Internet Contig Explorer (iCE)—A Tool for
Visualizing Clone Fingerprint Maps
Christopher D. Fjell,1Ian Bosdet, Jacqueline E. Schein, Steven J.M. Jones,
and Marco A. Marra
Genome Sciences Centre, British Columbia Cancer Agency, Vancouver, BC V5Z 4E6, Canada
Fingerprinted clone physical maps have proven useful in various applications, supporting both whole-genome
and region-specific DNA sequencing as well as gene cloning studies. Fingerprint maps have been generated for
several genomes, including those of human, mouse, rat, the nematodes Caenorhabditis elegans and Caenorhabditis
briggsae, Arabidopsis thaliana and rice. Fingerprint maps of other genomes, including those of fungi, bacteria,
poplar, and the cow, are being generated. The increasing use of fingerprint maps in genomic research has
spawned a need in the research community for intuitive computer tools that facilitate viewing of the maps and
the underlying fingerprint data. In this report we describe a new Java-based application called iCE (Internet
Contig Explorer) that has been designed to provide views of fingerprint maps and associated data. Users can
search for and display individual clones, contigs, clone fingerprints, clone insert sizes and markers. Users can
also load into the software lists of particular clones of interest and view their fingerprints. iCE is being used at
our Genome Centre to offer up to the research community views of the mouse, rat, bovine, C. briggsae, and
several fungal genome bacterial artificial chromosome (BAC) fingerprint maps we have either completed or are
currently constructing. We are also using iCE as part of the Rat Genome Sequencing Project to manage our
provision of rat BAC clones for sequencing at the Human Genome Sequencing Center at the Baylor College of
DNA clone fingerprint maps (Olson et al. 1986; Sulston et al.
1988; Marra et al. 1997; Marra et al. 1999) have proven useful
in various genomics investigations and hence the demand for
these maps has increased steadily. Fingerprint maps either
have been or are being generated for the genomes of many
intensively studied organisms, including human (McPherson
et al. 2001), mouse (Gregory et al. 2002), the model plant A.
thaliana (Marra et al. 1999), the laboratory rat (J. Schein and
the BCCA Genome Sciences Centre, unpubl.), Caenorhabditis
elegans (Coulson et al. 1986, 1991), and Caenorhabditis briggsae
(Marra et al. 1997). In addition, fingerprint maps either have
been or are being generated for the genomes of organisms of
agricultural importance, including cow (J. Schein and the
BCCA Genome Sciences Centre, unpubl.), rice (Mao et al.
2000), sorghum (Klein et al. 2000), and corn (Coe et al. 2002).
Also under construction are fingerprint maps of the genomes
of fungal and bacterial pathogens, including those of C. neo-
formans (Schein et al. 2002) and Ustilago hordeii (J. Kronstad
and M. Marra, unpubl.), the fungus Magnaporthe grisea (Zhu et
al. 1999), and Chlamydia (I. Bosdet, M. Marra, and B. Brun-
ham, unpubl.). Fingerprint maps for additional genomes are
The software most heavily used for analysis of fingerprint
data and display of fingerprint maps is FPC (Soderlund 1997,
2000; http://www.genome.arizona.edu/software/fpc). This
powerful and versatile software package, running under UNIX
or LINUX, has become the standard for genome map con-
struction and editing. We offer via FTP our maps in FPC for-
mat, so researchers can download the data and view them
using a locally installed copy of the FPC program. However, as
a consequence of its power and versatility, FPC is complex
and this, coupled with the requirement of a UNIX-based ar-
chitecture to install and run the software, has presented dif-
ficulties for some investigators wishing to view clone finger-
print databases. In addition, the databases may be large (up to
several hundred megabytes in size) and may frequently
change as data are updated. In our experience many investi-
gators often wish only to search the fingerprint maps for spe-
cific clones or markers and view these and the genomic seg-
ments (contigs) to which they belong. Due to the increasing
use of the clone fingerprint approach to physical map con-
struction and the corresponding increases in both the number
of genomes mapped and the number of researchers requiring
access to these maps, we saw a need for a simplified Internet-
There are several existing services available to view physi-
cal maps via the Internet. Web-FPC offers a limited view of
physical maps similar to FPC for maps such as rice, maize,
sorghum, zebrafish, and Arabidopsis thaliana (http://
www.genome.arizona.edu/software/fpc/). Other sites such as
Ensembl (http://www.ensembl.org/) and NCBI (http://
www.ncbi.nlm.nih.gov/mapview/) also offer views of BAC
maps integrated with sequence and other information. How-
ever, these existing tools did not provide the full functionality
we desired and were not adaptable to the more distant future
needs we foresaw. Therefore, the Internet Contig Explorer
(iCE) was devised to fill this need.
The aim of iCE was not to recreate the broad scope of
functions available already in FPC. Instead, our goal was to
provide a viewing system sufficient to satisfy most of the in-
vestigators who wished to browse the fingerprint data and the
maps built from them without the requirement and overhead
of downloading and updating datasets. In designing iCE we
E-MAIL firstname.lastname@example.org; FAX 604-877-6085
Article and publication are at http://www.genome.org/cgi/doi/10.1101/
1244 Genome Research
13:1244–1249 ©2003 by Cold Spring Harbor Laboratory Press ISSN 1088-9051/03 $5.00; www.genome.org
considered our previous interactions with investigators and
the most frequently requested types of information. As well,
we found we had novel requirements for managing our pro-
vision of rat BAC clones for sequencing at the Human Ge-
nome Sequencing Center at the Baylor College of Medicine.
Here we describe the design and implementation of iCE and
illustrate some of the features of the software.
The iCE system was designed to meet the immediate needs
of users to access physical map data, and provide an easily
maintained and extensible platform capable of future expan-
sion. For this reason, the Java programming language was
chosen for developing the software and an SQL database was
chosen for data storage. Both of these technologies are widely
used in the software development community and provide
robust and well-developed tools for development of the iCE
The iCE system is composed of two parts: a client Java
application and an SQL database. The client application runs
on the user’s machine, accessing data stored remotely on
the database server. The SQL data originates from one or
more FPC (Soderlund 1997, 2000; http://www.genome.
arizona.edu/software/fpc/) databases and is imported into the
SQL database using a separate C application (fpc_sql), which
is included as part of the iCE source distribution. To optimize
the delivery of data to the client application, a caching
scheme is implemented to preprocess SQL queries and store
the results as files on the iCE Internet server. These files are
compressed Java objects that are transferred and loaded into
the client application with minimal load on the client ma-
chine. The user has the option to allow duplicates of these
cache files to be added to their local disk storage to eliminate
delays due to transmission across the Internet. Data is trans-
mitted only as needed by the user, which is typically only a
small fraction of the database. In addition, to minimize the
data transferred, gel images are not downloaded until re-
quested by the user.
iCE was designed to give researchers easy access to the map
data through a graphical user interface. The main display win-
dow contains lists of all contigs, clones, and markers from the
database. The user may view a contig or clone by selecting it
from one of the lists as described below. Also, the user may
choose to search for contigs that contain clones matching a
particular marker. Each time the user selects a contig to view,
the contig is displayed in a new window. Each display shows
the orientation of clones with respect to one another, and all
markers and comments associated with each clone. The posi-
tions of restriction fragments are also displayed for all clones,
in the order determined by the clone positions in the contig.
This allows the user to see the individual restriction fragments
shared between clones. The gel images, which were analyzed
to determine the positions of restriction fragments, can also
On start-up, iCE connects via the Internet to the data-
base server and downloads the list of clone names, contigs,
and markers for the selected database and displays this infor-
mation in the main viewing frame (see Fig. 1). In addition,
user-defined lists are also shown: These user lists are arbitrary
lists of clones not necessarily associated with a particular con-
tig or marker. For example, these lists may represent clones
selected for DNA sequencing. The main viewing frame also
contains an Options tab where the user may customize the
displays, for example, specifying the maximum
number of clones to be displayed, allowing da-
tabase comments to be shown or changing the
magnification of the electrophoretic gel im-
The names of contigs, clones, markers,
and user-defined lists are shown in list boxes
on the main window (Fig. 1). To view an item
from one of these lists, the user selects an item
from the list with the mouse or types the name
in the appropriate text field. When first re-
quested, a contig, individual clone, or user list
will be displayed in a new display window. If
the item has already been displayed, the dis-
play will be brought to the front. If a contig is
requested, the contig and associated data are
downloaded. When a clone is selected, if the
clone is already displayed in a contig, the con-
tig window is brought to the front of the other
windows and the clone is highlighted. Other-
wise, data for the single clone is downloaded
and the clone is displayed within a list win-
dow, titled Miscellaneous. For a selected
marker, a list of all contigs containing clones
associated with the marker are determined
from the database. The user is then prompted
to select contigs from the list to display.
Each contig or list of clones is displayed in
a separate contig display frame, as show in Fig-
ure 2. Each contig display is listed in the drop-
and markers for the current FPC database are shown in the first three lists on the left side.
Arbitrary lists of clones, called user lists, are in the next list, with a log of application activity
on the right side of the frame. Options are available by choosing the Options tab. The top
of the frame displays the database and FPC build and contains a drop-down list to change
to a different contig display.
Main viewing frame of iCE, showing the Database pane. The contigs, clones,
Internet Contig Explorer
Contig display for an arbitrarily chosen contig in the mouse FPC database. The layout of the clones within the contig is shown on the left, with colored boxes representing the left
and right ends of the clone in the top area. Markers are shown in blue in the middle area, with the left and right ends indicating the left- and right-most ends of clones matching this marker. Remarks
for the clones are show in the bottom area. The processed gel images for the clones are shown on the right side of the display, with restriction fragment locations indicated with horizontal lines.
Green lines indicate restriction fragments confirmed by neighboring clones and red indicates unconfirmed restriction fragments. A ruler on the left side of the gel image area indicates the size
of restriction fragments (shown here) or mobility depending on the software configuration. Restriction fragments marked by the user are labeled with (x) symbols.
Fjell et al.
1246 Genome Research
down box at the top of the main frame, allowing
for convenient navigation between contigs. The
left side of a contig display frame is divided hori-
zontally into three areas. The area at the top dis-
plays the clones in the contig as colored boxes
with clone names. The left and right ends of the
boxes indicate the position of the clone as deter-
mined by consensus band map position in the
original FPC database (Soderlund et al. 1997). The
two areas below display markers and comments
associated with the clones. Selecting an item is
done with a left mouse click. When an item is
selected, the items associated with it are also
highlighted: Selecting a clone highlights the
markers and comments corresponding to the
clone; selecting a marker highlights the clones
associated with the marker. Clone boxes change
color to indicate selection state (green for unse-
lected, and yellow for selected). Border style iden-
tifies the clones as being parental, canonical, or
buried (solid black, etched, or no border, respec-
tively); these terms are as described in Soderlund
et al. 1997. Icons at the left end of the clone box
identify the clones as belonging to a user-defined
list (red dot), having an associated marker (blue
dot), or submitted for DNA sequencing (colors).
Pop-up menus with additional options for
modifying the clone and contig displays are
called up by right clicking on a clone box or
within the contig display, respectively. Buried
clones may be displayed or hidden; the vertical
order of the clones can be sorted by name, size,
left or right position within the contig, or the
position specified in a user list. Clones may also
be copied to new, temporary contig displays to
reduce the number of clones on an individual
display and allow a more flexible comparison of
restriction fragment locations between clones
that may not be located in a single contig (de-
scribed below) . Detailed information for clones
(Fig. 3) and contigs (Fig. 4) are also available from
these pop-up menus, described below. The contig
for a particular clone can also be requested from
a pop-up menu, allowing convenient navigation
from clones in user lists to their respective con-
The right side of a contig display contains
the electrophoretic gel images (Marra et al. 1997)
of the fingerprinted clones. Selection of a clone
name or gel image also selects the corresponding
clone on the left-hand side of the contig display
frame. If gel images were requested when the
contig display was created, these images and cor-
responding clone names are displayed; otherwise
the area underneath each clone name is initially
blank. To load a single gel image, the user double-
clicks the blank area under the clone name. To
load gel images for all clones on the display, the
user selects the “All images” button at the top of
the gel image display. Horizontal, colored lines
are drawn beside identified restriction fragments.
Green and red lines indicate restriction fragments
that are confirmed or unconfirmed, respectively,
by restriction fragments in neighboring clones, as
the parent to this clone, the clone type, number of restriction fragments (bands) and
remarks are shown at the top. Any markers associated with this clone and any clones
buried in this clone are listed below. At the bottom, the mobility and size of each
restriction fragment for the clone fingerprint are listed. The total size of restriction
fragments for this clone is displayed above the list of restriction fragments, along with
the sum of sizes of selected restriction fragments, colored purple. Fragments are se-
lected by selecting rows from the table. Restriction fragments that are unconfirmed or
marked by the user are identified by (*) in the respective columns.
Details for a mouse BAC clone. The contig and position within the contig,
Internet Contig Explorer
determined according to the current iCE configuration pa-
rameters. This set of neighboring clones is either a specified
number of clones to the left and right of the clone, or all
clones in the contig that are sufficiently similar to the clone,
depending on the current iCE configuration. Similarity be-
tween clones is calculated as the probability of seeing the
observed bands at the same positions for two clones by
chance (Sulston et al. 1988).
Restriction fragments of interest to the user can be
marked by clicking near the colored lines; marked restriction
fragments are indicated with an “x”. These marked fragments
are also identified on the clone details display so the size and
mobility of marked fragments can be determined. Normally,
the positions of restriction fragments for different clones are
only compared for clones within the same contig as described
above. To determine the restriction fragments shared between
clones that are not in the same contig, the user can copy these
clones to a separate display and perform the comparison us-
ing all clones on the display, regardless of position of the
clone in the contig or similarity between clones. This is done
using the “Confirm bands using all” button at the top of the
gel images. This allows the user to determine shared restric-
tion fragments between clones in arbitrary lists (such as user
lists) and to quickly determine the number and size of shared
restriction fragments between clones.
To view detailed information for a clone (Fig. 3), the user
selects “Show details” from the pop-up menu for a clone box.
This display includes the number of restriction fragments
(bands on the details display), position in the contig, markers,
the parent clone (burying clone), and any underlying buried
clones. The bottom of the display contains a table of the re-
striction fragment sizes and mobilities. The total size of the
restriction fragments and the size of the fragments on selected
rows are shown. Two columns of the table indicate uncon-
firmed fragments and fragments marked by the user on the
gel image. Buttons above the table allow the user to select all
fragments that are confirmed, unconfirmed, and marked.
These buttons allow the user to quickly identify sizes of
marked restriction fragments and size of restriction fragments
that are unique to a clone, and probably represent DNA not
found in neighboring clones.
To view detailed information for a contig (Fig. 4), the
user selects “Show statistics” from the pop-up menu for the
contig (this menu is raised when the user right-clicks the con-
tig background). This display includes all data available for
the contig and the clones it contains. The top display shows
information specific to the contig as generated by FPC, and
statistics on the number of clones and the sizes and number of
bands. The table at the bottom of the display shows informa-
tion for each clone on a separate line. This includes clone
name, left and right position in the contig, buried status, any
clones that are buried within it and its parent clone (burying
clone), size, and number of bands (restriction fragments), as
well as the mobilities of its restriction fragments. Columns of
data can be suppressed to prevent displaying undesired infor-
mation. This data can be printed and also written to file for
use by external software in a customizable format. For ex-
ample, the restriction fragment mobilities for a selection of
clones can be written to a file and read by spreadsheet soft-
ware such as Excel or StarOffice.
We have described iCE, a new software system for viewing
clone fingerprint mapping data at the British Columbia Can-
cer Agency Genome Sciences Centre and elsewhere. There are
now maps for ten organisms available via iCE, and these are
being used by the biological community. In the period from
October 2001 to February 2002, over ninety different external
users have accessed iCE databases, in more than five hundred
iCE was intended to incur low maintenance costs by
implementing two strategies: The iCE client was written using
the Java programming language in an object-oriented para-
digm, and the database system uses the industry-standard
SQL protocol. Since its initial conception, the iCE system has
undergone frequent changes in features and functionality
without requiring significant changes to existing database
structure or code.
Work is underway to extend iCE in several directions.
Most important are efforts to improve performance in speed
of data access and display responsiveness. It is also desirable to
allow users to continue to work with the iCE client without a
constant Internet connection. Features continue to be added
to allow for more convenient viewing and rearranging of data,
as well as better management of arbitrary lists of clones to be
used, for example, in a sequencing pipeline. The iCE software
and comprehensive documentation are available at http://
ice.bcgsc.ca. The iCE source code and related code for data-
base management is available from the authors under license,
at no cost, for academic use.
The iCE client application is written in Java 2 using the Java
Development Kit (JDK) 1.3.1 (http://java.sun.com). The Java
Runtime Environment (JRE) 1.3.1 is required on the client
machine. The Borland JBuilder 4.0 development environment
was used for code development (http://www.borland.com).
The iCE database uses MySQL DBMS Ver 8.19 (http://
www.mysql.com). The iCE client has been used successfully
top. The clones in the contig are listed in the bottom table. All clone
information can be displayed and sorted, including position of each
clone in the contig, any buried clones, clone size, and number of
bands. This data can be printed or sent to a file in a format to be read
by other applications.
Details for a contig. Values for the contig are shown at the
Fjell et al.
on Linux and Microsoft Windows 2000 on computers with an
Intel Pentium III processor and 512 MB RAM, and on Apple
computers running MacOS X. All contig, clone, and marker
data originate in FPC databases and are converted to the SQL
format using a custom application (fpc_sql) written in the C
We thank the many people who contributed to testing and
implementation of iCE at the Genome Sciences Centre.
Thanks to Justin Muir, Kirk Schoeffel, and Martin Krzywinski
for installing the iCE Web server. Thanks also to Steven Ness
for useful comments on an early version of the manuscript
and to Mike Holman at Washington University Genome Se-
quencing Center for helpful early discussions. This work was
funded by the National Human Genome Research Institute
(USA). We are grateful to the staff of the British Columbia
Cancer Agency Genome Sciences Centre for expert technical
and administrative assistance. M.A.M. is a Michael Smith
Foundation for Health Research Scholar.
The publication costs of this article were defrayed in part
by payment of page charges. This article must therefore be
hereby marked “advertisement” in accordance with 18 USC
section 1734 solely to indicate this fact.
Coe, E., Cone, K., McMullen, M., Chen, S.-S., Davis, G., Gardiner, J.,
Liscum, E., Polacco, M., Paterson, A., Sanchez-Villeda, H. et al.
2002. Access to the maize genome: An integrated physical and
genetic map. Plant Physiol. 128: 9–12.
Coulson, A., Sulston, J., Brenner, S., and Karn, J. 1986. Toward a
physical map of the genome of the nematode Caenorhabditis
elegans. Proc. Natl. Acad. Sci. 83: 7821–7825.
Coulson, A., Kozono, Y., Lutterbach, B., Shownkeen, R., Sulston, J.,
and Waterston, R. 1991. YACS and the C. elegans genome.
BioEssays 13: 413–417.
Gregory, S.G., Sekhon, M., Schein, J., Zhao, S., Osoegawa, K., Scott,
C.E., Evans, R.S., Burridge, P.W., Cox, T.V., Fox, C.A., et al. 2002.
A physical map of the mouse genome. Nature 418: 743–750.
Klein, P.E., Klein, R.R., Cartinhour, S.W., Ulanch, P.E., Dong, J.,
Obert, J.A., Morishige, D.T., Schlueter, S.D., Childs, K.L., Ale, M.
et al. 2000. A High-throughput AFLP-based method for
constructing integrated genetic and physical maps: Progress
toward a sorghum genome map. Genome Res. 10: 789–807.
Mao, L., Wood, T.C., Yu, Y., Budiman, M.A., Tomkins, J., Woo, S.,
Sasinowski, M., Presting, G., Frisch, D., Goff, S., et al. 2000. Rice
transposable elements: A survey of 73,000
sequence-tagged-connectors Genome Res. 10: 982–990.
Marra, M.A., Kucaba, T.A., Dietrich, N.L., Green, E.D., Brownstein,
B., Wilson, R.K., McDonald, K.M., Hillier, L.W., McPherson, J.D.,
and Waterston, R.H. 1997. High-throughput fingerprint analysis
of large-insert clones. Genome Res. 7: 1072–1084.
Marra, M.A., Kucaba, T., Sekhon, M., Hillier, L., Martienssen, R.,
Chinwalla, A., Crockett, J., Fedele, J., Grover, H., Gund, C. et al.
1999. A map for sequence analysis of the Arabidopsis thaliana
genome Nat. Genet. 22: 265–270.
McPherson, J.D., Marra, M., Hillier, L., Waterston, R.H., Chinwalla,
A., Wallis, J., Sekhon, M., Wylie, K., Mardis, E.R., Wilson, R.K. et
al. 2001. A physical map of the human genome. Nature
Olson, M.V., Dutchik, J.E., Graham, M.Y., Brodeur, G.M., Helms, C.,
Frank, M., MacCollin, M., Scheinman, R., and Frank, T. 1986.
Random clone strategy for genomic restriction mapping in yeast.
Proc. Natl. Acad. Sci. 83: 7826–7830.
Schein, J.E., Tanger, K.L., Chiu, R., Shin, H., Lengeler, K.B.,
MacDonald, W.K., Bosdet, I., Heitman, J., Jones, S.J., Marra,
M.A., et al. 2002. Physical maps for genome analysis of serotype
A and D strains of the fungal pathogen Cryptococcus neoformans.
Genome Res. 9: 1445–1453.
Soderlund, C., Longden, I. and Mott, R. 1997. FPC: a system for
building contigs from restriction fingerprinted clones. CABIOS
Soderlund, C., Humphray, S., Dunham, I. and French, L. 2000.
Contigs built with fingerprints, markers, and FPC V4.7. Genome
Res. 11: 934–941.
Sulston, J., Mallett, F., Staden, R., Durbin, R., Horsnell, T., and
Coulson, A. 1988. Software for genome mapping by
fingerprinting techniques. CABIOS 4: 125–132.
Zhu, H., Blackmon, B.P., Sasinowski, M., and Dean, R.A., 1999.
Physical map and organization of chromosome 7 in the rice blast
fungus, Magnaporthe grisea. Genome Res. 9: 739–750.
WEB SITE REFERENCES
http://www.genome.clemson.edu/fpc/; Web-based FPC physical
http://www.genome.arizona.edu/software/fpc/; FPC home page.
http://www.ensembl.org/; Ensembl genome browser.
http://www.ncbi.nlm.nih.gov/mapview/; NCBI Map Viewer.
http://ice.bcgsc.ca; iCE home page.
http://java.sun.com; JAVA home page, Sun Microsystems Inc.
http://www.borland.com; Borland Software Corp. (JBuilder vendor).
http://www.mysql.com; MySQL open source SQL database home
Received September 17, 2002; accepted in revised form March 14, 2003.
Internet Contig Explorer