3 Populations and Pathways
75
prefaced in the subtitles to indicate whether they are mainly used in either DNAbased (population based analyses) or RNA-based (functional/expression) studies.
A brief description of each tool is given, with reference to more detailed texts if
required.
3.1.1 DNA and RNA Studies: EST Libraries
EST is the abbreviation for “Expressed Sequence Tag”. An EST is a single sequencing read from a cloned piece of cDNA. The production of a cDNA occurs when
an RNA molecule corresponding to the expressed part of a gene is converted into a
DNA copy to increase stability. Each clone is then usually sequenced from one end
(generally the 5 end of the gene to avoid the complications of sequencing errors
associated with the polyA tail. It also allows more direct access to the open reading frame and thus increases the chance of finding potential gene matches using
sequence similarity searching of the databases). This data is entered into a public
database, dbEST (Boguski et al. 1993) (http://www.ncbi.nlm.nih.gov/dbEST/) on
the understanding that the sequencing quality may not be 100% and errors may
be present. ESTs can be produced from any cell/tissue/species and are a good
method for discovering and identifying genes in non-model organisms. They can
also be used to data-mine for markers for DNA studies (microsatellites and single
nucleotide polymorphisms (SNPs)). For DNA-based studies, the source of the EST
library is, not important however, this is an important consideration in expression
work, along with the type of library used.
Production of lots of ESTs from a particular tissue/cell type creates a library.
Libraries can be produced for just a few hundred sequences, or many thousands
(largely dependent on cost and ease of access to large-scale sequencing facilities).
There are essentially two approaches for producing EST libraries:
• Non-subtracted methodologies: cDNA is made directly from the cells or tissues
under investigation. So the number of times a sequence is present in a library is a
direct reflection on the quantity of the RNA quantity corresponding to that gene in
the cell. This may mean that highly expressed sequences (cf. actin in muscle) may
mask the detection of rarely expressed clones. Sequencing a library containing
many highly expressed sequences is also not ideal as sequencing the same gene
many times over is not efficient and only generally of utility when assessing the
nucleotide polymorphism of a specific locus). This issue is particularly important
when only a small number of clones are generated and sequenced, but knowing
the relative quantities of an RNA molecule (gene) across in various tissues can
be very useful.
• Subtracted and/or normalised methodologies: This involves hybridization steps to
compete out the most highly expressed sequences in the cell/tissue and to increase
the relative number of copies of rare differentially expressed sequences in the cell
(cf. Suzuki et al. 1997, Carninci et al. 2002, Otsuka et al. 2003). So with this latter
75
prefaced in the subtitles to indicate whether they are mainly used in either DNAbased (population based analyses) or RNA-based (functional/expression) studies.
A brief description of each tool is given, with reference to more detailed texts if
required.
3.1.1 DNA and RNA Studies: EST Libraries
EST is the abbreviation for “Expressed Sequence Tag”. An EST is a single sequencing read from a cloned piece of cDNA. The production of a cDNA occurs when
an RNA molecule corresponding to the expressed part of a gene is converted into a
DNA copy to increase stability. Each clone is then usually sequenced from one end
(generally the 5 end of the gene to avoid the complications of sequencing errors
associated with the polyA tail. It also allows more direct access to the open reading frame and thus increases the chance of finding potential gene matches using
sequence similarity searching of the databases). This data is entered into a public
database, dbEST (Boguski et al. 1993) (http://www.ncbi.nlm.nih.gov/dbEST/) on
the understanding that the sequencing quality may not be 100% and errors may
be present. ESTs can be produced from any cell/tissue/species and are a good
method for discovering and identifying genes in non-model organisms. They can
also be used to data-mine for markers for DNA studies (microsatellites and single
nucleotide polymorphisms (SNPs)). For DNA-based studies, the source of the EST
library is, not important however, this is an important consideration in expression
work, along with the type of library used.
Production of lots of ESTs from a particular tissue/cell type creates a library.
Libraries can be produced for just a few hundred sequences, or many thousands
(largely dependent on cost and ease of access to large-scale sequencing facilities).
There are essentially two approaches for producing EST libraries:
• Non-subtracted methodologies: cDNA is made directly from the cells or tissues
under investigation. So the number of times a sequence is present in a library is a
direct reflection on the quantity of the RNA quantity corresponding to that gene in
the cell. This may mean that highly expressed sequences (cf. actin in muscle) may
mask the detection of rarely expressed clones. Sequencing a library containing
many highly expressed sequences is also not ideal as sequencing the same gene
many times over is not efficient and only generally of utility when assessing the
nucleotide polymorphism of a specific locus). This issue is particularly important
when only a small number of clones are generated and sequenced, but knowing
the relative quantities of an RNA molecule (gene) across in various tissues can
be very useful.
• Subtracted and/or normalised methodologies: This involves hybridization steps to
compete out the most highly expressed sequences in the cell/tissue and to increase
the relative number of copies of rare differentially expressed sequences in the cell
(cf. Suzuki et al. 1997, Carninci et al. 2002, Otsuka et al. 2003). So with this latter