Artificial Intelligence (AI) is a broad discipline that uses powerful algorithms to emulate important aspects of human intelligence. Provided by the Industry 4.0 revolution, AI is increasingly applied in different fields from research to production. One of these fields is Materials Science and Engineering (MSE) which studies the relationships between processing, structure, properties, and performance of materials. The application of AI to MSE has triggered the invention of new materials to satisfy the demanding requirements in myriad sectors through the years. In this context, the “MSE paradigm” emerged as a framework to define these relationships supported by the available technologies at the corresponding time. This is how Polymer Matrix Composites (PMC) were synthesized. During the last years, they have turned from a futuristic solution to a necessity due to the wide range of advantages they offer concerning other conventional materials. The present work presents a modified approach to the MSE paradigm with the application of AI algorithms. An overview of the research advances from 2003 to 2019 in each fundamental link of the proposed MSE paradigm for PMC is exhibited in an organized fashion. This article must serve engineers and scientists working at the intersection of mechanical engineering, materials science and computer science to identify trendy topics in these fields. It aims to represent a starting point for developing innovative methods and proposing new research topics in the framework of the MSE paradigm powered by AI for PMC.

Oscillatory responses are ubiquitous in regulatory networks of living organisms, a fact that has led to extensive efforts to study and replicate the circuits involved. However, to date, design principles that underlie the robustness of natural oscillators are not completely known. Here we study a three-component enzymatic network model in order to determine the topological requirements for robust oscillation. First, by simulating every possible topological arrangement and varying their parameter values, we demonstrate that robust oscillators can be obtained by augmenting the number of both negative feedback loops and positive autoregulations while maintaining an appropriate balance of positive and negative interactions. We then identify network motifs, whose presence in more complex topologies is a necessary condition for obtaining oscillatory responses. Finally, we pinpoint a series of simple architectural patterns that progressively render more robust oscillators. Together, these findings can help in the design of more reliable synthetic biomolecular networks and may also have implications in the understanding of other oscillatory systems.

We show here that constant velocity steered molecular dynamics (SMD) simulations of α-helices in a vacuum present a well defined plateau in the force–extension relationship for homopolypeptides having more than (approximately) twenty residues. With the processes being far away from equilibrium, the energies strongly depend on the stretching velocity. Importantly, for a given velocity variation, the energy variation depends also on the helix sequence. Additionally, our observations show that homopolypeptides made of ten different amino acids (Ala, Cys, Gln, Ile, Leu, Met, Phe, Ser, Thr and Val) present a linear helix–coil transition.