There exist complex mathematical problems that are important in real world applications such as weather prediction, molecular modelling, network route optimization and more. In general, such problems are solved using supercomputers with higher computing efficiency but this also consumes high energy along with high production and maintenance cost. Network-based biocomputation (NBC) is an alternate computing approach, now at development stage, that can perform parallel computing in a highly energy efficient manner. Actin and myosin constitute one type of molecular motor system that has been utilized for the development of NBC. These proteins are key components in the sarcomere, the smallest functional unit of muscle and their interactions that underlie muscle contraction are powered by the cellular fuel adenosine triphosphate (ATP). To solve larger complex problems using actin-myosin based NBC, factors such as maintained biological function and longevity of operation are essential for practical relevance. In this thesis, the in vitro motility assay (IVMA) has been used as a central method to study actomyosin function and its operation within NBC devices. In the IVMA, actin filaments are propelled by myosin motors that are immobilized on functionalized surfaces in a flow cell. With the aim to improve motile fraction by reducing the interaction between actin and non-functional motor heads in the IVMA, two known methods were quantitatively compared in paper I, the affinity purification and the blocking actin method. Both approaches significantly improved the motile fraction to above 90% but affinity purification, due to the presence of ATP during incubation, induced significant reduction in sliding velocity, not seen with blocking actin. In paper III, critical parameters in the actomyosin IVMA system were investigated allowing us to extensively improve function and longevity, including: biocompatibility of flow cell components, effects of air exposure with oxygen scavenging and nanofabrication parameters such as plasma etching type and time, process of resist development, and surface silanization time. The above developments together with optimized network encoding of the problems enabled us (paper IV) to solve four instances of 3-SAT problem encoded in NBC with 99% probability of satisfiability. In parallel, (paper II) a method have been developed to recycle the surfaces with immobilized motor proteins by treatment of proteinase-K enzyme and detergent. This will allow re-cycling of advanced NBC chips. Finally, with aim to develop programmable gating for NBC, attempts have been made towards the integration of engineered light sensitive myosin XI motors with nanofabricated devices made up of Au/SiO₂, SiO₂/polymer and glass/polymer (paper V). In addition important factors such as standardized motor density, limiting of air exposure and longevity function have been optimized in the use of light sensitive motors.

Overall, this thesis reports critical insights for the upscaling of actomyosin based NBC. Described results, are also useful for the development of actomyosin based nanotechnological applications such as biosensing or diagnostics and other fundamental studies based on single molecule or drug testing.