9 Genomic Techniques and How to Apply Them to Marine Questions
317
can analyse as many as 96 samples at a time. The signal generated by the laser
detection system constitutes a sequence trace file in which the intensities of fluorescence of the four dyes corresponding to the four bases are plotted as a function of
speed of migration (which is equivalent to fragment size and therefore provides a
position within the sequenced region).
9.1.1.2 Shotgun Technique
The Sanger sequencing method generates sequence data for regions of up to about
one kilobase. A strategy is therefore needed to apply this method to the sequencing
of an entire genome. For bacterial genomes and small eukaryote genomes whole
genome shotgun sequencing is the method of choice. This technique involves fragmentation of the genomic DNA into pieces of a defined length, cloning them into
sequencing vectors, and introducing them into an E. coli host strain. A random selection of the resulting recombinant clones is then sequenced and the sequence reads
are assembled into contiguous regions (contigs) based on sequence overlaps. For
larger genomes the hierarchical shotgun approach provides an alternative method
(Green 2002). In this approach, a genome is decomposed into larger fragments, for
example large fragments cloned into Bacterial Artificial Chromosomes (BACs). The
BACs are then ordered into a minimal tiling path using Polymerase Chain Reaction
(PCR) or labour-intensive hybridization techniques. The selected subset of clones
(the minimum tiling path) are then individually sequenced using a shotgun approach
for each piece of DNA (Kaiser et al. 2003).
9.1.1.3 Bacterial Genome Assembly and Finishing
Several bioinformatics tools were developed at the same time as the first genome
sequencing projects were carried out. Examples include the basecalling and
sequence trace file quality clipping program PHRED (Ewing et al. 1998), the DNA
sequence assembly programs PHRAP (Green, 1996) and CAP3 (Huang and Madan,
1999) and the genome sequence finishing program Consed (Gordon et al. 1998). In
2003 the Bioinformatics Resource Facility at the university of Bielefeld developed
an optimized approach for whole genome shotgun sequencing (Kaiser et al. 2003),
which combined the advantages of fast high throughput shotgun sequence data generation with a sequence finishing and validation phase driven by large insert BAC or
fosmid clone libraries. The objective of this approach was to produce high quality
bacterial genome sequences in a time- and cost-effective manner. This sequencing
strategy consists of two main steps: (i) high-throughput generation of shotgun reads
using Sanger sequencing (up to an eight-fold coverage is performed) and (ii) a manually driven sequencing and linking phase, using end sequences of large insert BACor fosmid- libraries and finishing reads generated by primer walking.
To monitor the fist step, the high-throughput shotgun sequencing phase, a new
bioinformatics tool called SAMS (Sequence Analysis and Management System)
was developed (Bekel et al. 2009). SAMS processes raw sequence data (e.g. .scf
files) as follows: first a normalizing step involving basecalling and quality clipping
317
can analyse as many as 96 samples at a time. The signal generated by the laser
detection system constitutes a sequence trace file in which the intensities of fluorescence of the four dyes corresponding to the four bases are plotted as a function of
speed of migration (which is equivalent to fragment size and therefore provides a
position within the sequenced region).
9.1.1.2 Shotgun Technique
The Sanger sequencing method generates sequence data for regions of up to about
one kilobase. A strategy is therefore needed to apply this method to the sequencing
of an entire genome. For bacterial genomes and small eukaryote genomes whole
genome shotgun sequencing is the method of choice. This technique involves fragmentation of the genomic DNA into pieces of a defined length, cloning them into
sequencing vectors, and introducing them into an E. coli host strain. A random selection of the resulting recombinant clones is then sequenced and the sequence reads
are assembled into contiguous regions (contigs) based on sequence overlaps. For
larger genomes the hierarchical shotgun approach provides an alternative method
(Green 2002). In this approach, a genome is decomposed into larger fragments, for
example large fragments cloned into Bacterial Artificial Chromosomes (BACs). The
BACs are then ordered into a minimal tiling path using Polymerase Chain Reaction
(PCR) or labour-intensive hybridization techniques. The selected subset of clones
(the minimum tiling path) are then individually sequenced using a shotgun approach
for each piece of DNA (Kaiser et al. 2003).
9.1.1.3 Bacterial Genome Assembly and Finishing
Several bioinformatics tools were developed at the same time as the first genome
sequencing projects were carried out. Examples include the basecalling and
sequence trace file quality clipping program PHRED (Ewing et al. 1998), the DNA
sequence assembly programs PHRAP (Green, 1996) and CAP3 (Huang and Madan,
1999) and the genome sequence finishing program Consed (Gordon et al. 1998). In
2003 the Bioinformatics Resource Facility at the university of Bielefeld developed
an optimized approach for whole genome shotgun sequencing (Kaiser et al. 2003),
which combined the advantages of fast high throughput shotgun sequence data generation with a sequence finishing and validation phase driven by large insert BAC or
fosmid clone libraries. The objective of this approach was to produce high quality
bacterial genome sequences in a time- and cost-effective manner. This sequencing
strategy consists of two main steps: (i) high-throughput generation of shotgun reads
using Sanger sequencing (up to an eight-fold coverage is performed) and (ii) a manually driven sequencing and linking phase, using end sequences of large insert BACor fosmid- libraries and finishing reads generated by primer walking.
To monitor the fist step, the high-throughput shotgun sequencing phase, a new
bioinformatics tool called SAMS (Sequence Analysis and Management System)
was developed (Bekel et al. 2009). SAMS processes raw sequence data (e.g. .scf
files) as follows: first a normalizing step involving basecalling and quality clipping