The fragment network provides a convenient way to filter-out compounds that are dissimilar to the input hit(s). Overall, this search algorithm requires a compound input and 3 parameters: 1- the number of graph traversals (hops), 2- number of changes in heavy atom count (hac), 3- number of changes in ring atoms counts (rac). Please, read the reference (Hall, Murray and Verdonk, 2017) for the specifics of the methods.

{"references": ["Hall, R., Murray, C. and Verdonk, M. (2017). The Fragment Network: A Chemistry Recommendation Engine Built Using a Graph Database. Journal of Medicinal Chemistry, 60(14), pp.6440-6450."]}