"Frontmatter". In: Plant Genomics and Proteomics

G
ENOME
F
RACTIONATION
A
PPROACHES
Here a physical map is not a prerequisite for collecting the sequence infor-
mation. The initial activity in every case is to generate libraries from some
form of a fractionated genome. 
M
ETHYL
F
ILTRATION
L
IBRARIES
Shotgun libraries from crop plants and their propagation in strains of E. coli
that restrict methylated DNA, thereby greatly enriching these libraries 
for gene-containing regions, have been developed by Orion Genomics 
(Rabinowicz et al., 1999; Yuan et al., 2002). The library construction must
reduce or eliminate the organellar genomes that are essentially unmethy-
lated. From normal green tissue, up to 5% of the total DNA could be derived
from the chloroplast. If the gene space that is desired was only 15% of the
total genome, then the organellar “contamination” would be about 25% 
of all the clones. Therefore, starting from isolated nuclei from nongreen
tissue should reduce the unwanted chloroplast “contamination.” Although
G
E N O M E
F
R A C T I O N AT I O N
A
P P R O A C H E S
6 1
Chromosome
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
FIGURE 3.6.
Use of BAC end sequences to determine minimum tiling path.

---

the expressed sequences are hypomethylated, they are not necessarily com-
pletely unmethylated. Therefore, the DNA that is cloned must be short
enough so that no methylated cytosines are present in the clones of interest
and must also be random. This can be achieved by shearing to low molecu-
lar weight (<500 bp), with the fragments then size selected, ligated into 
the appropriate vector, and transformed into the appropriate E. coli host
strain. The resulting clones can be picked into the most appropriate format,
for example, into 384-well plates for automated plasmid isolation and
sequencing. 
The initial evidence indicates that the filtration may be more successful
when used with monocotyledonous than with dicotyledonous plants. Thus
in rice, 128 mb remained after methyl filtration, giving an enrichment factor
of nearly 4, whereas in wheat only 874 mb remained after filtration, giving 
a 19-fold equivalent enrichment. However a threefold enrichment was
observed in canola, soybean, and tomato, the same value irrespective of the
genome size. Thus it would appear that as the genome gets larger for mono-
cots the filter power increases, whereas as the genome size gets larger for
dicots the filter power is reduced (Bedell, 2002). 
M
ETHYLATION
R
ESTRICTION
L
IBRARIES
Methylation-sensitive restriction enzymes such as PstI have been used to
produce libraries enriched in hypomethylated sequences. Any approach that
uses methylation-sensitive restriction enzymes must take into account the
size of the inserts that will be obtained and the completeness of the cover-
age of the subsequent libraries. However, in a genome with little previous
genomic sequence, a shotgun sequencing project using PstI clones may yield
a large amount of data on genes of that organism. The resulting clones can
be picked into the most appropriate format, for example, into 384-well plates
for automated plasmid isolation and sequencing. 
C
OT
F
RACTIONATION
-B
ASED
L
IBRARIES
A preliminary Cot analysis can be performed if required to define the con-
ditions to be used to isolate particular, interesting regions of the genome,
although many species have already been characterized in this manner. From
the Cot analysis the genome size can be estimated as well as the number and
relative proportion of the repetitive and single-/low-copy components. The
reassociated fraction selected is already double stranded, although the frag-
ments may not have flush ends suitable for cloning. Therefore, these need
to be blunted and then cloned in standard vectors. The resulting clones can
be picked into the most appropriate format, for example, into 384-well plates
for automated plasmid isolation and sequencing.
6 2
3. S
E Q U E N C I N G
S
T R AT E G I E S

---