Reference

Glossary

238 terms across 10 categories — chromatography, mass spectrometry, content and mass, impurities and degradation, microbiology, peptide chemistry, statistics, quality systems, and this index's own conventions. Every technical term used on a record page resolves here.

238 terms10 categories

43 terms

Chromatography

Area percent: Purity expressed as the main peak's integrated area divided by the total integrated area of all peaks, times 100. A relative quantity. It is not a concentration and not a content.
Asymmetry factor (As): Peak shape measure taken at 10 % of peak height: the width of the rear half divided by the width of the front half. Symmetric = 1.00. Above 1.50 the index marks integration unreliable.
Backpressure: The pressure the pump develops to push mobile phase through the column. Rising backpressure across a sequence indicates a blocking frit or a fouled column inlet.
Baseline: The detector signal in the absence of an eluting analyte. On a gradient it drifts, and where the baseline is drawn determines the integrated area.
Capacity factor (k'): Retention relative to the unretained time: (t − t₀)/t₀. A k' below 1 means the compound is barely retained and is not usefully separated.
Column dead time (t₀): The time taken for an unretained species to traverse the column. Everything eluting at t₀ is unseparated.
Core–shell particle: A stationary-phase particle with a solid core and a porous shell. Shorter diffusion path, sharper peaks, lower backpressure than a fully porous particle of the same size.
Drift: Slow monotonic change in the baseline across a run. Caused by mobile-phase absorbance changing with organic fraction, or by column temperature not being controlled.
Drop-line integration: Separating two overlapping peaks with a vertical line dropped from the valley between them. Unbiased when the peaks are of comparable size.
Gradient: Programmed change in mobile-phase composition during a run. Gradient duration determines resolving power and is therefore part of the result, not metadata.
Gradient slope: Percentage points of organic solvent per minute. Lower slope, better resolution, longer run.
Guard column: A short disposable column before the analytical column, protecting it from particulates and strongly retained matrix. Shifts retention slightly and must be accounted for.
HILIC: Hydrophilic-interaction liquid chromatography. Polar stationary phase, high-organic mobile phase. Retains polar analytes that reverse phase cannot hold.
Ion-pair chromatography: Adding an amphiphilic counter-ion to the mobile phase so a charged analyte forms a neutral pair with enough hydrophobic character to be retained on C18.
Isocratic: Constant mobile-phase composition throughout the run. Simpler, and unsuitable for samples spanning a wide hydrophobicity range.
Mobile phase: The liquid carrying the sample through the column. In peptide work typically water/acetonitrile with 0.1 % trifluoroacetic acid.
Peak apex: The point of maximum detector response within a peak. Where the UV spectrum and mass spectrum are usually taken.
Peak integration: Determining a peak's area by defining its start, end and baseline. The step where a purity figure is actually decided, and the one least often documented.
Peak purity: Evidence from a photodiode-array or mass detector that the spectrum is constant across a peak, indicating no co-eluting species with a different chromophore. Silent on diastereomers.
Photodiode array (PDA): A UV detector acquiring a full spectrum at every time point rather than one or two wavelengths. Enables peak-purity assessment and the 280/214 ratio.
Plate count (N): Column efficiency: N = 5.54 (t/w½)². Higher is sharper. The index requires N ≥ 20,000 for a main peak.
Plate height: Column length divided by plate count. A physical measure of the efficiency per unit length, independent of column length.
Re-equilibration: Returning the column to starting conditions after a gradient. At least ten column volumes; short re-equilibration causes retention drift across a sequence.
Reporting threshold: The area percent below which peaks are not reported. The index uses 0.05 %. Excluded peaks are excluded from the total, so purity is systematically overstated by their sum.
Identification threshold: The level above which an impurity must be structurally identified rather than merely quantified. ICH Q3A sets it by daily dose.
Qualification threshold: The level above which an impurity must be shown to be biologically qualified. Not applicable to research material and included here for completeness.
Resolution (R): Separation between two peaks: R = 2(t₂ − t₁)/(w₁ + w₂). R = 1.5 is baseline resolution. The index requires R ≥ 1.20 to the nearest named impurity.
Retention time: Time from injection to peak apex. A property of the compound, the column and the gradient — comparable across laboratories only after gradient normalisation.
Reverse-phase chromatography: Separation on a non-polar stationary phase, typically C18-bonded silica, with a polar mobile phase. Retains by hydrophobic interaction.
C18: Octadecyl-bonded silica: the standard reverse-phase stationary phase for peptides. Retention scales with the analyte's exposed hydrophobic surface, not with its mass.
Shoulder: An unresolved peak appearing as an asymmetric bulge on the leading or trailing edge of a larger peak. The main source of integration ambiguity.
Solvent front: The disturbance at the start of a chromatogram caused by the injection solvent. Compounds eluting within it cannot be integrated.
System suitability: Injections demonstrating the chromatographic system was performing acceptably when the sample was run. Required by every pharmacopoeia, documented on almost no research-peptide certificate.
Tailing factor (T): Peak shape measure at 5 % of peak height: (a + b)/2a. USP convention; related to but not identical with the asymmetry factor.
Tangent skim: Integrating a small peak on the trailing edge of a much larger one by drawing a tangential baseline. The convention above a height ratio of about 10:1.
Trifluoroacetic acid (TFA): Mobile-phase additive at 0.1 %. Gives excellent peptide peak shape and suppresses electrospray ionisation. Also the most common peptide counter-ion.
Valley: The minimum of the detector signal between two adjacent peaks. Valley height relative to the smaller peak determines whether the two can be reliably integrated separately.
Valley-to-valley integration: Baselining a small peak from valley to valley rather than to the true baseline. Assigns the overlap to the larger peak and understates the smaller one.
Void volume: The volume of mobile phase within the column. Determines the dead time. A void marker injection calibrates the retention axis and is cheap and rarely done.
Wavelength, detection: The UV wavelength at which absorbance is measured. 214 nm sees the amide backbone; 280 nm sees only aromatic residues. Purity figures at the two are not comparable.
Chromophore: The part of a molecule responsible for UV absorbance. In peptides: the amide bond below 230 nm, and Trp/Tyr near 280 nm.
Ghost peak: A peak arising from material retained on the column from a previous injection and released by a later gradient. Detected by a blank injection.
Carry-over: Sample residue transferred from one injection to the next, usually in the autosampler needle or loop. Quantified by a blank after a high-concentration injection.

25 terms

Content & mass

Amino-acid analysis (AAA): Acid hydrolysis followed by chromatographic quantification of the liberated residues. The reference method for peptide content where no standard of the intact peptide exists.
Anhydrous basis: A content figure recalculated as if the sample contained no water. Must be stated, or a mass balance will not close.
Assay: Quantification of an analyte against a material of known content. Distinct from area percent, which compares peaks within one chromatogram.
Counter-ion: The anion paired with a peptide's basic residues in the isolated salt. Usually trifluoroacetate or acetate. Mass in the vial that is not peptide.
Fill weight: The gross mass of material in a vial as weighed. Not the peptide mass — it includes water, counter-ion and residual solvent.
Gravimetric fill check: Weighing a vial before and after removal of contents to determine the actual fill. The only direct measurement of label-claim compliance.
Karl Fischer titration: Stoichiometric determination of water by reaction with iodine and sulfur dioxide. The reference method for residual moisture in lyophilised material.
Coulometric titration: A titration in which the titrant is generated electrochemically and quantified from the charge passed. For water: 10.71 coulombs per milligram.
Label claim: The mass of compound stated on the vial. Whether it refers to gross fill or to peptide content is almost never specified.
Mass balance: Water + counter-ion + residual + total peptide content = 100 % of cake mass. Exhaustive and mutually exclusive. The index closes it on every record where the terms exist.
Peptide content: The fraction of total mass that is peptide, %w/w. Distinct from purity, which is a fraction of detected chromatographic area.
Target-compound content: Total peptide content multiplied by chromatographic purity. The fraction of the vial that is the named compound. The number that determines recovered mass.
Recovered mass: Gross fill × target-compound content / 100. The milligrams of named compound actually obtainable from the vial.
Relative response factor: The detector response of an impurity per unit mass, relative to the parent. Assumed to be 1.00 in almost all purity determinations, and usually is not.
Residual moisture: Water remaining in a lyophilised cake. 1–4 % in well-dried material; 8–13 % for hygroscopic short peptides.
Residual solvent: Process solvent remaining in the product. DMF, DMSO, acetonitrile and dichloromethane are the usual candidates in peptide synthesis. Controlled by ICH Q3C.
GC headspace: Gas chromatography of the vapour above a heated sealed sample. The standard technique for residual solvents.
Salt-free basis: A content figure recalculated as if the sample contained no counter-ion. Like the anhydrous basis, must be stated explicitly.
Salt form: Which counter-ion the compound was isolated as. Trifluoroacetate is 114 Da per site; acetate is 60. At equal purity, acetate vials contain more peptide.
Sulfated ash: The inorganic residue remaining after ignition with sulfuric acid. A measure of inorganic contamination and part of the mass balance.
Quantitative NMR (qNMR): Absolute quantification by NMR against an internal standard of certified purity. Method-independent and rarely used in this market.
Nitrogen determination: Content estimation from total nitrogen, by Kjeldahl or combustion. An orthogonal check on AAA. Confounded by any nitrogen-containing excipient.
Ion chromatography: Separation of ions on an ion-exchange column with suppressed conductivity detection. The reference technique for counter-ion quantification.
ICP-MS: Inductively coupled plasma mass spectrometry. Elemental analysis. Required for metal-complexed peptides and for elemental impurity profiles.
Elemental impurity: Trace metal contamination, controlled by ICH Q3D. Palladium, copper, nickel and chromium are the relevant risks in peptide synthesis. Not reported anywhere in this index.

31 terms

Impurities & degradation

Acetylation: Addition of an acetyl group, +42.011 Da. Either a deliberate synthesis capping step or an unwanted side reaction.
Aggregate: A non-covalent assembly of two or more molecules. Dissociated by the acidic organic mobile phase of reverse-phase HPLC, so invisible to it. Seen by size-exclusion chromatography.
Aspartimide: A cyclic imide formed at an Asp-X motif, −18.011 Da. Hydrolyses to a mixture of α- and β-aspartyl peptides, the latter mass-identical to the parent.
Asp-Gly motif: An aspartate followed by a glycine. The most aspartimide-prone sequence in peptide chemistry. Epitalon's defining analytical problem.
β-Aspartyl isomer: A peptide in which the backbone connects through the aspartate side-chain carboxyl. Mass-identical to the parent and barely resolved. Requires a dedicated method.
Capping: Deliberately acetylating unreacted amines during synthesis so failed couplings terminate as short capped species rather than continuing as deletion sequences.
Deamidation: Conversion of asparagine to aspartate or glutamine to glutamate, +0.984 Da. A storage degradation route, accelerated above pH 7.
Isoaspartate: The rearranged product of deamidation via a succinimide intermediate. Mass-identical to aspartate-containing peptide.
Deletion sequence: A peptide missing one or more residues from an incomplete coupling. Mass difference equals the missing residue mass. A synthesis finding: it does not form in a vial.
Diastereomer: A stereoisomer arising from inversion at one chiral centre. Same formula, same mass, same UV spectrum, often much less active. The impurity no mass spectrum finds.
Diketopiperazine: A cyclic dipeptide, −18.011 Da, formed by intramolecular attack at the N-terminus. Common in proline-rich short peptides.
Dimer: Two molecules covalently linked, usually through a disulfide. Mass is roughly double the parent, minus 2.016 Da for a disulfide linkage.
Disulfide scrambling: Cysteines paired in an incorrect combination. Mass-identical to correct pairing and only detectable by peptide mapping or comparison against a folded reference.
Open-chain form: A cyclic peptide with its disulfide reduced, +2.016 Da. A handling artefact, resolved chromatographically, and functionally inactive.
Epimerisation: Inversion of a single stereocentre during synthesis, typically during carboxyl activation. Produces a diastereomer.
Racemisation: Loss of stereochemical purity toward a 50:50 mixture. In stepwise peptide synthesis the practical concern is epimerisation at individual residues.
Fibrillation: Ordered self-assembly into amyloid-like fibrils. Glucagon is the fastest fibrillator in this index; native amylin is why pramlintide exists.
Fmoc: Fluorenylmethyloxycarbonyl, the standard temporary amine protecting group in solid-phase synthesis. Residual Fmoc appears at +222.068 Da.
Oxidation: Addition of oxygen. Methionine to sulfoxide (+15.995) then sulfone (+31.990); tryptophan to oxindole and beyond; cysteine to sulfonic acid.
Oxidation ladder: The series of peaks from sequential oxidation of multiple oxidisable residues: parent, mono-sulfoxide, bis-sulfoxide. Diagnostic — a single early peak with no second rung is probably something else.
Sulfoxide: The +15.995 Da methionine oxidation product. Elutes earlier than the parent on reverse phase. A handling finding, not a synthesis finding.
Pyroglutamate: Cyclisation of an N-terminal glutamine or glutamate, −17.027 or −18.011 Da. Also the native N-terminus of GnRH, where acid stress opens it to +18.011.
Regioisomer: An isomer differing in the position of a substituent, for example an acyl group on a different side chain. Mass-identical.
Related substance: A process- or degradation-derived impurity structurally related to the target. The pharmacopoeial term for what a purity method quantifies.
Impurity fingerprint: The vector of named impurities and their levels in a lot. Characteristic of one manufacturing run; used by the index to identify shared bulks.
Shared bulk: One physical batch of material sold by two or more suppliers under different lot numbers. Normal in a contract-synthesis market; means two results are one measurement.
TFA adduct: A trifluoroacetylated peptide, +95.982 Da, formed during cleavage or from prolonged exposure to trifluoroacetic acid.
Truncation: A peptide missing residues from one terminus. Distinguished from an internal deletion by the mass difference matching a contiguous run of residues.
tBu residual: Incompletely removed tert-butyl side-chain protection, +56.063 Da. Indicates a short or weak cleavage step.
Forced degradation: Deliberately stressing a sample under acid, base, oxidative, thermal and photolytic conditions to prove the method can resolve degradants.
Stability-indicating method: A method demonstrated to separate the analyte from all its degradation products. A stability claim from a non-stability-indicating method means nothing.

30 terms

Index conventions

n=: The number of records a statistic was computed from. Displayed as a badge beside every statistic on this site; amber below 8, suppressed below 3.
Provenance: How a sample was obtained: buyer-sourced, supplier-supplied or laboratory-retained. A first-class field in this index because it changes the weight of the evidence.
Buyer-sourced: Purchased through the supplier's ordinary retail channel by the submitter, with no supplier involvement. The strongest provenance class here.
Supplier-supplied: Provided by the supplier for testing. The supplier chose the vial. Reads a mean 0.31 pp higher than buyer-sourced across this index.
Assay report: The primary object in this index: one analysis of one sample of one batch by one laboratory. Everything else on the site is a rollup of these.
Batch rollup: All reports in the index for one physical lot, with a reproducibility assessment across independent tests.
Within-batch spread: The range of independent purity results on one lot. A statement about bulk homogeneity and fill control, distinct from median purity.
COA divergence: The supplier's declared purity minus the median of independent measurements on the same lot. Flagged above 1.5 pp by rule AX-R05.
Documentation-completeness score: The index's 0–100 rubric for what a supplier's certificates actually state. Rubric published in full at /suppliers/methodology/.
Gradient normalisation: Converting a retention time measured on one gradient duration to the index reference scale, so retentions from different laboratories can be compared.
Index reference method RG-2: The index's nominated reverse-phase method: C18 2.6 µm, 0.1 % TFA, 5→65 % acetonitrile over 30 min at 40 °C, 214 nm. All retention windows are stated against it.
Method limited: A verification status: the method used does not meet the index specification for the compound class. The result is published with the limitation named.
Partial: A verification status: a measurement the compound requires is missing — a metal determination on a complex, an anomeric ratio on NAD+.
Purity only: A verification status: no identity confirmation was ordered. The record describes homogeneity and does not establish what the material is.
Verification status: The index's summary of how complete a record is: verified, identity consistent, purity only, method limited, partial, or not applicable.
Rule AX-R01: Mass accuracy outside the class limit. Records identity as not established rather than failed, because most instances are calibration problems.
Rule AX-R02: A certificate's declared purity and declared total impurities do not sum to 100 %. A documentation finding: the certificate is not interpretable as written.
Rule AX-R03: A certificate reports a largest single impurity exceeding its total impurities. Arithmetically impossible; a template error.
Rule AX-R04: Repeated round-number purity declarations from one supplier. Requires a pattern of at least three, because one round result is unremarkable.
Rule AX-R05: Declared purity more than 1.5 pp above the median of at least two independent tests on the same lot. The index's most consequential rule.
Rule AX-R06: Recovered mass outside ±10 % of label claim. The rule that fires most often and matters most to anyone buying by the milligram.
Rule AX-R07: Declared water, counter-ion, peptide content and residual do not account for 100 % of mass. Usually an unstated reporting basis.
Rule AX-R08: Endotoxin above the index screening threshold of 5.0 EU/mg. A screening figure, not a pharmacopoeial limit.
Rule AX-R09: Impurity fingerprints matching across suppliers, indicating one bulk under two labels. Not improper; means two results are one measurement.
Rule AX-R10: A supplier-supplied sample reading more than 0.8 pp above buyer-sourced material from the same lot.
Rule AX-R11: Retention time outside the compound window after gradient normalisation. Two thirds of instances are undocumented method differences.
Rule AX-R12: Peak asymmetry above 1.50 or resolution below 1.20 to the nearest impurity. Makes the reported purity a range rather than a number.
Submission: A report offered to the index by a buyer or laboratory. Becomes a record after normalisation and provenance classification.
Annotation: A short signed technical comment attached to a record. The index's only community surface — there is no forum and no voting.
Suppressed statistic: A statistic the index declines to publish because n < 3. The individual values are shown instead.

21 terms

Mass spectrometry

Adduct: An ion formed by association with a cation other than a proton — sodium at +21.982, potassium at +37.956. Not an impurity; the same molecule with a different cation.
Average mass: The abundance-weighted mean mass over the whole natural isotope distribution. What a low-resolution instrument reports. About 600 ppm above the monoisotopic mass for a peptide.
Base peak: The most intense peak in a spectrum. For a 4 kDa peptide by electrospray it is usually the 4+ charge state.
Charge state (z): The number of protons carried by an ion. Observed m/z = (M + z × 1.007276)/z. Two adjacent charge states determine z without assumption.
Charge-state envelope: The distribution of intensities across charge states. Its shape carries conformational information: compact molecules ionise at lower charge.
Deconvolution: Recovering a neutral mass from a set of observed m/z values across charge states. Self-checking: two charge states give two independent estimates.
Electrospray ionisation (ESI): Producing gas-phase ions from solution by applying a high potential to a fine spray. Multiply charges large molecules and is compatible with LC.
High-resolution mass spectrometry: Sufficient resolving power to distinguish species differing by a small fraction of a dalton. Required to exclude 1 and 2 Da alternatives.
Ion suppression: Reduction in analyte ionisation caused by a co-eluting species or a mobile-phase additive. Trifluoroacetic acid is a strong suppressor.
Isotope pattern: The set of peaks arising from natural isotopic variants of a molecule. Its spacing gives the charge state directly: 1/z daltons.
Lock mass: A reference compound infused continuously so calibration drift can be corrected in real time. Its absence limits mass accuracy to the last external calibration.
Mass accuracy: Difference between measured and theoretical mass, in ppm of the theoretical value. A relative measure — the same ppm is a different absolute error at different masses.
Mass defect: The difference between a nominal integer mass and the exact mass. Constrains which elemental compositions are plausible for an observed impurity.
Monoisotopic mass: The mass of the isotopologue in which every atom is its most abundant isotope. All theoretical masses on this site are monoisotopic.
MS/MS: Tandem mass spectrometry: selecting an ion, fragmenting it, and measuring the fragments. Required to distinguish isobaric species such as pyroglutamate from aspartimide.
ppm: Parts per million. Used for mass accuracy: ppm = (observed − theoretical)/theoretical × 10⁶.
Resolving power: A mass spectrometer's ability to separate two close masses, expressed as m/Δm. A quadrupole gives ~1,000; a TOF 20,000–50,000.
Space charge: Mutual repulsion between ions at high density, shifting measured m/z. Why an overloaded injection can read a different mass from an on-scale one.
Time-of-flight (TOF): A mass analyser measuring the time for ions to traverse a flight tube. High resolving power; calibration is temperature-sensitive.
Unit resolution: Resolving power sufficient to separate masses one dalton apart but not to determine exact mass. Cannot exclude a +1 or +2 Da alternative species.
Quadrupole: A mass analyser using oscillating electric fields. Robust, inexpensive, unit resolution.

13 terms

Microbiology

Bacterial endotoxin: Lipopolysaccharide from gram-negative bacterial outer membranes. Pyrogenic, filter-passing, autoclave-stable, and invisible to every chromatographic method.
Endotoxin unit (EU): The activity unit for bacterial endotoxin, defined against a reference standard. Roughly 0.1–0.2 ng of E. coli LPS per EU.
LAL: Limulus amebocyte lysate. The horseshoe-crab-derived reagent whose clotting cascade detects endotoxin. Available as gel-clot, turbidimetric and chromogenic assays.
Gel-clot: A limit-test LAL format: a firm gel forms or it does not. Gives a pass or fail against a stated sensitivity, never a number.
Kinetic chromogenic: A quantitative LAL format reading the onset time of a colour change against a standard curve. Sensitive to 0.005 EU/mL.
Recombinant factor C (rFC): A recombinant alternative to LAL using the cascade's first enzyme. Equivalent sensitivity, no horseshoe crabs.
Inhibition/enhancement control: A spike of standard endotoxin added to the sample to show the matrix does not interfere. Must recover at 50–200 % or the result is invalid.
Maximum valid dilution: The greatest dilution at which an endotoxin limit can still be detected. Sets how far a sample may be diluted to overcome interference.
Threshold pyrogenic dose (K): 5.0 EU per kilogram per hour for parenteral routes, 0.2 EU/kg for intrathecal. Converting it to a per-milligram limit requires the dose.
Sterility test: Incubation of the article or its filtrate in growth media for at least 14 days. Destructive, low-powered, and not a statement about the lot.
Membrane filtration: The preferred sterility technique: filter the article through 0.45 µm, then incubate the membrane in media.
Method suitability (sterility): Demonstration that the article does not inhibit growth of challenge organisms. Omitting it on an antimicrobial peptide produces a false pass.
Bioburden: The viable microbial count in a non-sterile article. Distinct from sterility, which is a presence/absence test.

22 terms

Peptide chemistry

Aib: α-Aminoisobutyric acid. An achiral, sterically hindered residue used to resist enzymatic cleavage. Present in every incretin co-agonist in this index; its hindered couplings drive epimerisation nearby.
Acylation: Attachment of a fatty-acid chain to a side chain, conferring albumin binding and a long half-life. The defining modification of the long-acting GLP-1 analogues.
AEEA linker: 2-(2-(2-aminoethoxy)ethoxy)acetic acid. The spacer between the peptide backbone and the fatty-acid chain in semaglutide. Its loss gives the diagnostic −145 Da impurity.
Coupling efficiency: The fraction of resin-bound chains that react at each synthesis step. At 99.5 % per step, a 40-mer is 82 % full-length before purification.
C-terminal amide: A peptide ending in -CONH₂ rather than -COOH. Frequently required for activity; its hydrolysis to the free acid is a +0.984 Da impurity.
Lactam bridge: A cyclisation through an amide bond between a side-chain carboxyl and a side-chain amine. Melanotan II and bremelanotide are lactam-cyclised.
Macrolactamisation: The cyclisation step forming a lactam bridge. Incomplete cyclisation leaves the linear precursor at +18.011 Da — the commonest process failure in the melanocortin class.
PEGylation: Covalent attachment of polyethylene glycol. The polymer is polydisperse, so a PEGylated peptide has no single mass and no meaningful single purity figure.
Polydispersity: A distribution of molecular masses rather than a single value. Characteristic of synthetic polymers and therefore of PEG conjugates.
Maleimide: A thiol-reactive group used for covalent conjugation, as in CJC-1295 DAC. Hydrolyses to an inert maleamic acid at +18.011 Da — pure and non-functional are compatible states.
Pseudoproline: A dipeptide building block that temporarily disrupts secondary structure on-resin, used to get through aggregation-prone sequences.
Resin: The insoluble polymer support on which solid-phase synthesis is performed. Loading and swelling behaviour determine achievable purity for difficult sequences.
Solid-phase peptide synthesis (SPPS): Stepwise chain assembly on an insoluble support, with excess reagent driving each coupling. The origin of almost every peptide in this index.
Cleavage cocktail: The acidic mixture that releases the peptide from the resin and removes side-chain protection. Scavengers in it suppress alkylation artefacts.
Preparative HPLC: Chromatography at scale to purify the crude peptide. The fraction cut sets the purity/yield trade, and a wide cut is visible in the impurity profile.
Salt exchange: Converting a peptide from one counter-ion to another, typically trifluoroacetate to acetate, by repeated lyophilisation from the new acid or by ion exchange. Rarely complete.
Lyophilisation: Freeze-drying: freezing the solution then subliming the ice under vacuum. Cake structure and residual moisture are the observable outputs.
Collapse temperature: The temperature above which a freeze-concentrated matrix loses structure during primary drying. Exceeding it gives a collapsed cake and high residual moisture.
Melt-back: A lyophilisation failure in which the cake structure is entirely lost, leaving a powder or glassy residue. Expect high residual moisture.
Metal complex: A peptide coordinating a metal ion, as in GHK-Cu and thymulin. An acidic reverse-phase mobile phase dissociates the complex on-column.
Recombinant: Produced by expression in a host organism rather than by chemical synthesis. Brings folding, glycosylation and host-cell-protein questions that a purity figure does not address.
Host cell protein: Protein from the expression organism carried through purification. A required specification for recombinant products and reported on none in this index.

14 terms

Physical

Cake: The porous solid remaining after lyophilisation. Its appearance is a direct readout of cycle control.
Cake shrinkage: Separation of the cake from the vial wall. Indicates moisture uptake after drying, or a closure-integrity failure.
Closure integrity: Whether the stopper and crimp maintain the container's seal and vacuum. A failure lets in moisture and oxygen.
Headspace oxygen: Oxygen in the gas space above the product. Correlates with sulfoxide content and is essentially never measured.
Bacteriostatic water: Water containing 0.9 % benzyl alcohol as a preservative. The benzyl alcohol absorbs strongly in the UV and adds peaks to a chromatogram.
Reconstitution: Dissolving a lyophilised cake in diluent. Procedure matters: directing a stream at the cake or shaking can nucleate aggregates that were not in the vial.
Reconstitution clarity: The visual appearance of the reconstituted solution. Recorded verbatim in this index because the descriptive detail is the diagnostic content.
Cold chain: Maintaining a controlled low temperature through shipping and storage. Its failure has a specific analytical signature: oxidation and, for disulfides, reduction.
Desiccant: A moisture-absorbing material included in packaging. Its presence or absence changes the water content measured on arrival for hygroscopic material.
Freeze–thaw: Repeated freezing and thawing of a reconstituted solution. A common aggregation trigger and a standard forced-degradation stress.
Hygroscopic: Readily absorbing atmospheric moisture. Proline-rich short peptides and NAD+ are the worst offenders in this index.
Particulate, sub-visible: Particles below the visible threshold, counted by light obscuration under USP <788>. Not measured anywhere in this index.
Photostability: Resistance to light. Tryptophan-containing peptides and the quinolinium adjuncts degrade in clear glass; amber glass is the mitigation.
Vial fill volume: The nominal liquid volume filled before lyophilisation. Determines cake geometry and, with concentration, the mass per vial.

16 terms

Quality system

Accuracy: Closeness of a measured value to the true value. Demonstrated by recovery experiments during validation.
Precision: Closeness of repeated measurements to each other. Says nothing about accuracy: a precise method can be consistently wrong.
Certificate of analysis (COA): A document stating test results for a specific lot. In this market frequently lacking a method section, a lot number, or both.
Chain of custody: The documented sequence of possession of a sample from acquisition to analysis. Determines what a result can be attributed to.
Linearity: Proportionality of detector response to concentration across the working range. Required for an assay; irrelevant to area percent.
Range: The concentration interval over which accuracy, linearity and precision have been demonstrated.
Limit of detection (LOD): The lowest amount detectable, conventionally at a signal-to-noise ratio of 3.
Limit of quantitation (LOQ): The lowest amount quantifiable with acceptable precision, conventionally S/N ≥ 10 with RSD ≤ 10 %.
Robustness: A method's tolerance to deliberate small changes in conditions — temperature, pH, organic fraction.
Specificity: Demonstration that the response is due to the analyte and nothing else. The hardest characteristic to establish for peptides because of mass-identical isomers.
Specification: A decision about what result is acceptable. Not a measurement. Conflating the two is how "≥ 99.0 %" becomes "99.0 %" on a certificate.
Validation: Documented demonstration that a method is fit for its purpose, per ICH Q2(R2). Not published for any supplier method in this index.
Reference standard: A material of known content used to quantify an analyte. Absent for most research peptides, which is why content is rarely measured.
Retest date: The date by which a material should be re-tested rather than discarded. Distinct from an expiry date and less often stated.
Out of specification (OOS): A result outside the acceptance criteria. Requires investigation before it can be invalidated; a result cannot be discarded because it is inconvenient.
Pharmacopoeial monograph: An official specification and method for a named substance. Exists for oxytocin, desmopressin, gonadorelin, leuprolide, triptorelin and glucagon among the compounds here.

23 terms

Statistics

Bland–Altman analysis: Paired comparison of two measurement methods by plotting difference against mean. Yields a bias and limits of agreement. Used for inter-lab agreement here.
Bias: A systematic offset between a measurement and a reference. Distinct from scatter: bias in one direction across many observations is a different finding from random disagreement.
Limits of agreement: Mean difference ± 1.96 standard deviations of the difference. The interval within which 95 % of paired differences fall.
Bootstrap: Estimating a confidence interval by resampling the observed data. Useful where a supplier cell is too small for a parametric interval.
Confidence interval: A range that would contain the true value in a stated proportion of repeated samples. Wide intervals on small cells are the honest presentation.
Digit preference: Non-uniform distribution of terminal digits in reported figures. A real integrated area does not land on a round number repeatedly.
Interquartile range (IQR): The difference between the 75th and 25th percentiles. The index's preferred spread statistic because purity distributions are left-skewed.
Median: The 50th percentile. Preferred over the mean throughout this index because purity distributions have long left tails.
Median absolute deviation (MAD): The median of absolute deviations from the median, scaled by 1.4826 to estimate σ for normal data. Robust to outliers.
Quantile: A value below which a stated proportion of observations fall. The index uses the type 7 definition and says so.
Outlier: An observation beyond the Tukey fences at 1.5 IQR from the quartiles. Plotted individually on every box plot here rather than discarded.
Tukey fence: Q1 − 1.5 IQR and Q3 + 1.5 IQR. The convention for defining box-plot whiskers and outliers.
Percentage point (pp): The unit of a difference between two percentages. 98.7 % and 97.9 % differ by 0.8 pp, not 0.8 %.
Power: The probability of detecting an effect of a given size if it exists. Small supplier cells here have low power and the index states so rather than concluding no difference.
Regression to the mean: The tendency of an extreme observation to be followed by a less extreme one. Why a retest after a bad result usually looks better regardless of any change.
Repeatability: Precision under identical conditions in one session, same analyst, same instrument. The narrowest kind of precision.
Intermediate precision: Precision across days, analysts and instruments within one laboratory. Wider than repeatability and more realistic.
Reproducibility: Precision across laboratories. The index estimates it from paired same-batch comparisons.
Selection bias: Distortion arising because the sample was not drawn at random. The index's largest unmeasured confounder: lots that get tested are lots someone chose to test.
Survivorship: A related bias: the material that reaches a laboratory is not representative of material that does not. Compounded here by who chooses to submit and when.
Standard uncertainty (u): The uncertainty of a measurement expressed as a standard deviation. Combined across sources in quadrature.
Expanded uncertainty (U): Standard uncertainty multiplied by a coverage factor, usually k = 2 for approximately 95 % confidence.
Uncertainty budget: An itemised account of every contribution to a measurement's uncertainty. No supplier certificate in this index contains one.