Computational models for metatranscriptome analysis

点击翻译按钮获取中文摘要

Metagenomics has fundamentally changed the exploration of the microbial world because it allows to study organisms in their natural environment as part of complex communities. Metatranscriptomics based on sequencing of RNA in environmental samples has become a key technology to elucidate gene expression in microbial communities. In transcriptomics, high-throughput sequencing of RNA (RNA-Seq) has successfully been applied to gene expression analysis and well-established workflows exist for detection and interpretation of differential expression patterns. The goal of the proposed project is to examine how the existing pipelines for differential expression analysis can be extended and modified for an application to metatranscriptomic data. Current approaches to metatranscriptome analysis have mainly been adopted from comparative metagenomics, which does not provide integrated tools for differential expression analysis. The project will investigate how the analytical power of metatranscriptomics and the reproducibility of results can be improved by the incorporation of specific models from machine learning and RNA-Seq analysis. The design and evaluation of an optimized metatranscriptomics pipeline will be performed in close collaboration with the G¨ottingen Genomics Laboratory (G2L), which provides an outstanding expertise in metagenome analysis. In particular, the project will support an ongoing cooperation with the G2L for analysis of comprehensive metatranscriptome data from different soil communities.