Fast and Optimal Sequence-to-Graph Alignment Guided by Seeds

We present a novel A* seed heuristic enabling fast and optimal sequence-to-graph alignment, guaranteed to minimize the edit distance of the alignment assuming non-negative edit costs. We phrase optimal alignment as a shortest path problem and solve it by instantiating the A* algorithm with our novel seed heuristic. The key idea of the seed heuristic is to extract seeds from the read, locate them in the reference, mark preceding reference positions by crumbs, and use the crumbs to direct the A* search. We prove admissibility of the seed heuristic, thus guaranteeing alignment optimality. Our implementation extends the free and open source AStarix aligner and demonstrates that the seed heuristic outperforms all state-of-the-art optimal aligners including GraphAligner, Vargas, PaSGAL, and the prefix heuristic previously employed by AStarix. Specifically, we achieve a consistent speedup of >60x on both short Illumina reads and long HiFi reads (up to 25kbp), on both the E. coli linear reference genome (1Mbp) and the MHC variant graph (5Mbp). Our speedup is enabled by the seed heuristic consistently skipping >99.99% of the table cells that optimal aligners based on dynamic programming compute.