Developing spatial analytical methods as open source libraries is an important endeavor to enable open and replicable science. However, despite the fact that large geospatial data and geospatial cyberinfrastructure (GeoCI) resources are becoming available, many libraries and toolkits are only initialized and designed for analytics in a desktop environment. Coupling spatial analytical functionality with big data and high-performance computing will result in immediate benefits for multidisciplinary research in terms of addressing challenging socioeconomic and environmental problems, as well as supporting remote collaboration between participants from physically distributed research groups, and assisting informed decision-making. In this article, we present the design and implementation of a general workflow to integrate state-of-the-art open source libraries with GeoCI resources. We also solve various interoperability and replicability issues that arise during the implementation process. The popular open source Python Spatial Analysis Library (PySAL) was selected to build the interoperable Web service, WebPySAL, which was then successfully integrated in GeoCI. With this integration between spatial analytics and cyberinfrastructure, the new GeoCI platform provides easy-to-use, efficient, and interactive exploratory spatial analysis functions to public users. The GeoCI capability is demonstrated through two regional economic case studies of (1) evaluating global spatial autocorrelation and identifying local clusters in the spatial pattern of median household incomes for US counties (with global and local Moran’s I statistics) and (2) modeling the space-time dynamics of per capita incomes at the state level (with spatial Markov statistics).