Buried Treasure

Computational analysis of metagenomics data: basic analyses and why they are not so basic

Jeroen Raes giving an excellent summary of the many challenges in analysis of metagenomics data. In many cases statistically significant differences are do to “something else”: differences in sampling methods (e.g. filter sizes used for ocean samples), sequencing technologies (e.g. differences in read length), and sequence processing (e.g. algorithms used for assembly and gene finding) can all have a major impact on the final results. The details of how each sample was produced and analyzed really matter if you want to compare metagenomes.