Given the importance of such phenomena, increasing research has been dedicated to their understanding, but the many interactions between them and their influence on the decisions of cells as a collective network or cluster remain unclear. We present a new approach that combines Structure Preserving Neural Networks, which study cell movements as a purely mechanical system, with other Machine Learning tools (Artificial Neural Networks), which allow taking into consideration environmental factors that can be directly deduced from an experiment with Computer Vision techniques. This new model, tested on simulated and real cell migration cases, predicts complete cell trajectories following a roll-out policy with a high level of accuracy. This work also includes a mitosis event prediction model based on Neural Networks architectures which makes use of the same observed features. Introduction Cell migration mechanisms are known to be present in many fundamental processes throughout the evolution of living organisms. As cells are living units that perform complex tasks, undergo constant reactions and transformations, interact with other cells and can respond to their surroundings, their migration can be presented as the result of a large number of internal and external factors. The influence of many environmental factors such as chemical gradients that can be created with biomaterials [6] or that might appear in organic environments [7], density gradients caused by cell accumulation [8], or even the presence of dead cells (which can be of interest in wound healing or tumor growth processes) [9] has been thoroughly studied. Other external factors related to cell collective movement and the tensile forces that can appear between them have also been analyzed [10, 11], with several works in this field showing that cells can use protuberances to attach themselves to other cells, which later exert pulling or pushing forces to guide their movement [12]. Despite the precision that the proposed models can achieve while explaining the relation between these factors and cell movements, there is a general lack of a global approach to the problem. Due to the possible interrelations between environmental properties, many studies simplify the problem by creating conditions where the studied gradient or feature is the dominant source of instability, and thus the main reason behind cell migration [13].