A novel computational methodology for large-scale screening of MOFs is applied to gas storage with the use of machine learning technologies. This approach is a promising trade-off between the accuracy of ab initio methods and the speed of classical approaches, strategically combined with chemical intuition. The results demonstrate that the chemical properties of MOFs are indeed predictable (stochastically, not deterministically) using machine learning methods and automated analysis protocols, with the accuracy of predictions increasing with sample size. Our initial results indicate that this methodology is promising to apply not only to gas storage in MOFs but in many other material science projects.npj Computational Materials (2017) 3:40 ; doi:10.1038/s41524-017-0045-8 INTRODUCTION Metal-organic frameworks (MOFs) or porous coordination polymers are a rapidly growing family of hybrid inorganic-organic nanoporous materials, which belong to the category of coordination polymers.1-3 These relatively new materials consist of a threedimensional periodic network, constructed from molecular building blocks, such as metal clusters and organic linkers (Fig. 1). The possible combinations of these numerous building blocks under different topologies result is an almost unlimited number of potential MOFs! Since their discovery 4 MOFs have attracted significant scientific attention due to their extraordinary properties. As "skeleton" materials, they pose very large pores and outstanding apparent surface area. If we were able to unwrap the surface of only one gram of these "very empty" materials, we could cover the area of a football court! These unique characteristics of the MOFs made them excellent candidates for catalysis and gas storage applications.MOFs have shown exceptional performance in gas storage and separation. Both useful and harmful gases can be absorbed in their pores in very large amounts. The storage of hydrogen,