Fig. 2

The adapted Dijkstra algorithm used in Haploflow to find fattest paths through the unitig graph. Instead of determining the shortest paths from the source to all vertices, this algorithm determines the fattest path. The fatness is initialized as 0 for all vertices, but the source and then the graph is searched using a breadth-first search and based on the fact that the fattest path from a source s to a sink t is based on the edge with the lowest coverage along this path (lines 9 to 12)