EvoMiner: frequent subtree mining in phylogenetic databases

Abstract

The problem of mining collections of trees to identify common patterns, called frequent subtrees (FSTs), arises often when trying to interpret the results of phylogenetic analysis. FST mining generalizes the wellknown maximum agreement subtree problem. Here we present EvoMiner, a new algorithm for mining frequent subtrees in collections of phylogenetic trees. EvoMiner is an Apriori-like level-wise method, which uses a novel phylogeny-specific constant-time candidate generation scheme, an efficient fingerprinting-based technique for downward closure, and a lowest common ancestor based support counting step that requires neither costly subtree operations nor database traversal. Our algorithm achieves speed-ups of up to 100 times or more over Phylominer, the current state-of-the-art algorithm for mining phylogenetic trees. EvoMiner can also work in depth first enumeration mode, to use less memory at the expense of speed. We demonstrate the utility of FST mining as a way to extract meaningful phylogenetic information from collections of trees when compared to maximum agreement subtrees and majority rule trees ---two commonly used approaches in phylogenetic analysis for extracting consensus information from a collection of trees over a common leaf set. Disciplines Theory and AlgorithmsThis article is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/cs_techreports/11Noname manuscript No. The final publication is available at http://link.springer.com Abstract The problem of mining collections of trees to identify common patterns, called frequent subtrees (FSTs), arises often when trying to interpret the results of phylogenetic analysis. FST mining generalizes the well-known maximum agreement subtree problem. Here we present EvoMiner, a new algorithm for mining frequent subtrees in collections of phylogenetic trees. EvoMiner is an Apriori-like level-wise method, which uses a novel phylogeny-specific constant-time candidate generation scheme, an efficient fingerprinting-based technique for downward closure, and a lowest common ancestor based support counting step that requires neither costly subtree operations nor database traversal. Our algorithm achieves speed-ups of up to 100 times or more over Phylominer, the current state-of-the-art algorithm for mining phylogenetic trees. EvoMiner can also work in depth first enumeration mode, to use less memory at the expense of speed. We demonstrate the utility of FST mining as a way to extract meaningful phylogenetic information from collections of trees when compared to maximum agreement subtrees and majority rule trees -two commonly used approaches in phylogenetic analysis for extracting consensus information from a collection of trees over a common leaf set.

SciPhyloMiner: um Workflow para Mineração de Dados Filogemônicos de Protozoários

“…Dessa forma, apresentamos aplicações existentes que mineram árvores filogenômicas e que facilmente poderiam ser incorporadas ao SciPhyloMiner. Um dos algoritmos mais usados na identificação de padrões comuns, o FST, foi implementado pelo algoritmo Evominer (Deepak et al, 2014), acoplado ao workflow SciPhyloMiner. Deepak et al (2014) (Chen et al, 2008) é uma aplicação baseada na Web para a procura de árvores em banco de dados (TreeBASE); a entrada é uma árvore que é comparada contra o TreeBase via consulta por palavra-chave ou nome do táxon, podendo navegar pelos ancestrais e filhos desse táxon.…”

“…Um dos algoritmos mais usados na identificação de padrões comuns, o FST, foi implementado pelo algoritmo Evominer (Deepak et al, 2014), acoplado ao workflow SciPhyloMiner. Deepak et al (2014) (Chen et al, 2008) é uma aplicação baseada na Web para a procura de árvores em banco de dados (TreeBASE); a entrada é uma árvore que é comparada contra o TreeBase via consulta por palavra-chave ou nome do táxon, podendo navegar pelos ancestrais e filhos desse táxon. O PhyloSift (Darling et al, 2014) é uma ferramenta para análises filogenéticas em genomas focado na anotação e análise funcional e taxonômica de metagenomas.…”

Leaf realization problem, caterpillar graphs and prefix normal words

“…It is worth mentioning that our motivation in the study of induced subtrees having the maximum number of leaves comes from [19,20], where remarkable structures on regular lattices are presented. Other similar problems have also been studied, such as maximum leaf spanning subtrees [21,22], frequent subtrees mining [23] and induced subtrees [24].…”

A Novel Coverage Pattern Mining Method for Unordered Tree

“…Modeling these data with unordered tree and mining frequent subtrees can effectively find the hidden knowledge. At present, many frequent subtree mining algorithms have been proposed, such as FP-tree tree [1] and PFP-tree [2] mine frequent subtrees based on the idea of projection, while EvoMiner [3] and MCFP-tree [4] adopt the ideas of Apriori. However, these algorithms usually output large scale frequent subtrees with redundant information which affect the efficiency of subsequent processing.…”