# Purity Checks for Reference Cultures A purity check answers one question, does this culture contain exactly one organism and nothing else, while an identity check answers a different question, is that one organism the strain you believe it to be. The rule of thumb for choosing among purity check methods is simple. Put a colonial morphology and stain check on every single subculture because it is cheap and fast, and layer on molecular or genomic methods when identity, provenance, or genetic stability must be proven rather than assumed. This article walks through that logic at the bench. It covers what a reference culture is, why purity matters, when to check, the main methods available with their pros and cons, a side-by-side comparison, worked scenarios, and the mistakes that most often undermine microbiology quality control in real laboratories. ## What a Reference Culture Is and Why Purity Matters A reference culture is a well-characterized microbial strain maintained under controlled conditions and used as a known comparator. Quality control strains for antimicrobial susceptibility testing, proficiency testing challenge organisms, seed strains for biological production, and probiotic working strains all fall under this umbrella. The defining feature is trust. Downstream results are only as credible as the assumption that the culture in the vial is what the label says, alive, and alone. That assumption fails more often than laboratories like to admit. A frozen vial can carry a low-level bacterial contaminant introduced during the original harvest. A working stock can pick up a colony picker cross-contamination event that no one notices for months. A seed strain passed repeatedly in vitro can drift genetically while still growing happily on the same medium it always did. None of these failures announce themselves in growth rate or gross appearance. The microbial genetics literature makes the stakes concrete. Work on live attenuated animal vaccine strains in South Korea sequenced the complete genomes of five bacterial seed strains recovered directly from commercial products, including *[Erysipelothrix rhusiopathiae](/knowledge/bacteria/livestock-bacteria/erysipelothrix-rhusiopathiae)*, *[Clostridium chauvoei](/knowledge/bacteria/livestock-bacteria/clostridium-chauvoei)*, two *Bordetella bronchiseptata* strains, and *Mycoplasroides gallisepticum*, specifically to verify whether manufacturers were using the same seed strains at the strain level and to support consistent mass production and quality control of biological products [1]. The same group performed the equivalent exercise for eleven bacterial strains used in inactivated vaccines, including *Pasteurella multocida*, *[Actinobacillus pleuropneumoniae](/knowledge/bacteria/livestock-bacteria/actinobacillus-pleuropneumoniae)*, *Glaesserella parasuis*, *[Mannheimia haemolytica](/knowledge/bacteria/livestock-bacteria/mannheimia-haemolytica)*, *Avibacterium paragallinarum*, and three *Escherichia coli* strains, again framing complete genome data as the basis for confirming seed strain identity over time [2]. These papers exist because seed strain fidelity is not assumed in regulated biological production. It is demonstrated. Strain-level fidelity matters outside manufacturing too. The Malaysian Society of Gastroenterology and Hepatology position statement on probiotics emphasizes that probiotic benefits are strain specific, and that quality control in manufacturing and careful strain identification are prerequisites for selecting a probiotic for clinical use [3]. A probiotic product labeled with a strain name is a claim, and purity and identity checking are how a laboratory or manufacturer backs that claim. ## Purity, Identity, and Genetic Stability: Three Different Questions ![Comparison of purity, identity and genetic stability checks by the question asked, level reached and example failure.](/article-images/illustrations/purity-check-methods-reference-cultures-microbiology-three-questions-compared.webp) Laboratories conflate three distinct checks, and the confusion wastes money and misses problems. Separating them clarifies which method to reach for. **Purity** asks whether the culture contains a single organism. Contamination can be a different species, a different genus, or a different strain of the same species. A mixed culture of *E. coli* and *Klebsiella pneumoniae* fails a purity check, and so does a 99-to-1 mixture of two *E. coli* strains that look identical on a plate. Purity checks operate mostly at the visible and cultural level. **Identity** asks whether the organism is the taxon on the label. A pure culture of *Enterococcus faecalis* labeled as *Streptococcus pneumoniae* is pure but wrong. Identity checks operate at species level and above, and in regulated settings at strain level. **Genetic stability** asks whether the organism has changed over time in culture. This matters most for live attenuated strains, seed lots, and any culture passed serially. A study of *Brucella abortus* S19 and RB51 vaccine strains examined 63 commercial batches produced between 2006 and 2009 from nine manufacturers, typed them by multiple-locus variable-number tandem repeat analysis (MLVA16, a method that sizes repeat regions at 16 genomic loci), and passed reference strains and selected batches through ten serial in vitro passages [4]. The strains proved genetically stable and homogeneous overall, but batches from one S19 manufacturer and one RB51 manufacturer carried a distinct genotype, each differing from the reference vaccine strain at the Bruce07 locus by a single repeat unit, one gain and one deletion respectively [4]. That finding shows two things at once. Genetic drift can be small enough to escape every phenotypic check, and it can also be caught when the right molecular method is applied. There is a useful parallel outside microbiology. Genetic monitoring of inbred laboratory animal colonies is described as a key element of quality control and colony management, and one study mined genome-wide microsatellite markers from ddRAD sequencing data of Swiss albino mice, identified 508 and 353 short sequence repeat motifs in outbred and inbred groups, and validated eight of those markers by PCR and gel electrophoresis [5]. The principle transfers directly. Living biological materials held for long periods drift or get swapped, and a small panel of genetic markers is an efficient way to notice. ## When to Check: A Bench Workflow Purity checks are not a one-time event at receipt. A practical schedule built into laboratory quality management looks like this. 1. **On receipt.** Examine the lyophilized vial or frozen stock, note any physical abnormality, then open it under conditions appropriate to the organism and plate for isolation on a nonselective medium such as sheep blood agar. 2. **After resuscitation.** Once the first subculture has grown, perform the primary purity check, colonial morphology review plus a stain, before any stock is frozen down. 3. **Before freezing a master stock.** Never freeze a stock that has not passed a purity check. A contaminated master seed propagates the problem into every future working stock. 4. **Before and after each passage.** Each subculture is an opportunity for contamination and a selective pressure for drift. The Brucella work passed strains through ten in vitro passages and used MLVA16 to confirm stability across them, framing the method as suitable for inclusion in official in vitro tests for live vaccine quality control [4]. 5. **Before releasing a working lot.** A working stock released to other analysts or other laboratories should have a documented purity result attached to the release record. 6. **On a defined calendar for long-term stocks.** Stocks held for years should be rechecked periodically, and any culture that behaves unusually on subculture, changes colonial appearance, or gives an unexpected biochemical or susceptibility pattern should trigger an immediate purity and identity recheck regardless of schedule. ## Purity Check Methods for Reference Cultures: Options, Pros and Cons The best purity check methods for reference cultures in microbiology form a ladder from visual inspection to [full genome sequencing](/blog/guides/full-genome-sequencing). Each rung answers a slightly different question, and each has a characteristic blind spot. The sections below explain each option on its own terms. ### Colonial Morphology and the Streak Plate This is the oldest and most used purity check in bacteriology. The culture is streaked for isolation onto a nonselective agar, commonly 5% sheep blood agar, and incubated under conditions appropriate to the organism, typically 35 to 37 degrees C for 18 to 24 hours, with added carbon dioxide or anaerobic incubation where the organism requires it. After incubation, the analyst examines the plate with transmitted and reflected light, looking for a single colony morphology, uniform hemolysis, uniform pigmentation, size, opacity, and edge pattern. Discrete colonies of uniform appearance are picked from well-separated areas of the plate and used for all subsequent work. **Pros.** The method is fast, cheap, and universally available. It detects gross contaminants of a different species almost immediately, since a second colony type on the plate is direct visual evidence of a mixed culture. It also produces a viable subculture, so the check and the working preparation happen in one step. **Cons.** Colonial morphology has a low ceiling of sensitivity. Two strains of the same species often produce indistinguishable colonies, so strain-level contaminants pass unnoticed. Slow-growing organisms can be overgrown by the dominant strain and never form visible colonies. Fastidious organisms and mycoplasmas may not produce visible colonies on standard media at all, which means a clean-looking plate is not evidence of purity for those organisms. Finally, the check depends entirely on the analyst's eye, so it needs defined photographic or written descriptions of the expected colonial morphology for the reference strain. ### Direct Microscopy and Stains Gram stain from a single isolated colony remains the workhorse confirmatory check. A thin smear is fixed, stained with crystal violet, iodine, decolorized with alcohol or acetone, and counterstained with safranin. The analyst records whether the organism is gram positive or gram negative, its cellular morphology (cocci, diplococci, rods, coccobacilli), and whether the field contains a single uniform cell type or a mixture. For organisms with distinctive features, additional stains add information, such as acid-fast staining for mycobacteria or a capsule stain where relevant. A wet mount can assess motility, and a few organisms show characteristic cell shapes or arrangements that support identity. **Pros.** Microscopy is inexpensive, takes minutes, and directly shows mixed cell types that could indicate contamination. It also catches one important failure mode of the plate check, a contaminant that grows invisibly within the colonies of the dominant strain but has a different cell morphology under the microscope. **Cons.** Some important reference organisms defeat the Gram stain. Mycoplasmas lack a cell wall, do not retain the Gram stain reliably, and require specialized culture media or molecular methods, which is why the *M. gallisepticum* seed strain in the live vaccine genome study needed genomic rather than routine stain-based characterization [1]. Microscopy also cannot distinguish strains of the same species, and low-level contamination (well below one cell per field) will not be seen. The check is subjective unless the laboratory defines what the expected microscopic picture looks like for each reference strain. ### Biochemical and Physiologic Profiling Classical identification runs a defined panel of substrates, sugars, and enzymatic tests, either as tube media or commercial panel systems, and compares the profile to a database. Supplementary physiologic tests, such as growth on specific media, hemolysis pattern, oxidase, catalase, and optochin or bacitracin susceptibility where applicable, round out the picture. For a reference culture, the value is not identification per se but pattern stability. A well-characterized reference strain has a known biochemical profile, and any deviation is a red flag for contamination, mutation, or mislabeling. **Pros.** Profiling is standardized, relatively inexpensive, and produces a written record suitable for a quality file. A changed profile can catch a swapped or contaminated culture even when colonies look uniform. **Cons.** Profiling takes a day or more, some systems require fresh overnight growth, and closely related strains of the same species usually give identical profiles, so the method has no strain-level resolving power. Some fastidious organisms are difficult to profile at all. Because profiles are descriptive rather than definitive, an unchanged profile does not prove that nothing changed elsewhere in the genome. ### MALDI-TOF Mass Spectrometry Matrix-assisted laser desorption ionization time-of-flight mass spectrometry profiles the ribosomal proteins of a colony and matches the spectrum against a reference database. A smear of a single colony is spotted on a target plate, overlaid with matrix solution, and a spectrum is acquired in minutes. The method has become standard in clinical laboratories for species-level identification of routine isolates. **Pros.** Turnaround is fast, per-sample consumable cost is low once the instrument exists, and the method easily detects a mixed spectrum when two species are present in the spot. It is an excellent rapid identity check on a colony that has already passed the plate and stain checks. **Cons.** The instrument is a major capital purchase, so it is unavailable to many small laboratories. Database coverage determines what can be identified, and entries for unusual or older reference strains can be missing or mislabeled. Species with very similar ribosomal protein spectra can be misidentified at the species or strain level. Like every colony-based method, MALDI-TOF only sees whatever was picked, so a low-level contaminant outside the picked colony escapes detection. ### PCR and 16S rRNA Gene Sequencing [Polymerase chain reaction](/knowledge/molecular-biology/polymerase-chain-reaction) (PCR) amplifies a defined DNA target, and broad-range 16S rRNA [gene sequencing](/knowledge/molecular-biology/gene-sequencing) reads the roughly 1,500-base bacterial small subunit [ribosomal RNA](/knowledge/bioinformatics/ribosomal-rna-structure-taxonomic-profiling) gene and compares the sequence to curated public databases such as those maintained by NCBI and EMBL-EBI. A colony or a loopful of growth is lysed, the gene is amplified with broad-range primers, and the sequence is matched to reference entries. Specific PCR assays, for example genus- or species-specific targets, serve as rapid screens for expected or unexpected organisms. **Pros.** Sequencing reaches species identification where biochemical panels and MALDI-TOF databases fail, particularly for slow-growing, unusual, or taxonomically revised organisms. It gives an objective, shareable record, a sequence trace, that can be compared across laboratories and years. Broad-range PCR can also reveal an unexpected second organism if a mixed template produces a messy chromatogram. **Cons.** A single-copy core gene has limits. Many species within a genus share nearly identical 16S sequences, so the method often cannot separate close relatives, let alone strains. PCR contamination with amplified product from previous runs is a real hazard in laboratories that process many samples, and mixed templates can be dominated by one organism so that a minor contaminant contributes no visible sequence. The method also requires molecular infrastructure and trained staff. ### MLVA and Other Variable-Number Tandem Repeat Typing Multiple-locus variable-number tandem repeat analysis sizes repeat regions at a defined set of loci by PCR and capillary electrophoresis, producing a numeric genotype. MLVA16 for *Brucella* sizes 16 loci, and because tandem repeats mutate at measurable rates, the method detects small genomic differences between and within strains. **Pros.** The Brucella study demonstrated exactly what makes MLVA attractive for reference culture work. Across 63 commercial batches and ten serial passages, MLVA16 distinguished reference strains from divergent commercial batches and detected single repeat unit changes at the Bruce07 locus, leading the authors to conclude that MLVA16 can discriminate small genomic variations and should be included in official in vitro tests [4]. The method is cheaper than sequencing, produces portable numeric genotypes, and directly measures the property that matters for serially passed stocks, genetic stability over passages. **Cons.** The panel is organism specific. A Brucella MLVA scheme does not transfer to *Pasteurella* or *E. coli*, so each reference organism needs its own validated loci and primers, which exist only for well-studied groups. Repeat loci sample a tiny fraction of the genome, so changes elsewhere go undetected. Interpretation of single-locus differences requires judgment, since one repeat unit can be a biologically meaningless slippage event or a marker of a genuinely different lineage, and the Brucella work itself documented both single-locus divergence in commercial batches and overall genetic homogeneity of the reference strains [4]. ### Whole Genome Sequencing with Hybrid Assembly Complete genome sequencing reads the entire genetic content of the strain. The highest quality assemblies in the recent vaccine strain literature used a hybrid approach, combining Illumina NovaSeq 6000 short reads with Oxford Nanopore MinION long reads to produce gap-free, complete, well-annotated genomes [1][2]. Short reads give accuracy, long reads resolve repeats and plasmids, and the hybrid assembly produces a closed genome that serves as a permanent reference for the strain. The output directly supports purity and identity work in several ways. Genome size becomes a fingerprint, and the published set covers a wide range, 1.78 Mb for *E. rhusiopathiae* NL-11, 2.40 Mb for *C. chauvoei* B, 5.34 Mb for each of the two *B. bronchiseptida* strains, 0.95 Mb for the *M. gallisepticum* strain, and 2.28 to 5.35 Mb across the inactivated vaccine strains [1][2]. Assembly quality metrics quantify purity at the sequence level. All eleven inactivated vaccine strain assemblies in the second study showed completeness above 99% and contamination below 1%, which the authors cite as the basis for reliable downstream characterization [2]. Annotation then lets analysts verify the presence of expected genes, toxin loci, and attenuating mutations, and lets regulators confirm at strain level that a manufacturer is using the declared seed [1][2]. **Pros.** [Whole genome sequencing](/blog/guides/whole-genome-sequencing) is the most complete answer available. It resolves strain-level identity, detects mixed DNA from more than one organism, documents the entire gene content, and creates a permanent digital reference against which every future subculture can be compared. For seed lots and biological products, it is the standard of proof, and both vaccine strain studies frame complete genomes as the robust foundation for identity verification and quality control of biological products [1][2]. **Cons.** The method is the most expensive and slowest option, requires sequencing infrastructure or an external provider, and demands bioinformatic competence to assemble, annotate, and interpret. Cost and turnaround make it impractical as a per-subculture routine check. Its contamination metric reflects the extracted DNA population, so an organism whose DNA extracts poorly could be undercounted, and the <1% contamination figure describes sequence data quality rather than a live-culture observation [2]. Sequencing also tells you nothing about viability or phenotypic expression, so it complements rather than replaces plate, stain, and growth checks. ## Side-by-Side Comparison The table below is the core of the choice. Cost tiers are qualitative because real prices vary widely by region, instrument access, and contract sequencing rates. | Method | What it confirms | Time to result | Cost tier | Main blind spot | Best role | |--|--|--|--|--|--| | Streak plate and colonial morphology | Single colony type present, viable culture | Same day to 24 h | Low | Same-species contaminants, nonvisible organisms on the medium used | Every subculture, first-line check | | Gram stain and microscopy | Single uniform cell type, expected morphology | Under 1 h | Low | Wall-less organisms, low-level contamination, no strain resolution | Every subculture, paired with the plate | | Biochemical and physiologic profiling | Species-level identity, pattern stability | 1 to 2 days | Low to moderate | No strain resolution, slow for fastidious organisms | Receipt of new strains, periodic reconfirmation | | MALDI-TOF MS | Species identity, mixed spectra | Under 1 h | Low per sample, high capital | Database coverage, closely related species, colony bias | Rapid identity confirmation where an instrument exists | | PCR and 16S rRNA sequencing | Species or targeted identity, DNA-level confirmation | 1 to 3 days | Moderate | Closely related species and strains indistinguishable, PCR contamination | Unusual or difficult organisms, permanent records | | MLVA or VNTR typing | Strain identity, genetic stability over passages | 1 to 2 days | Moderate | Organism-specific panels, samples a tiny part of the genome | Serially passed seed lots of covered organisms [4] | | Whole genome sequencing, hybrid assembly | Complete identity, genome content, sequence-level purity | Days to weeks | High | Not a viability or phenotype check, cost limits routine use | Master seed lots, regulatory confirmation, permanent references [1][2] | Read the table as a workflow, not a menu of equals. The first two rows run on essentially every subculture of every reference culture. The middle rows run at defined milestones, on receipt, on suspicion, and on schedule. The last rows run rarely, at the moments when strain-level proof is required, and the genetic stability row applies only to organisms with a validated typing scheme. ## How to Choose: Worked Scenarios Three scenarios show how the criteria, speed, sensitivity, cost, and the question actually being asked, decide the choice. **Scenario 1. A clinical laboratory receives a new quality control strain for antimicrobial susceptibility testing.** The stock arrives lyophilized from a reputable supplier, but laboratory quality management still requires local verification. The analyst reconstitutes the vial, streaks for isolation on blood and chocolate agar, incubates appropriately, and reviews colonial morphology against the supplier's description. A Gram stain from a single colony confirms a single uniform cell type. Because the strain will anchor susceptibility results for years, the analyst adds identification by the laboratory's routine method, MALDI-TOF if available or a biochemical panel, and records the profile and image in the strain file. Genome sequencing is not indicated at this step. The question is whether the culture is viable, alone, and the right species, and the ladder stops there. **Scenario 2. A biological manufacturer manages a master seed lot and working seed lots across years of production.** Here the question is strain-level fidelity over time, and the bar is regulatory. The manufacturer maintains a master seed with minimal passages, derives working seeds under documented passage limits, and applies a molecular stability check across those passages. The Brucella model is the template. Serial in vitro passages were tested by MLVA16 and found genetically stable, which is precisely the evidence a producer needs that its passage scheme does not alter the strain, and the same study documented that commercial batches from some manufacturers carried a distinct Bruce07 genotype, which is exactly the kind of finding an official in vitro test should surface [4]. For the master seed itself, complete genome sequencing with hybrid assembly establishes the reference sequence once, and the published vaccine strain projects show how genome data then anchor identity confirmation and consistent production over time [1][2]. **Scenario 3. A laboratory's in-house probiotic working strain, held for repeated teaching and research use, suddenly shows two colony types on subculture.** The immediate answer comes from the plate itself, a mixed culture is visible and the subculture fails purity. The analyst picks one colony of each type, Gram stains both, and identifies each by MALDI-TOF or 16S sequencing. If the second organism is a recognizable laboratory contaminant, the stock is discarded, the working and master stocks are traced, and the incident is logged in the quality system. If the two colony types are the same species and possibly the same genus and species pair that only sequencing can separate, the culture is escalated to molecular resolution. The probiotics position statement's insistence on strain-specific identification and manufacturing quality control is the governing principle here, since a strain's claimed properties cannot be attributed to a stock whose composition is uncertain [3]. The pattern across all three scenarios is consistent. Start at the cheap and fast end, escalate only when the question demands it, and match the molecular method to the organism and the question, species-level identity for routine confirmation, strain-level markers for stability, whole genomes for seed lots and permanent records. ## Building Purity Checks into Laboratory Quality Management A purity check that lives in an individual analyst's head is not a quality control. The checks above only protect the reference culture system when they are written into the laboratory's quality management structure. **Write the expected phenotype down.** For each reference strain, the strain file should state the expected colonial morphology on the routine media, the expected microscopic appearance, and, where applicable, the expected biochemical or mass spectrometry profile. Without a defined expectation, a changed appearance has nothing to be compared against. **Define acceptance criteria and actions.** The procedure should state what passes and what happens on failure. A single colony type matching the description passes. Mixed morphology fails and triggers discard, trace-back through recent subcultures, and recheck of the master stock. An unusual biochemical or spectrometric result triggers identity rechecking before any result generated with that strain is released. **Record everything with dates and passage numbers.** Every purity check should be tied to a lot number, a passage count, an analyst, and a date. The Brucella study could only demonstrate stability across ten passages because passage history was explicit and the genotypes were recorded batch by batch [4]. A laboratory that cannot say how many passages a working stock has had cannot make a stability claim of any kind. **Trend the results.** Microbiology quality control is strongest when results are reviewed in aggregate. Recurring colony variants, a strain that repeatedly throws small-colony variants, or repeated contamination events tied to one incubator or one analyst's technique are trends that only appear on review. The value of routine, structured quality parameters is visible in other regulated supply chains as well. A systematic review of antimalarial drug quality in East Africa evaluated 300 samples against defined quality control parameters including assay and percentage purity, and found failure rates of 15.93% for dissolution, 5.00% for assay, and 1.90% for mass uniformity, failures that only structured routine testing could reveal [6]. Biological reference stocks deserve the same discipline, and the manufacturing quality control framework articulated for probiotics, with defined selection criteria and strain-level verification, is a direct model [3]. **Schedule reverification.** Master stocks get the deep characterization once, at establishment, including genome sequencing where the stakes justify it. Working stocks get the full ladder on receipt and periodically thereafter. The genome sequence of the master becomes the comparator for any future whole-genome check, which is how the vaccine strain projects intend their data to be used, as a lasting reference for confirming that the same seed strain is in use over time [1][2]. ## Common Mistakes and Limitations The following mistakes account for most purity check failures in practice. **Checking the wrong colony.** Picking a single well-isolated colony and calling the plate pure is the most common error. If any colony on the plate differs, the culture is mixed, regardless of how clean the picked colony looks. The check applies to the whole plate. **Trusting a clean plate for noncultivable or fastidious contaminants.** Mycoplasmas and other wall-less or highly fastidious organisms may never form visible colonies on the media used for the primary organism. A plate check alone cannot clear such a culture, and molecular or genomic methods are needed, which is why the mycoplasmal vaccine strain in the genome study required full sequencing rather than plate review [1]. **Confusing purity with identity.** A culture can be perfectly pure and perfectly wrong. Pure cultures still need identity confirmation at the milestones described above, and identity checks do not substitute for the per-subculture purity checks. **Assuming selective media prove purity.** Growth on a selective medium only shows that the target organism survives that medium. Contaminants suppressed by the selective agents remain in the culture and reappear when conditions change. Purity checks belong on nonselective media. **Ignoring passage number.** Every serial passage is both a contamination risk and a drift risk. The Brucella work demonstrates that a validated molecular typing method can verify stability across defined passages and detect divergence between batches and their reference strains, and it explicitly recommends such testing for official in vitro quality control [4]. A laboratory that passes a strain indefinitely without passage limits or periodic stability testing has no basis for claiming the strain is unchanged. **Over-reading a single genomic metric.** Completeness above 99% and contamination below 1% in a genome assembly describe the quality of the sequence data and the assembly, not the living culture in every downstream vial [2]. Genomic results must be tied back to the specific culture lot they were generated from. **Skipping the record.** An unwritten purity check does not exist in a quality system. Images, stains, profiles, and sequences belong in the strain file with dates. **One caveat on individual application.** The methods and scenarios here describe laboratory practice for reference cultures, and individual cases, particularly clinical isolates or production problems with animal or human health implications, need review by the responsible veterinarian, microbiologist, or quality officer before any action is taken. ## Frequently Asked Questions ### How often should a purity check be performed on a reference culture? At every subculture, plus at defined milestones. Each resuscitation, each passage, each new frozen stock, and each working lot release should have a documented purity check, with molecular or genomic checks added on a schedule appropriate to how the stock is used. ### Can one method guarantee that a culture is pure? No. Every method has a blind spot, which is why laboratories layer checks. Plate and stain checks catch visible contaminants, biochemical and spectrometric methods confirm species identity, and molecular or genomic methods reach strain level and detect changes invisible to the eye. ### What is the difference between a purity check and an identity check? A purity check asks whether the culture contains a single organism, while an identity check asks whether that organism is the one on the label. A culture can pass one check and fail the other, so both are needed. ### Do I need whole genome sequencing for every stock in the freezer? No. Whole genome sequencing belongs on master seed lots, strains used in regulated production, and any culture where strain-level proof is required. Routine working stocks are managed with plating, staining, and periodic identity confirmation, with the master genome serving as the reference for any future comparison. ### How do purity checks differ for organisms that do not stain or grow on standard media? Plate and stain checks lose much of their power for wall-less organisms such as mycoplasmas and for highly fastidious species, because these organisms may not stain reliably or form visible colonies on standard media. For these, media appropriate to the organism, molecular assays, or genomic methods carry the purity and identity confirmation. ### What should I do and document when a purity check fails? ![Decision path from a failed purity check through quarantine, master stock recheck, contaminant identification and quality record.](/article-images/illustrations/purity-check-methods-reference-cultures-microbiology-purity-failure-triage.webp) Do not use the culture. Discard or quarantine the affected stock, recheck the master stock from which it was derived, trace back through recent subcultures to find where contamination entered, identify the contaminant if feasible, and record the event, the investigation, and the corrective action in the quality system before reestablishing a clean working stock. ## Sources 1. [High-quality genome assembly and annotation of live animal vaccine bacteria strains in South Korea.](https://pubmed.ncbi.nlm.nih.gov/40898044/) 2. [High-quality genome assembly and annotation of inactivated animal vaccine bacteria strains in South Korea.](https://pubmed.ncbi.nlm.nih.gov/42576204/) 3. [Use of probiotics in clinical practice with special reference to diarrheal diseases: A position statement of the Malaysian Society of Gastroenterology and Hepatology.](https://pubmed.ncbi.nlm.nih.gov/33490608/) 4. [Genetic stability of Brucella abortus S19 and RB51 vaccine strains by multiple locus variable number tandem repeat analysis (MLVA16).](https://pubmed.ncbi.nlm.nih.gov/23933375/) 5. [Comparison of ddRAD derived genome-wide SSR markers in outbred and inbred Swiss albino mice.](https://pubmed.ncbi.nlm.nih.gov/40350065/) 6. [Quality of Antimalarial Drugs in East Africa: A Systematic Review.](https://pubmed.ncbi.nlm.nih.gov/36277242/)