DNA barcoding: a new diagnostic
tool for rapid species recognition,
identification, and discovery
James Hanken
Museum of Comparative Zoology
Harvard University, USA
What is DNA barcoding?
• Definition: Derivation of a short DNA sequence(s) that
enables species identification or recognition in a particular
domain of life (eucaryotes).
• Focus to date—in animals—has been on a 658 base-pair
(bp) fragment of the mitochondrial gene, cytochrome oxidase
subunit I (COI).
• The Barcode of Life Initiative (BOLI) would resolve barcodes
for named species and use a barcoding approach to assess
undescribed biological diversity.
• Very controversial!
What isn’t DNA barcoding?
• It is not intended to, in any way, supplant or
invalidate existing taxonomic practice.
• It is not DNA-taxonomy; it does not equate
species identity, formally or informally, with a
particular DNA sequence.
• It is not intended to duplicate or compete with
efforts to resolve deep phylogeny, e.g.,
Assembling the Tree of Life (ATOL).
“the role of any molecular diagnostic is to aid research, not to serve as an
end in itself. Barcoding … is independent of questions as to whether
individual taxa are species, what species are (or should be), and where they
fit in a unified tree of life…. Barcoding is not an end in itself, but will boost
the rate of discovery. The unique contribution of DNA barcoding to …
taxonomy and systematics is a compressed timeline for the exploration and
analysis of biodiversity.”
Potential applications
1) Facilitating identification and recognition of named
(described) species:
• linking life history stages, genders;
• differentiating cryptic species;
• identifying gut contents;
• human disease vectors;
• agricultural pests;
• biosecurity (?).
2) Surveying and inventorying biodiversity; e.g.,
flagging potentially new (undescribed) species.
Differentiating cryptic species
• Astraptes fulgerator, skipper butterfly.
• Wide-ranging; southern U.S. to northern
Argentina.
• In northwestern Costa Rica, comprises
complex of 10 sympatric species that
are distinct in DNA sequence (COI),
larval coloration, food plants, and subtle
morphological traits.
D. Janzen, et al., submitted
Sympatric larvae of Astraptes
fulgerator
Food plant:
Trigonia (2
species); larvae
will starve if
reared on plants
used by other
larval types.
Food plant:
Celtis iguanaea
Strengths
• Offers alternative taxonomic identification tool for
situations in which morphology is inconclusive.
• Focus on one or a small number of genes provides
greater efficiency of effort.
• Cost of DNA sequencing is dropping rapidly due to
technical advances.
• Potential capacity for high throughput and processing
large numbers of samples.
• Once reference database is established, can be applied
by non-specialist.
Weaknesses
• Assumes intraspecific variation is negligible, or at least
lower than interspecific values.
• No single gene will work for all taxa (e.g., COI is not
appropriate for vascular plants, or even for some
animals).
• Single-gene approach is less precise than using multiple
genes; may introduce unacceptable error.
• Some of the most attractive aspects rely on future
technology, e.g., handheld sequencer.
Proportion of species pairs
Between-species sequence
divergence in COI (%)
1.0
n = 13,320
1.0
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0
2 4 8 16 32 64%
Annelida, Arthropoda, Chordata, Mollusca,
Echinodermata, Nematoda, Platyhelminthes
n = 17
2 4 8 16 32 64%
Cnidaria (corals, anemones,
jellyfish, sea pens, etc.)
COI sequence divergence among
North American birds
K2P (%)
Barcoding must adhere to standards for
specimen and data management
Sequence data
Voucher specimens
and electronic
databases
Digital images
Pilot projects
1) Goals and objectives:
• Validate barcoding approach, in general, and the use of COI, in
particular (for animals); i.e., proof of principle.
• Assess feasibility of large-scale effort, e.g., identify bottlenecks,
cost, logistic issues.
2) Possible targets:
• “All taxa”—primates, turtles, mosquitoes, tephritid fruitflies, birds,
sphinx moths, salamanders, etc.
• Regional faunas, e.g., Gulf of Maine megafauna.
• Existing inventories, e.g., INBio and ACG (Costa Rica).
3) Would rely principally on museum specimens.
Role of museums (and impact)
• Barcoding must validate existing taxonomy before it can be offered
as an identification tool, and especially before it can be used to
discover new species.
• Pilot projects will utilize museum specimens.
• New inventory efforts will yield large numbers of vouchers, which
must be properly accessioned, databased, and stored.
• Results will flag many new species requiring formal description.
•
Additional burden on museums, herbaria, etc., but may also
offer new sources of support.

DNA barcoding is under way…
See also Nature 426: 514 (4 Dec 2003)
and appeals to the biotech sector
Anticipated next steps (2004)
• January: Working group submits a grant proposal to
found a coordinating Secretariat to develop a
Barcode of Life Initiative (BOLI) based in Washington,
DC; decision expected late March.
• May: Founding meeting of the BOLI Consortium
(museums and herbaria) at U.S. Smithsonian
Institution (NMNH).
• Late fall: First International Barcode of Life
Conference; venue TBD.