"Frontmatter". In: Plant Genomics and Proteomics

N is the number of clones generated
P is the required probability that the sequence is present
I is the average insert size in base pairs
C is the genome size in base pairs
The number of clones that need to be screened to find a single copy
sequence in libraries that were constructed from DNAs of various plant
species is given in Table 2.1.
N
P
I C
=
-
(
)
-
(
)
ln
ln
1
1
C
L O N I N G
S
Y S T E M S
2 7
T
ABLE
2.1. T
HE NUMBER OF CLONES NECESSARY FOR HAVING A
99.995%
PROBABILITY
(10X 
COVERAGE
) 
OF FINDING A SINGLE COPY SEQUENCE IN A
LIBRARY WITH
100,000 
BP INSERTS FOR VARIOUS PLANT SPECIES
Species
Genome size (Mbp)
Number of clones required
Arabidopsis thaliana
0.125
12,500
Maize
2.73
273,000
Wheat
17
1,700,000
Fritellaria assyrica
124
12,400,000

---

B
ACTERIAL
A
RTIFICIAL
C
HROMOSOME
L
IBRARIES
These are now a staple resource in the plant genomics community. The
libraries can be maintained at low temperatures and are very easily adapted
for use in high-throughput automated processes. The insert size that can be
accommodated is sufficiently large to generate a manageable library for most
plant genomes. The clones can be picked and stored in 96- or 384-well plates
for use with most liquid handling systems. Automated procedures for the
isolation of BAC DNA followed by the fingerprinting of these clones (see
Chapter 3) have made such libraries the material with which most physical
maps are generated. Two different vectors are available for making BAC
libraries. These are the standard bacterial artificial chromosome (BAC)
vector, and the binary BAC (BIBAC) vector. The BIBAC vector is based 
on the standard BAC vector for genomic libraries, with the addition of
regions from the binary vector system for Agrobacterium-mediated plant
transformation (http://hbz7.tamu.edu/homelinks/tool/bac_content.htm;
http://www.research.cornell.edu/Biotech/BIBAC/BIBACHomePage.html).
This provides the opportunity for the direct transfer of the recombinants to
A. tumefaciens and subsequent use for plant transformation. One of the pos-
sible drawbacks of the BIBAC vector is that its larger size may interfere with
the automated DNA fingerprinting processes because of the large number
of overlapping bands generated from the larger vector DNA. The vectors are
in constant modification, and a vector that includes the best features of both
the P1 and BAC systems is shown in Figure 2.2.
G
ENERATION OF
BAC L
IBRARIES
The source DNA for BAC libraries can be isolated in at least two ways.
Extractions can be made from whole cells, in which case the organellar
genomes will comprise a substantial fraction of the clones. Alternatively,
nuclei can be isolated and then the DNA purified from these nuclei. In this
latter case any organellar sequences identified in the library are likely to
come from copies that have been integrated into the nuclear genome. The
steps involved in preparing the BAC library include:
∑ The megabase-sized DNA is isolated from cells or nuclei.
∑ The DNA is then embedded in agarose plugs and partially digested
by the restriction enzymes of choice.
∑ The size range of 100–350 kb from the partially digested DNA is
selected after separation of the partial digest on a gel.
∑ A second size selection can be performed if required to eliminate
small trapped fragments from the first gel run.
∑ The size-selected DNA is then ligated with a BAC vector of choice,
the latter having been first digested with the appropriate enzyme and
2 8
2. T
H E
B
A S I C
T
O O L B O X
— A
C Q U I R I N G
F
U N C T I O N A L
G
E N O M I C
D
ATA

---

then dephosphorylated; the ligation mixture is then electroporated
into the appropriate Escherichia coli host strain.
∑ BAC transformants are then usually selected on LB plates containing
an antibiotic, X-Gal, and IPTG. 
∑ White recombinant colonies are picked robotically and stored as 
individual clones in 96- or 384-well microtiter plates as glycerol 
stocks at –80°C. 
∑ The library can then be replicated to provide working copies and a
master (original) copy. 
∑ Before extensive use, the library should be evaluated for at least three
quality factors:
∑ Insert size distribution 
∑ Chloroplast and mitochondrial DNA content 
∑ Genome representation as determined by screening the library
with single-copy markers that are dispersed throughout the
genome. 
C
L O N I N G
S
Y S T E M S
2 9
PCYPAC2N
18754 bp
SACBII
KANAMYCIN
SP6
P1 Lytic Replicon 
T7
TN903
PBR322-SEQUENCES
PBR322
PBR322
LOXP
PUC-LINK
SACB PROMOTER
PLASMID
SacII
BamHI (1)
BamHI (2739)
NotI (2778)
NotI (18720)
XhoI (9158)
XhoI (13985)
ScaI (21)
ScaI (1789)
ScaI (2723)
EcoRI (2692)
EcoRI (5733) 
EcoRI (7917)
EcoRI (15982)
EcoRI (16468)
EcoRI (16574)
HindIII (57)
HindIII (3748)
HindIII (4197)
HindIII (4709)
HindIII (7523)
HindIII (8770)
HindIII (13465)
HindIII (16838)
FIGURE 2.2.
Map of the PAC/BAC vector PCYPAC2N from
http://www.chori.org/bacpac/pcypac2.htm.

---

Full protocols for generating such libraries are available from various 
sources (Peterson et al., 2000; http://hbz7.tamu.edu/homelinks/tool/
bac_content.htm; http://www.research.cornell.edu/Biotech/BIBAC/BIBA
CHomePage.html).
U
TILIZATION OF
BAC L
IBRARIES
The BAC libraries can have a number of different uses. 
∑ One increasingly popular use is to fingerprint large numbers of BACs
to develop physical maps. This is described in detail in Chapter 3. 
∑ Libraries can also be used to isolate “similar” (potentially syntenic)
regions across species. To do this, the library must be screened 
with sets of specifically designed probes. The probes are usually
either cDNAs or oligonucleotides that have been conserved 
across species, the latter termed “overgo oligos.” A program 
available at http://www.mousegenome.bcm.tmc.edu/webovergo/
OvergoDescription.asp can be used for designing overgo probes. 
Relatively nonredundant sequences are usually selected for the
design of a pair of overlapping oligos (overgos) for each sequence,
either from expressed sequence tag (EST) sequences or BAC end
sequences. The screenings are performed by hybridization, usually to
high-density nylon filters onto which a large number of BAC clones
have been spotted. The clones are usually double spotted to avoid
false positives, because a significant amount of work is involved in
analyzing each positive BAC clone. Overgos (in a microtiter plate, for
example) can be pooled and the BAC filters hybridized with the
labeled pool. The BACs that hybridize to the pooled overgo probes
can be rearrayed and hybridized with columns and rows from the
original pooled plate. In this way, specific BACs that hybridize to spe-
cific overgo probes can be identified in just n + m + 1 hybridizations
(where n and m are the number of rows and columns in the original
plate). Thus all 384 oligos in a 24 ¥ 16-well plate can be unambigu-
ously assigned to specific BACs with just 41 hybridization reactions
(see Figure 2.3). 
∑ This approach can be used to test synteny across species. The syn-
tenic approach involves the selection of BAC clones either containing
a known gene or located at particular positions on a specific chro-
mosome. The sequence comparisons of the chosen BACs and the
chromosomes will provide evidence for the conservation, or other-
wise, of these regions across ever-increasing evolutionary distance.
Obviously, for this approach to be viable it is necessary to have a large
BAC library for each of the species under investigation because 
syntenic regions, almost by definition, must contain low, or single-
3 0
2. T
H E
B
A S I C
T
O O L B O X
— A
C Q U I R I N G
F
U N C T I O N A L
G
E N O M I C
D
ATA

---

C
L O N I N G
S
Y S T E M S
3 1
5
′
3
′
3
′
5
′
24-mer
24-mer
Overgo Oligo Labeling and Hybridization
Approximately 40-mer selected for overgo generation. 
Two 24-mers with 8 bp overlap synthesized
8 bp overlap
Fill in with Klenow and labeled
dATP and dCTP
Double stranded 40-mer
* * * * * *
* * * * * *
Row 1
Labeled oligos arrayed in microtiter plate
Column 1
Rearrayed BACs hybridized
with pooled labeled oligos
from wells in column 1
Rearrayed BACs hybridized
with pooled labeled oligos
from wells in row 1
The 2 BACs spotted in duplicate all contain the oligo in well #1.
FIGURE 2.3.
BAC screening with overgo oligo probes. The overgo probes are
labeled and pooled. Each pool is then hybridized to a high-density filter of BAC
clones to identify which of the BAC clone(s) contain each of the overgo oligos.

---

copy, sequences for their initial selection. The development and
public availability of BAC libraries for many species are currently
being funded, and strategic decisions must be taken as to which
would be the most informative libraries to have (see
http://ucjeps.berkeley.edu/bryolab/greenplantpage.html). 
∑ BACs containing a high density of genes can be selected by using
hybridization with expressed sequences to determine the gene 
richness of the region. This is certainly one way of trying to identify
gene-rich regions of the genome, as a prelude to sequencing the
“important” regions of large complex plant genomes (see Chapter 3).
One of the decisions to be made in the development of any BAC resource
is the identity of the particular plant or line that will be the source of the DNA
from which the BAC library will be prepared. In many cases this is a com-
munity (i.e., those interested in the particular plant species or family) deci-
sion as to which would be the most appropriate starting point. Once the initial
libraries have been prepared it will be easier to use the information garnered
from these libraries to survey a much larger diverse population to understand
the amount of variation that is available. An example of this process in action
is the choice of the two initial varieties for the preparation of BAC libraries
from banana (Musa). The Global Musa Genomics Consortium, a publicly
funded group of 27 collaborating organizations from 13 countries, was
launched in July 2001. As a result of the deliberations of this group two
banana BAC libraries, one from M. acuminata “Calcutta 4” and the other from