Molecular Identification of Forensically Important Insects

Forensic entomology's postmortem interval estimates rest on knowing which species laid the eggs or colonised the body. Get the species wrong and the development table you use is wrong too. For common blow flies in familiar habitats an experienced entomologist working with fresh, well-preserved adult specimens can make reliable identifications by morphology alone. But those ideal conditions do not describe most casework.

• Early-instar larvae: first and second instars of many forensically important species are morphologically indistinguishable from each other. The posterior spiracle patterns used to key third instars are not yet differentiated.
• Puparia: the pupa is sealed inside a hardened casing derived from the last larval skin. No external morphological features allow species identification without dissection, and dissection destroys the specimen.
• Damaged adults: postmortem scavenging, desiccation, mould, and handling can strip the wings, antennae, and setae from adult specimens, removing the key features relied on for morphological keys.
• Species complexes: genera like Calliphora and Lucilia contain sibling species with overlapping morphology but different thermal development rates, so misidentification directly corrupts PMI calculations.

These failure modes were recognised decades before molecular tools existed. Entomologists reared larvae to adulthood before identifying them, or preserved them in formalin for histology. Both approaches cost time. Rearing introduces risk: a larva that dies during rearing, or that moulted unseen and is now a different instar than the collected notes say, is not just an identification problem but an evidence integrity problem.

The COI barcode region used in animal identification was standardised by Paul Hebert and colleagues at the University of Guelph in 2003, when they demonstrated that a ~658 bp region of the mitochondrial COI gene could discriminate species across a wide taxonomic range. For Diptera, the group containing almost all forensically important flies, inter-species divergence typically exceeds 2% while intra-species divergence stays below 1%, creating a gap that allows reliable assignment.

1. Tissue sampling
   A small tissue piece, typically a leg segment or a small fragment of larval cuticle, is removed from the specimen. The rest of the specimen is retained as a voucher. For larvae one leg from a frozen or ethanol-preserved specimen is standard; the remaining specimen stays in the case file.
2. DNA extraction
   Commercial kit-based extraction (silica-column or CTAB method) releases DNA. Ethanol-preserved specimens yield good quality. Formalin-fixed specimens yield poor or no DNA because formalin cross-links nucleic acids with proteins.
3. PCR amplification
   Universal dipteran primers (LCO1490 / HCO2198 or the improved versions C_LepFolF / C_LepFolR) amplify the target region. Degraded samples may need nested PCR or short-amplicon primers designed for low-molecular-weight template.
4. Sequencing and alignment
   Sanger sequencing produces a trace file that is trimmed, quality-checked, and assembled into a consensus sequence. The consensus is aligned to reference sequences in BOLD or NCBI GenBank.
5. Database query and interpretation
   A BOLD ID Engine search returns ranked matches with percent similarity. A match above ~98% is generally accepted as species-level; 95-97% suggests genus-level only; below 95% the identification is unreliable and must be flagged. The analyst reports the best match, the similarity score, and the depth of database coverage for the target taxon.

BOLD (Barcode of Life Data System, hosted at the University of Guelph) is the primary reference for COI-based identification. At the time of writing it contains millions of specimen records with associated sequences, images, and collection data. Critically, BOLD ties every sequence to a physical voucher specimen whose morphological identification has been verified by the submitting researcher. This traceability is the system's main quality advantage over the general NCBI GenBank.

GenBank accepts deposited sequences without the same curation requirement. It is larger than BOLD and essential as a backup, but it contains sequences with misidentified vouchers, sequences derived from reared specimens without confirmed wild-type identity, and sequences lacking geographic metadata. Searching GenBank via BLAST and then cross-referencing with BOLD is the standard two-step approach when an initial BOLD query returns ambiguous or low-similarity results.

An important practical scenario: a case involves a cluster of puparia collected from under a body that had been outdoors for several weeks. The casings are dark brown and opaque, the internals hardened. No morphological features are accessible without destroying the specimen. Molecular sampling, removing one hind leg from each of several casings while keeping the rest intact, yields enough DNA for COI sequencing from most specimens, and the voucher puparia remain available for independent review.

For first-instar larvae, the tissue available per specimen is minimal. Options include extracting from a single specimen and hoping for enough yield (works for well-preserved larvae stored in 95% ethanol at -20°C), or pooling two or three larvae from a single collection point when the analyst has confirmed through microscopy that they appear to be the same species. Pooling introduces risk if the collection contains a mix of species, so it must be reported and justified in the case notes.

• Field preservation protocol: kill larvae in near-boiling water (prevents tissue contraction) then transfer to 95% ethanol. This is the single highest-impact step for downstream molecular work.
• Temperature during transport: storing preserved specimens at 4°C or -20°C slows enzymatic degradation and maintains DNA integrity for months to years.
• Subsample early: take a molecular subsample at collection when possible, before morphological processing. Dissection and slide-mounting consume or destroy tissue.

COI works for the large majority of forensically important Diptera. But in species complexes where COI alone cannot discriminate, additional markers help. Cytochrome b (cytb) is the most commonly used secondary mitochondrial marker in forensic entomology. Nuclear markers, including ITS2 (internal transcribed spacer 2), have been validated for specific blow fly groups.

Next-generation sequencing (NGS) and mitogenome approaches are appearing in the research literature. Whole mitochondrial genome sequencing provides far higher resolution than a single-gene barcode and is increasingly cost-competitive as sequencing prices fall. For casework it is not yet routine, partly because reference mitogenomes are scarce for many taxa, partly because the bioinformatics pipeline requires expertise not present in most forensic labs. It is a technology to watch rather than one in daily use.

Species-specific qPCR assays have been developed for a handful of the most common blow fly species. They are faster than Sanger sequencing, amenable to degraded template, and interpretable in a lab without sequencing infrastructure. Their limitation is specificity: each assay targets one species, so an unexpected species on a case produces a negative result that is misread as no blow fly rather than no Calliphora vicina. Assay panels covering a region's common species can mitigate this, but they require regional validation studies.

Forensic scientists are trained to report findings with uncertainty quantified. Molecular identifications must follow the same discipline. A BOLD match at 99.1% similarity to Calliphora vicina is strong but not absolute. The analyst should report the similarity score, the number of database entries queried, the next closest species and its similarity, and any known coverage gaps in the database for that geographic region.

In a typical evidence report the entomologist presents: (1) the morphological assessment where possible; (2) the molecular result with similarity score and database name; (3) a combined conclusion that names the species or, where uncertainty remains, the genus or species complex; (4) what that identification means for the development table used in the PMI calculation. Separating these four elements prevents the jury from treating a molecular match as an absolute certainty where it is not.