A deep learning framework for identifying essential proteins based on multiple biological information.

Background: Essential Proteins are demonstrated to exert vital functions on cellular processes and are indispensable for the survival and reproduction of the organism. Traditional centrality methods perform poorly on complex protein-protein interaction (PPI) networks. Machine learning approaches based on high-throughput data lack the exploitation of the temporal and spatial dimensions of biological information.

Photophysical and electrochemical properties of a dysprosium-zinc tetra(4-sulfonatophenyl)porphyrin complex.

A dysprosium-zinc porphyrin, [DyZn(TPPS)H3O]n (1) (TPPS = tetra(4-sulfonatophenyl)porphyrin), was prepared through solvothermal reactions and structurally characterized by single-crystal X-ray diffraction analyses. Complex 1 features a three-dimensional (3-D) porous open framework that is thermally stable up to 400 °C. Complex 1 displays a void space of 215 Å(3), occupying 9.