9 Genomic Techniques and How to Apply Them to Marine Questions
369
(Brazma et al. 2001). It is aimed at standardizing the necessary content of a
submission to a public database.
The MicroArray Gene Expression Markup Language (MAGE-ML) format was
developed by the MGED Society based on XML to provide a standardized document format to exchange microarray data (Spellman et al. 2002). However,
this document format is highly complex and should therefore be only used for
data exchange between software applications. It has been complemented by a
tab-delimited spreadsheet-based format called MAGE-TAB (Rayner et al. 2006),
which is much simpler in structure and should now be preferred for large-scale
submissions.
9.4.4 Summary of the Gene Expression Analysis Section
Gene expression analysis is a highly flexible and promising approach within marine
genomics and expression data can greatly aid the inference of functions from
sequence data. Quantitative analysis of transcriptomes can be performed using
microarrays in combination with quantitative RT-PCR for validation. If only a few
known genes are to be investigated, the use of microarrays is not necessary but
quantitative RT-PCR could be used instead.
While qRT-PCR and microarrays require the sequence of the transcript to be
known but the SAGE method generates transcript sequence data during the experimentation and can therefore be applied to genomes that have not been sequenced.
Genome wide expression profiling is only possible with the SAGE and microarrays
approaches. The most popular method of these two is the microarray technology,
as it allows mRNA levels of thousands of genes to be studied in parallel in a cost
efficient manner. Also, SAGE can be applied only to Eukaryotes.
Good experimental design and an appropriate level of replication are crucial for
microarray experiments. The optimal setup depends on the biological question. In
principle biological replicates should be preferred over technical replicates and at
least three biological replicates should be made.
There are a growing number of commercial providers of microarrays and
related services. For large scale comparative studies there is no clearly preferable platform, thus the choice of array provider should be based on criteria
such as budget, availability of designs and services, and in particular practical
experience with the methods. However, setting up an in-house array production
pipeline is not recommended for small to medium sized labs because of high setup costs and the investment in time. For large labs or consortia, this might still
be an option, while for occasional applications a full-service provider should be
considered.
Each experiment has a certain level of unavoidable experimental variation,
however technical variation can be reduced by following experimental protocols
rigidly.
Data analysis is a complex task and requires bioinformatics and statistical expertise. Depending on the experimental setup, simple analyses such as statistical tests
369
(Brazma et al. 2001). It is aimed at standardizing the necessary content of a
submission to a public database.
The MicroArray Gene Expression Markup Language (MAGE-ML) format was
developed by the MGED Society based on XML to provide a standardized document format to exchange microarray data (Spellman et al. 2002). However,
this document format is highly complex and should therefore be only used for
data exchange between software applications. It has been complemented by a
tab-delimited spreadsheet-based format called MAGE-TAB (Rayner et al. 2006),
which is much simpler in structure and should now be preferred for large-scale
submissions.
9.4.4 Summary of the Gene Expression Analysis Section
Gene expression analysis is a highly flexible and promising approach within marine
genomics and expression data can greatly aid the inference of functions from
sequence data. Quantitative analysis of transcriptomes can be performed using
microarrays in combination with quantitative RT-PCR for validation. If only a few
known genes are to be investigated, the use of microarrays is not necessary but
quantitative RT-PCR could be used instead.
While qRT-PCR and microarrays require the sequence of the transcript to be
known but the SAGE method generates transcript sequence data during the experimentation and can therefore be applied to genomes that have not been sequenced.
Genome wide expression profiling is only possible with the SAGE and microarrays
approaches. The most popular method of these two is the microarray technology,
as it allows mRNA levels of thousands of genes to be studied in parallel in a cost
efficient manner. Also, SAGE can be applied only to Eukaryotes.
Good experimental design and an appropriate level of replication are crucial for
microarray experiments. The optimal setup depends on the biological question. In
principle biological replicates should be preferred over technical replicates and at
least three biological replicates should be made.
There are a growing number of commercial providers of microarrays and
related services. For large scale comparative studies there is no clearly preferable platform, thus the choice of array provider should be based on criteria
such as budget, availability of designs and services, and in particular practical
experience with the methods. However, setting up an in-house array production
pipeline is not recommended for small to medium sized labs because of high setup costs and the investment in time. For large labs or consortia, this might still
be an option, while for occasional applications a full-service provider should be
considered.
Each experiment has a certain level of unavoidable experimental variation,
however technical variation can be reduced by following experimental protocols
rigidly.
Data analysis is a complex task and requires bioinformatics and statistical expertise. Depending on the experimental setup, simple analyses such as statistical tests