Recently, researchers introduced their new tools for strain identification and data extraction in metagenomics research. This article was published on nature, which has detailedly described the function and advantage of the new tools.
The traditional methods of strain
identification are SNP
-based approaches. It is convenient and basically can be used for all kinds of strain
identification. However, the only limitation of this method is that it asks for small-read of sequencing
. This could be a problem for strains with a large group of data
sets like Bifidobacterium longum(a kind of probiotics which comes from human intestine).
The Constrain, however, has fixed this problem. Compared to the detection on species-level, the Constrain has two merits. Firstly, based on quantitative metagenomics, Constrain does not need any additional types of data
. It has a ten-fold sequence
coverage, and it can cluster and concatenate SNPs
Secondly, the Constrain has talents in longitudinal studies. It can identify the additional function of strains which can not be detected by species-level analysis
Another tool is Latent Strain Analysis
. This tool is based on the hashing function and the advantages are clear: the LSA has a high sensibility which can select strains from related species; it can find new microbes.
The LSA can assemble 90% genomes
. It can not distinguish very similar regions.
For a really long time researchers try to find effective and accurate methods of strain
-detection. Because different strains in the same bacterial species may have totally different characteristics, which is closely related to many diseases. There are few bioinformatic tools which are useful for metagenomic research.
Due to the large quantity of bacteria
family, bioinformatic tools and analysis
are important for research on the metagenomics research. The researcher of this study said that this is only one step of their exploration. They and other researchers will continually find more functional tools to help them understand the function of strains in their environment.
Brian Cleary, Ilana Lauren Brito, Katherine Huang, Dirk Gevers, Terrance Shea, Sarah Young & Eric J Alm: Detection of low-abundance bacterial strains in metagenomic datasets by eigengenome partitioning doi:10.1038/nbt.3329
Chengwei Luo, Rob Knight, Heli Siljander, Mikael Knip, Ramnik J Xavier & Dirk Gevers: ConStrains identifies microbial strains in metagenomic datasets doi:10.1038/nbt.3319
Tal Nawy: The strain
in metagenomics doi:10.1038/nmeth.3642