diffuStats

"diffuStats" is an R package that facilitates label propagation and diffusion processes over biological networks. This methodology is widely employed in computational biology to contextualize molecular entities and prioritize novel candidates within a research area. The diffusion process can be conceptualized in various ways, involving choices around the graph kernel, score definitions, and the incorporation of posterior statistical normalization. These choices significantly influence the outcome of the diffusion analysis.

diffuStats addresses this variability by providing a comprehensive graph kernels and diffusion scores suite. Additionally, it offers a parallel permutation analysis for normalized scores, simplifying the computation of these scores and enabling their systematic benchmarking to identify the most effective diffusion process configuration for specific research objectives.