Synthesizing large-scale species trees using the strict consensus approach

Abstract

Supertree problems are a standard tool for synthesizing large-scale species trees from a given collection of gene trees under some problem-specific objective. Unfortunately, these problems are typically NP-hard, and often remain so when their instances are restricted to rooted gene trees sampled from the same species. While a class of restricted supertree problems has been effectively addressed by the parameterized strict consensus approach, in practice, most gene trees are unrooted and sampled from different species. Here, we overcome this stringent limitation by describing efficient algorithms that are adopting the strict consensus approach to also handle unrestricted supertree problems. Finally, we demonstrate the performance of our algorithms in a comparative study with classic supertree heuristics using simulated and empirical data sets.

Exact median-tree inference for unrooted reconciliation costs

“…Consequently, analyzing the accuracy of such heuristics is of critical importance. Recent algorithmic advances in exact computing solutions for median tree reconciliation problems made it possible to evaluate the accuracy of local search heuristics addressing these problems [ 31 , 50 ]. However, these algorithmic advances address only tree reconciliation problems for rooted input trees.…”

“…Moreover, this landscape has typically numerous local optima that can easily trap heuristic approaches [ 27 ]. So far, these local search heuristics lack any provable correctness or precision [ 28 , 29 ].…”

“…In practice, however, gene trees also can be incomplete. This can be addressed by filling in the gene trees (e.g., [ 29 ]), and then applying UrExact. Another option is to apply the minus method [ 39 ]; that is when computing the pairwise unrooted reconciliation costs between a gene tree and a candidate species tree, the species tree is constrained to the host-species of the gene tree’s leaf-genes.…”

Phylogenetic comparison measurements and their application towards the accurate inference of evolutionary histories

“…One of the most important properties for the median tree methods is the so-called Pareto for clusters property (Neumann, 1983). Pareto for clusters is a highly desired property, both from theoretical as well as application perspectives (Bryant, 2003;Moon and Eulenstein, 2017). This property was introduced in the context of consensus methods that seek to "summarize" a collection of input trees over the same species set.…”

“…Generally, any median tree method with the respective distance measurement satisfying at least weak Pareto can largely benefit from using SCM algorithms. For example, recently, the NP-hard gene duplication median tree problem (GD) (Ma et al, 2000) was efficiently and effectively approached (Moon and Eulenstein, 2017) by combining the greedy SCM method (Huson et al, 1999) with the exact dynamic programming GD solution (Chang et al, 2013) and the popular GD heuristic, DupTree (Wehe et al, 2008).…”