Resolving contradicting binets and finding similarities in networks

To understand the (physical) differences between species, biologist research the common ancestry of these species. This research area is called phylogenetics and the visualisation of the ancestry is displayed with the use of phylogenetic trees or networks. Generally, the phylogenetic networks are not known completely or with certainty. Therefore, it is important to develop algorithms that can construct large networks from smaller ones. It is possible that the data used for the algorithms is faulty, which can result in contradictions in the data. In order to still be able to construct a network, some of the data has to be disregarded. The algorithm discussed in this report will have as input data a set of binets: small phylogenetic networks with two leaves each. If the binet set contains contradictions, it is not possible to construct a large network that is consistent with all the input binets. This report will discuss several of the possibilities to decide which binets to omit. Even if the data does not contain any contradictions, the ultimate network can still be uncertain. It can occur that more than one network is able to explain the given data and based on the data alone it is not possible to determine which is the correct one. If this is the case, it is still possible to retrieve some certain information about the correct network from the data. For the case of binary, level-1 networks, a proof will be given that all the possible networks constructed from a complete binet set will contain the same number of reticulations (for example hybridization events).

Subject

phylogenetic networks
binets

To reference this document use:

http://resolver.tudelft.nl/uuid:2defcd3f-02be-4e7d-acd2-bb9261d1b533

Part of collection

Student theses

Document type

bachelor thesis

Rights

Files

PDF

Report.pdf

569.04 KB

Close viewer