RESUMO
A probabilistic discrete model for 2D protein crystal growth is presented. This model takesinto account the available space and can describe growing processes of a different nature due to theversatility of its parameters, which gives the model great flexibility. The accuracy of the simulation istested against a real recrystallization experiment, carried out with the bacterial protein SbpA fromLysinibacillus sphaericus CCM2177, showing high agreement between the proposed model and theactual images of the crystal growth. Finally, it is also discussed how the regularity of the interface(i.e., the curve that separates the crystal from the substrate) affects the evolution of the simulation.
RESUMO
A new method for estimating the contact point in AFM force curves, based on a local regression algorithm, is presented. The main advantage of this method is that can be easily implemented as a computer algorithm and used for a fully automatic detection of the contact points in the approach force curves on living cells. The estimated contact points have been compared to those obtained by other published methods, which were applied either for materials with an elastic response to indentation forces or for experiments at high loading rates. We have found that the differences in the values of the contact points estimated with three different methods were not statistically significant and thus the algorithm is reliable. Also, we test the convenience of the algorithm for batch-processing by computing the contact points of a force curve map of 625 (25×25) curves.