MLTreeMap - Maximum likelihood placement of environmental DNA sequence reads into curated reference phylogenies

Abstract

Traditional microbiology has proven to be insufficient for studying entire microbial communities in situ, because only a small fraction of microbes can be grown in pure culture. The idea of circumventing this bottleneck by directly sequencing DNA from the environment led to a new field of research, called metagenomics. As a consequence of its approach, metagenomics provides a very unbiased view of all organisms contained in a sample, but it also has to cope with heavily fragmented sequence data. MLTreeMap, which is presented in this thesis, is a software framework designed to give insights into phylogenetic and functional properties of metagenomes and of the underlying microbial communities. It does so by detecting and phylotyping a series of relevant marker genes on the submitted DNA fragments. Among these genes are protein coding phylogenetic markers, 16S and 18S rRNA genes and markers for important functional pathways. Examples of the latter are genes coding for the key enzymes of photosynthesis, nitrogen fixation, methane fixation and ammonia oxidation. MLTreeMap is available as a web-server at http://mltreemap.org and also as a stand-alone version. It has been published in BMC Genomics in 2010 [1].