8 Marine Biotechnology
291
exploit the huge reservoir of genes in these types of environments have still to be
refined, in particular to gain access to rare phylotypes or species.
In spite of considerable efforts to develop novel bioinformatic methods, the
assembly and analysis of discrete microbial genomes from complex communities
is still a difficult task. This has triggered the development of alternative approaches
among which single-cell genomics is becoming increasingly popular. This method
involves a first step based on single cell isolation followed by whole-genome amplification using ϕ29 DNA polymerase in a reaction called multiple displacement
amplification (Dean et al. 2002). It opens interesting perspectives in various cases,
notably when the targeted species are still uncultured or when the amount of DNA
available is extremely limited.
The overall result of sequencing projects dedicated to both specific genomes and
to metagenomes is the unprecedented access to unique gene sequences. To put this
novel gene resource to good use is the major challenge of modern day genomics. The
traditional combination of in silico analysis (data mining) and functional screening
is still the cornerstone for discovery in biotechnological applications. Yet with the
increasingly large and complex datasets being generated, we are at a stage where
we will soon be drowning in data whilst still thirsting for knowledge. In the future,
novel multidisciplinary approaches will need to be developed to efficiently exploit
our increasing genomic knowledge.
8.3.1 Complete Genomes
For a long time it was thought that the sequencing of the complete genome of a
type species would give access to the majority of the gene pool of this species and
that a few dozen representative genomes from the bacterial and archaeal domains
would describe most of the gene pool of the microbial world. This is clearly
not the case. The genomic variation within a single species was inferred from
comparative genomics of pathogens including E. coli and Streptococcus agalactiae
(Bielaszewska et al. 2007). These studies demonstrated that it is not possible to characterize a species from a single genome sequence and that the number of genome
sequences (pan-genome) needed to describe a species may vary depending on the
species. As a consequence, for some important pathogens, the number of strains
undergoing genome sequencing is increasing rapidly (for example: 2 complete and
14 incomplete sequences for Vibrio cholerae, and probably more than 50 within 2
years). Until now, all genome sequences of marine species of biotechnological interest have been obtained from single clonal cultures established from single cell. In
the process, part (possibly, most!) of the variability has been lost and the corresponding set of genes may not be easily retrieved by genome analysis and data mining.
This is why functional screening is still needed, in parallel with data mining, to
explore strain variability. In the near future, as with pathogens, projects dedicated to
genome sequencing of several strains belonging to a single species of biotechnological interest, are likely to become widespread. This is already the case for the fungi
291
exploit the huge reservoir of genes in these types of environments have still to be
refined, in particular to gain access to rare phylotypes or species.
In spite of considerable efforts to develop novel bioinformatic methods, the
assembly and analysis of discrete microbial genomes from complex communities
is still a difficult task. This has triggered the development of alternative approaches
among which single-cell genomics is becoming increasingly popular. This method
involves a first step based on single cell isolation followed by whole-genome amplification using ϕ29 DNA polymerase in a reaction called multiple displacement
amplification (Dean et al. 2002). It opens interesting perspectives in various cases,
notably when the targeted species are still uncultured or when the amount of DNA
available is extremely limited.
The overall result of sequencing projects dedicated to both specific genomes and
to metagenomes is the unprecedented access to unique gene sequences. To put this
novel gene resource to good use is the major challenge of modern day genomics. The
traditional combination of in silico analysis (data mining) and functional screening
is still the cornerstone for discovery in biotechnological applications. Yet with the
increasingly large and complex datasets being generated, we are at a stage where
we will soon be drowning in data whilst still thirsting for knowledge. In the future,
novel multidisciplinary approaches will need to be developed to efficiently exploit
our increasing genomic knowledge.
8.3.1 Complete Genomes
For a long time it was thought that the sequencing of the complete genome of a
type species would give access to the majority of the gene pool of this species and
that a few dozen representative genomes from the bacterial and archaeal domains
would describe most of the gene pool of the microbial world. This is clearly
not the case. The genomic variation within a single species was inferred from
comparative genomics of pathogens including E. coli and Streptococcus agalactiae
(Bielaszewska et al. 2007). These studies demonstrated that it is not possible to characterize a species from a single genome sequence and that the number of genome
sequences (pan-genome) needed to describe a species may vary depending on the
species. As a consequence, for some important pathogens, the number of strains
undergoing genome sequencing is increasing rapidly (for example: 2 complete and
14 incomplete sequences for Vibrio cholerae, and probably more than 50 within 2
years). Until now, all genome sequences of marine species of biotechnological interest have been obtained from single clonal cultures established from single cell. In
the process, part (possibly, most!) of the variability has been lost and the corresponding set of genes may not be easily retrieved by genome analysis and data mining.
This is why functional screening is still needed, in parallel with data mining, to
explore strain variability. In the near future, as with pathogens, projects dedicated to
genome sequencing of several strains belonging to a single species of biotechnological interest, are likely to become widespread. This is already the case for the fungi