Species Identification by Cytochrome-b, COI Barcoding and 16S rRNA

The cytochrome-b gene encodes a transmembrane protein in the mitochondrial electron transport chain. Its forensic value has nothing to do with its biochemical function. What matters is that it sits on the mitochondrial genome (present in hundreds to thousands of copies per cell, making it recoverable from trace material), it is 1,140 base pairs long in most vertebrates, its ends are flanked by conserved primer-binding sites, and its middle third accumulates substitutions at a rate sufficient to distinguish species while remaining stable enough to be sequenced reliably in degraded samples.

The standard forensic amplicon targets approximately 350-400 base pairs within the 3' portion of the gene, using universal vertebrate primers such as L14724 and H15915 (Kocher et al., 1989) or their shorter derivatives. This fragment reliably discriminates between closely related species. Tiger (Panthera tigris) and leopard (Panthera pardus) differ at cyt-b by roughly 5-7%, enough for unambiguous identification even from degraded trophy material. Elephant species (Loxodonta africana vs Elephas maximus) differ at cyt-b by approximately 9%, which matters when a tusk is seized and its provenance needs to be placed under the Appendix I listing of CITES. Domestic cattle (Bos taurus), water buffalo (Bubalus bubalis), and horse (Equus caballus) can all be distinguished using 359 bp cyt-b fragments with a simple PCR-RFLP approach, which was the basis of the first widely reported horse-meat adulteration testing.

In 2013, a Food Safety Authority of Ireland investigation found horse-meat DNA in 37% of beef-burger products sampled from major UK and Irish retailers. Subsequent European Food Safety Authority-coordinated testing across 27 member states used cyt-b PCR-RFLP and sequencing to confirm that horse meat originating from Romania and Poland had entered the beef supply chain at multiple points. The UK Food Standards Agency used the same marker set. In India, the Food Safety and Standards Authority of India (FSSAI) has adopted species-identification methods for meat products, with cyt-b cited in its 2019 food testing protocols.

The COI barcode is the 658 base-pair Folmer fragment of the cytochrome-c oxidase subunit I gene, amplified with the LCO1490/HCO2198 primer pair (Folmer et al., 1994) or one of several derivative primer sets developed for difficult groups. COI evolves faster than cyt-b in many invertebrate lineages, making it better for species discrimination across the animal kingdom as a whole, though cyt-b remains preferred for mammals and birds where the primer sets are more extensively validated.

The Barcode of Life Data System (BOLD Systems, boldsystems.org) was established at the University of Guelph in 2003 with the explicit purpose of hosting COI barcodes linked to voucher specimens, collection locality, and taxonomic authority. As of 2024, BOLD holds reference barcodes for over 250,000 species spanning insects, fish, birds, reptiles, and mammals. A forensic query against BOLD uses the Species Level Barcode Records (SLBR) dataset: a hit sharing 98% or more sequence identity with a BOLD reference barcode is treated as probable species-level identification in most published validation studies; below 95% indicates genus or higher-level assignment.

Forensic laboratories using COI include the US National Fish and Wildlife Forensics Laboratory (USFWFL) in Ashland, Oregon, which validates sequences against both BOLD and GenBank before reporting a species identification to the agent handling a CITES casework file. In Australia, the Department of Agriculture uses COI for rapid species screening at mail parcels suspected of containing protected wildlife parts. In India, the Wildlife Institute of India (WII, Dehradun) has a molecular systematic laboratory that uses both cyt-b and COI for wildlife-crime casework, and its protocols have been admitted in cases heard before High Courts and the National Green Tribunal. The admissibility standards for sequence-based identification evidence overlap with the broader Daubert framework examined in admissibility and ethics: Daubert, Frye and ELSI.

Fish species identification provides a textbook COI casework context. Sea bass (Dicentrarchus labrax) and Nile perch (Lates niloticus) cannot be distinguished once filleted and skinned. In 2020, a UK Food Standards Agency investigation used COI barcoding to confirm that 35 of 116 fish samples sold as "sea bass" in UK restaurants were misidentified, including cheaper species such as basa (Pangasius bocourti). In a parallel investigation by the Food Safety Authority of Ireland, the same marker confirmed substitution of Atlantic cod (Gadus morhua) with Pacific pollock (Theragra chalcogramma) in fish-and-chip products.

The bacterial ribosome is a 70S particle composed of a 30S small subunit (carrying the 16S rRNA gene, approximately 1,550 base pairs) and a 50S large subunit (carrying 23S and 5S rRNA genes). The 16S rRNA gene is present in one to fifteen copies per bacterial genome, is essential for cell survival, and therefore evolves slowly at its core while accumulating species-diagnostic variation in nine hypervariable regions designated V1 through V9. For forensic species identification, the V3-V4 region (approximately 470 base pairs) amplified with the 341F/806R primer pair provides sufficient resolution to distinguish bacterial species in mixed environmental samples using Illumina MiSeq.

SILVA (silva-project.eu) curates aligned 16S, 18S, and 23S/28S rRNA sequences with quality filtering, chimera checking, and taxonomic annotation updated through a consortium release cycle. The Ribosomal Database Project (RDP, US) and Greengenes (now Greengenes2, 2022, Australia/US) are the other major reference sets. For forensic biothreat attribution, the NCBI RefSeq collection of fully sequenced bacterial genomes provides the highest-resolution reference because it allows whole-genome or multi-locus comparison beyond the 16S region.

The Bacillus anthracis identification problem illustrates the limitation of 16S rRNA at intraspecies level. Bacillus anthracis is genetically near-identical to Bacillus cereus by 16S rRNA sequence. The two species share more than 99.5% 16S sequence identity and cannot be distinguished by 16S alone. The Amerithrax case (Section 4 of this module's microbial forensics topic) required whole-genome sequencing and morphotype analysis to go beyond species and reach the strain-level attribution that built the case. For identification purposes in a clinical or border-control context, however, 16S rRNA reliably places an unknown isolate at the genus level and often at the species level for the approximately 95% of bacterial species that are well-separated in 16S sequence space.

The ITS region, consisting of ITS1, 5.8S rRNA, and ITS2 flanked by the 18S and 28S subunits, is the primary barcode target for fungi. The UNITE database (unite.ut.ee, hosted at the University of Tartu, Estonia) curates fungal ITS sequences linked to species hypotheses. ITS1-5.8S-ITS2 is amplified with the ITS1/ITS4 primer pair (White et al., 1990). Forensic mycology uses ITS barcoding to identify fungal species from environmental samples in cases where mould growth patterns, food contamination, or spore profiles contribute to a forensic question. Palynology, covered in the plant forensics section of this module, uses a different set of nuclear markers (Section 5).

BOLD Systems, SILVA, UNITE, and GenBank differ in their curation strategies, and those differences matter in a court setting where a defence expert can challenge the reference entry. BOLD requires that a submitted barcode be linked to a physical voucher specimen with a determined taxonomy from a specialist. SILVA applies automated and manual quality filters to rRNA sequences and releases versioned reference datasets that can be cited by accession. GenBank accepts submissions from any registered user with minimal quality checks beyond format compliance, meaning that a GenBank BLAST hit carries less inferential weight than a BOLD or SILVA hit unless the reference entry itself is scrutinised.

In a forensic context, the standard practice at accredited laboratories is to query unknown sequences against at least two reference databases and to report the taxonomy only when both return consistent results above a defined identity threshold. The USFWFL in Ashland requires a minimum of 98% identity with a reference sequence in BOLD or GenBank, supported by a phylogenetic tree showing the unknown clustering within a species-level clade, before reporting a species identification in writing to law enforcement. The UK Centre for Environment, Fisheries and Aquaculture Science (CEFAS) follows a similar dual-database requirement for fish species identification.

The CITES framework places molecular identification results into a prosecution context. Under CITES (the Convention on International Trade in Endangered Species of Wild Fauna and Flora, 1973), species are listed in Appendix I (trade prohibited), Appendix II (controlled trade), or Appendix III (national protection). A laboratory report identifying seized material as Panthera tigris (Appendix I) or Elephas maximus (Appendix I) is a material finding in any prosecution under CITES-implementing national legislation: the US Endangered Species Act, the UK Control of Trade in Endangered Species (Enforcement) Regulations, or India's Wildlife Protection Act 1972 (Schedule I). The molecular species identification must therefore meet an evidentiary standard, including documentation of the method, the database version consulted, and the identity score returned.

The 2013 European horsemeat scandal exposed a supply-chain fraud stretching across at least eight countries. The trigger was a Food Safety Authority of Ireland (FSAI) study that detected horse-specific DNA in beef products using a real-time PCR assay targeting a 117 base-pair fragment of horse cyt-b, run alongside a reference real-time PCR assay for cattle cyt-b to confirm that beef DNA was also present. The quantitative PCR approach (horse-to-beef DNA ratio) showed that some products contained horse meat at proportions exceeding 80% of total meat content. Subsequent investigations using Sanger sequencing and BLAST against GenBank confirmed species identification in regulatory submissions to the European Food Safety Authority (EFSA).

The regulatory framework that governs these investigations differs across jurisdictions. In the European Union, Regulation EC No 1169/2011 requires that the species composition of meat products be accurately labelled. Enforcement is the responsibility of national food safety authorities operating under the Official Controls Regulation (EU) 2017/625. In the UK after Brexit, the Food Safety Act 1990 and the Food Information (England) Regulations 2014 cover the same ground. In India, the Food Safety and Standards (Labelling and Display) Regulations 2020, enforced by FSSAI, require species declaration on meat product labels. In the US, the Federal Meat Inspection Act and USDA Food Safety and Inspection Service regulations require accurate species labelling, and FSIS maintains a molecular methods database for species authentication.

The forensic methodology for food fraud cases parallels wildlife casework. DNA is extracted from a processed food sample (heat treatment, grinding, and pH extremes reduce DNA quality, so short amplicons are preferred). A species-specific PCR or multiplex real-time PCR is used to detect and quantify target species. Sequencing is performed if the real-time PCR result is ambiguous or contested. The result is reported against a validated method, and the validation documentation must accompany any regulatory or prosecutorial submission.

In India, cyt-b and COI have been used in cases involving illegal trade in beef from protected bovine species (gaur, Bos gaurus) and camel (Camelus dromedarius), which is restricted in several states. The Central Forensic Science Laboratory (CFSL, New Delhi) and several state FSLs have validated molecular species identification protocols for meat casework. In Singapore, the Agri-Food and Veterinary Authority has adopted COI barcoding for import control of fish products.