"Frontmatter". In: Plant Genomics and Proteomics

CHAPTER
3
S
E Q U E N C I N G
S
T R AT E G I E S
O
VERVIEW
The central goal in any genome sequencing project is the discovery of all the
genes in that organism, which in essence can only be known for certain when
every nucleotide in the genome is known and placed in its appropriate
context. On December 14, 2000 the first complete genome sequence of a plant
was reported in which the analysis of sequenced regions covered 115.4 Mb
of a total genome size of 125 Mb (The Arabidopsis Genome Initiative, 2000).
How was this achieved? Essentially, the Arabidopsis genome was sequenced
with the traditional approach to genome sequencing. This involved cutting
the genome into a large number of subfragments, cloning the pieces, and
then grouping the pieces together as large regions of overlapping clones. The
chromosomal locations of all of these large groups were determined with
the high-density genetic map. Groups of researchers then took on the 
task of sequencing a specific chromosome or chromosome segment. This
sequencing strategy is slow and expensive but provides the most precise and
complete sequence across the entire genome. The resulting sequence only
represented the inbred progeny of a single isolate of Arabidopsis, so any vari-
ation between ecotypes was not immediately available. 
Complete genome sequences, even those that are made up of a very large
number of relatively short segments, of a variety of plants will provide pow-
erful tools for biologists. The sequences will aid in understanding how gene
families have been created, amplified, and diverged, resulting in the creation
of new biological activities and specificities. The gene content of related
species can be investigated and compared to identify which pathways are
shared among many species and which are restricted to some parts of the
plant kingdom. The advantage of having a full set of genes for a compre-
Plant Genomics and Proteomics, by Christopher A. Cullis
ISBN 0-471-37314-1 Copyright © 2004 John Wiley & Sons, Inc.
4 7

---

hensive characterization of gene expression especially, for example, when
plants are growing under stressed conditions, will lead to the understand-
ing and subsequent manipulation of plant growth for improved agronomic
performance. The efficient use of high-throughput approaches like micro-
array hybridization and analysis will be much more efficient if the complete
suite of genes for the plant is available for experimentation. Not only 
will geneticists have access to the genes underlying quantitative trait loci
(QTLs), but they will also be able to generate essentially an unlimited
number of DNA markers. One of the major uses of a genome sequence
would be to perform map-based cloning of genes and to associate candidate
genes with important traits. A well-integrated physical and genetic map is
essential for map-based cloning, rendering a pure shotgun sequence less
valuable than an anchored sequence. Therefore, the genes identified by
sequencing projects will provide the basis for determining why and how
some characteristics are shared among particular lineages of organisms
while others have a more limited representation in the branches of the tree
of life. A complete genome sequence is a first step toward the understand-
ing of biological processes, but it still must be followed by detailed studies
of gene function.
The costs and benefits of following a complete genome strategy as
opposed to just trying to identify the genes themselves need to be consid-
ered. Relatively few plant species have a comprehensive sequence analysis
available. Across-species comparisons are very valuable, but how many
species are needed to be able to use comparative and syntenic relationships
to draw valid conclusions? Therefore, how many genomes need to be
sequenced? What are the alternative approaches to getting sufficient infor-
mation that do not necessarily aim to generate the complete genome
sequence but will serve to add enormous value to the existing genome
resources?
The traditional approach to genome sequencing (the stepwise sequenc-
ing of overlapping clones) is slow and expensive. A faster and less expen-
sive method is the shotgun sequence analysis of small-insert clones (Venter
et al., 1996). This latter approach was used to generate the rice genome
sequence by Syngenta’s Torrey Mesa Research Institute and the Beijing
Genomics Institute (Goff et al., 2002). This method, which produced draft
sequences, only cost approximately 10% of the International Rice Genome
Sequencing Project (ISGRP), which followed the traditional sequencing
method. Shotgun sequencing alone does not provide the locations of the
sequence segments on the genetic or physical map. In the traditional
sequencing approach, however, the positions of the sequenced regions are
already known. An alternative to generating a whole genome sequence is to
devise a gene enrichment strategy that will result in the sequences of all the
genes but without the need to generate the sequences of all the repetitive
regions. This gene enrichment strategy will provide a paradigm for a cost-
4 8
3. S
E Q U E N C I N G
S
T R AT E G I E S

---

effective sequencing of large complex genomes of plants that would other-
wise be cost prohibitive to produce by whole genome sequence.
In this chapter we consider the processes for generating both 
whole genome sequences and partial genome sequences enriched for genes.
The organization of a whole genome sequence by the traditional and
shotgun methods is considered. Various forms of gene enrichment strategies
including expressed sequence tags (ESTs), kinetic and methylation-
dependent fractionations of the genome, and the use of transposons are
explained. The reliance on bioinformatics to assemble such sequences is also
considered. A compilation of the “best approach” to obtaining useful
genome sequences is described. Finally, the potential plant targets for
genome sequencing in the near future and the effect of technology on any
predictions are discussed.