The numbers of metagenomics studies are exponentially increasing in recent years. Because we lack comprehensive microbial genome databases for most environments, the microbiome field is moving towards de novo metagenomics, using reference-independent techniques to assemble putative genes and in silico refinement of draft genomes from short shotgun sequencing reads. The challenges with this technique are to critically assess the quality of assembly and to avoid chimeric contigs, which can result from sequencing or assembly errors.
On Friday, November 24 at 3pm in 3N10A, at the human microbiome journal club, we will review the performance of popular assembly algorithms using two benchmarking studies: Van der Walt et. al. (2017) that evaluated the metagenomics assembly using defined microbial communities and Greenwald et. al. (2017) which proposed a workflow to select metagenomic assembler based on the research question, the computational resources available and the bioinformatic expertise of the researcher.
More specifically, we will discuss
1. What are the criteria to critically assess the quality of a metagenomic assembly?
2. Why is it necessary to compare the performance of multiple algorithms for a given dataset?
3. What is the required computational power to conduct de novo assembly?
Hopefully, by the end of this week’s journal club, we’ll excite you enough that you will join us to continue our discussion over a beer at the Bioinformatics and Beers meet up at The Phoenix.