Abstract
Crystal structure prediction has been applied successfully to discover novel materials and associated physical properties. It was well tested computationally to reproduce existing compounds and even it became very powerful to find new materials at extreme conditions such as pressure, temperature, and confined geometries. In pressure dimension, most materials undergoes phase transition, and in these days we are witnessing discovery of several interesting physical properties such as room temperature superconductivity, superionicity in ice and hydrous minerals, electride formation, and unexpected superoxide, which are related to structural phase transitions. Crystal structure prediction based on density functional theory plays a critical role in aforementioned studies as in silico experiments. I will present our theoretical predictions, which lead to experimental synthesis at high pressure.