FY08 LDRD Final Report Probabilistic Inference of Metabolic Pathways from Metagenomic Sequence Data

D'haeseleer, P

关键词: BIOLOGICAL PATHWAYS;    COMMUNITIES;    ECOLOGY;    FUNCTIONALS;    GENES;    STRUCTURAL CHEMICAL ANALYSIS;    TESTING;   

DOI  :  10.2172/948980
RP-ID  :  LLNL-TR-410988
PID  :  OSTI ID: 948980
Others  :  TRN: US200909%%378

学科分类：生物科学（综合）

美国|英语

来源: SciTech Connect

【 摘 要 】

Metagenomic 'shotgun' sequencing of environmental microbial communities has the potential to revolutionize microbial ecology, allowing a cultivation-independent, yet sequence-based analysis of the metabolic capabilities and functions present in an environmental sample. Although its intensive sequencing requirements are a good match for the continuously increasing bandwidth at sequencing centers, the complexity, seemingly inexhaustible novelty, and 'scrambled' nature of metagenomic data is also proving a tremendous challenge for analysis. In fact, many metagenomics projects do not go much further than providing a list of novel gene variants and over- or under-represented functional gene categories. In this project, we proposed to develop a set of novel metagenomic sequence analysis tools, including a binning method to group sequences by species, inference of phenotypes and metabolic pathways from these reconstructed species, and extraction of coarse-grained flux models. We proposed to closely collaborate with the DOE Joint Genome Institute to align these tools with their metagenomics analysis needs and the developing IMG/M metagenomics pipeline. Results would be cross-validated with simulated metagenomic data using a testing platform developed at the JGI.