Fig. 4

Adapted Dijkstra Algorithm for GFA (ADAG). All paths emanating from a given segment (i.e., the reconstructed seed segment) are enumerated, whilst keeping track of the distance traveled (in bp) and the total k-mer count. Since segments have two directions, the path search is performed from both the 3′ and 5′ ends, also keeping track of the orientation in which segments are traversed. The search continues until all segments that can be reached have been visited or when a specified limit (distance from the start segment or number of iterations) is exceeded. Once finished, the shortest paths from the end segments are found by backtracking through the graph to the starting segment. Variables: seg = the 3′ and 5′ segments ends (i.e., each segment has 2 orientations); prev = previous segment in shortest path; dist = distance from current segment to start; TK = total k-mer count from current segment to seed; visited = segment has been visited by algorithm; end = end-segment (true/false); LN = segment length; KC = segment k-mer count; NA = Not Available; Inf = Infinity