The problem of determining the minimum cost hypothetical ancestral sequences for a given cladogram is known to be NP-complete (Wang and Jiang, 1994). Traditionally, point estimations of hypothetical ancestral sequences have been used to gain heuristic, upper bounds on cladogram cost. These include procedures with such diverse approaches as non-additive optimization of multiple sequence alignment, direct optimization (Wheeler, 1996), and fixed-state character optimization (Wheeler, 1999). A method is proposed here which, by extending fixed-state character optimization, replaces the estimation process with a search. This form of optimization examines a diversity of potential state solutions for cost-efficient hypothetical ancestral sequences and can result in greatly more parsimonious cladograms. Additionally, such an approach can be applied to other NP-complete phylogenetic optimization problems such as genomic break-point analysis.

中文翻译：

---

基于搜索的优化。

为给定的克拉德图确定最小成本假想祖先序列的问题已知是NP完全的（Wang和Jiang，1994）。传统上，假想祖先序列的点估计已用于获得启发式克拉德图成本的上限。这些包括具有多种方法的程序，例如多序列比对的非加性优化，直接优化（Wheeler，1996）和固定状态字符优化（Wheeler，1999）。在此提出了一种方法，该方法通过扩展固定状态字符优化来用搜索代替估计过程。这种优化形式检查了具有成本效益的假想祖先序列的各种潜在状态解决方案，并且可能导致简明的简图。另外，