Protein interactions and cellular responses are fundamental pillars of molecular systems biology. Decoding these complex signaling pathways requires advanced computational methods. One promising direction of algorithm development is using graph algorithms to identify proteins involved in signaling pathways. Despite the availability of tools, many researchers grapple with software and user experience constraints. In response, we have developed the Signaling Pathway Reconstruction Analysis Streamliner (SPRAS), a robust containerized framework that enables users to easily reconstruct signaling pathways by connecting proteins of interest within molecular interaction networks. It seamlessly integrates graph algorithms designed for pathway reconstruction with downstream visualization and clustering analysis.

In this paper, we contribute and integrate three random-walk-based algorithms to SPRAS, including one we developed and two that appear in the literature. Random walk approaches have been highly successful in predicting candidate proteins involved in a signaling pathway, and integrating them into SPRAS will greatly expand the framework’s ability for pathway reconstruction. We illustrate their importance by using the random walk algorithms now available in SPRAS to explore potential proteins involved in cell-cell fusion in flies. With the addition of these new algorithms, SPRAS will become an essential tool for unraveling the mysteries of biological interactions.