**6. Groups and metagenomics**

Phytoplankton, free and attached bacteria, aggregates of particulate organic matter and grazers, such as rotifers, ciliates and flagellates, and protozoa and copepods are common groups of microorganisms in BFT. As the use, identification and study of microbes in aquaculture have become a usual practice in the last decade [69]. For a long time, techniques based on culture media were used as the main strategy to know the microbial composition of biotic communities, including biofilm and BFT; however, this was a very superficial approach considering that >80% of the bacteria thriving in any environment are readily culturable or unculturable at all [70]. The overcome of culture-independent techniques such as denaturing gradient gel electrophoresis (DGGE) but particularly high-throughput sequencing (next-generation sequencing) increased the depth and coverage of studies aiming to study the microbial diversity of these kinds of conglomerates [71, 72].

Metagenomics is therefore a relative recent genomics subdiscipline that has emerged as a promising scientific tool to analyze the complex genomes contained within microbial communities. However, its use is not yet common in some agro-industrial disciplines such as aquaculture. The reason of this relies in the high cost of this technology; however, prices have significantly decreased during the last decade, and now it is possible for individual laboratories to perform metagenomics studies using high-throughput sequencing.

The study of microbial diversity can be studied with the highest resolution so far, for instance, ribosomal genes such as 16S and 23S have been used as a targeted loci approach for diversity studies of prokaryotic communities. Herein, universal genes are used for the amplification of particular hypervariable regions of these genes [4]; these hypervariable regions contain elements to differentiate organisms. In this regard, this technology offers the possibility to reveal most of the bacteria thriving in any biofloc biomass. However, current sequencing technologies can cover only a fraction (~600 bp) of the ribosomal genes used for taxonomic classification, which means that only 2 or 3 of the 10 hypervariable regions of the 16S gene can be used for this classification. Researchers have made efforts to elucidate whose sequences are the most information richness [73]; however, there is still a loss of information contained on the regions that cannot be covered by these sequencing platforms.

Novel technologies such as single-molecule real-time (SMRT) sequencing have been developed [74]. This particular technology has advantages such as the generation of long reads and high accuracy. Long reads could be useful for sequencing not only ribosomal genes but also larger fragments of DNA, which would serve for multilocus classification. Whether the price for doing metagenomics of bacteria is still high, it is expected to decrease along the following years.

In spite of the incomplete coverage of ribosomal genes by most of the current high-throughput sequencing platforms, the amplification of two or three hypervariable regions of these genes is still a very useful tool to know most of the bacteria contained in biofilms and bioflocs [72] and inclusively to detect novel species, study dynamic population patterns, probiotic activity, etc. Furthermore, metagenomics based on single-gene surveys and random shotgun studies of all accessible genes in any environment could be two useful approaches to study the biological activity and communications of these complex bacterial networks. Whether the use of BFT systems is now a reality and promises to be revolutionary strategy, their biology studied through novel genomic tools is still to provide mass information of these biotic communities.
