Fine Read­ing: Metage­nomics

by Elio

Noth­ing like be­ing lazy on a nice Au­tumn week­end. If you are so in­clined, and yet you are won­der­ing about meta­gen­omics and specif­i­cally about what it has done for our un­der­stand­ing of the mi­cro­bial world, here's a sat­is­fy­ing piece by Hugen­holtz and Tyson. For those who can't wait to read it, here are some of the main con­clu­sions.

Foto by echer­ries

Among other things, mi­cro­bial metage­nomics:

• re­veals nat­ural com­mu­ni­ties in their rich­ness and com­plex­ity. Metage­nomics can not only iden­tify the prin­ci­pal part­ners of a mi­cro­bial con­sor­tium but also many of their meta­bolic ca­pa­bil­i­ties.
• has great po­ten­tial for stum­bling across novel find­ings, some of which have sig­nif­i­cantly al­tered our view of global processes. One ex­am­ple is pro­te­orhodopsin, the light-dri­ven pro­ton pump, dis­cov­ered to be abun­dant in mi­crobes in ma­rine habi­tats where there is light.
• is able to re­con­struct whole genomes, though cur­rently only for prokary­otes, and it works best in sim­ple com­mu­ni­ties with a dom­i­nant species.
• tells us about the vari­a­tion in nat­ural pop­u­la­tions, thus pro­vid­ing in­sight into the evo­lu­tion­ary processes that have shaped those genomes. For ex­am­ple, in the most abun­dant pho­to­syn­the­sizer in oceans, Prochloro­coc­cus, great­est vari­a­tion is seen in ge­nomic is­lands, likely re­sult­ing from virus-as­sisted lat­eral trans­fers.
• demon­strates that we have scarcely be­gun to dis­cover the di­ver­sity that is out there.

But there are lim­i­ta­tions. As se­quenc­ing costs drop, the amass­ing of huge num­bers of se­quences is out­strip­ping com­pu­ta­tional ca­pac­ity. Eu­kary­ote genomes are still to be avoided. And if you dream of do­ing com­par­isons of every se­quence against every se­quence, you'll likely have to wait for a new gen­er­a­tion of com­put­ers. Nice to know that the bot­tle­neck is not with mi­cro­bi­ol­ogy, but in some­body else's baili­wick, in­stead.