Educational guides and overviews on research peptides, what they are and how they're studied in the laboratory. For research use only.
Orthogonality is a property of the whole set of protecting groups, not of any one member. What the trityl/Acm/Mob and trityl/Acm/tBu triads cost, which deprotection reagent damages which residue, why the order of closure is a design decision, and why the connectivity still has to be measured rather than assumed.
Read moreOxidation does not know which cysteines you meant to join. When a single thermodynamic fold makes random oxidation legitimate, when orthogonal protection is the only honest route, and how concentration, redox buffer and oxidant choice decide which isomer you get.
Read moreLyophilised peptide solids commonly return an amorphous halo, and a halo is a weaker statement than it looks. This guide covers what X-ray powder diffraction licenses you to conclude about crystallinity and polymorphism, why the halo also fits disordered nanocrystalline material, how calorimetry, solid-state NMR and vibrational spectroscopy fill the gap, where residual water pulls the glass transition, and how preferred orientation and grinding can manufacture the result you report.
Read moreLong-term, intermediate and accelerated arms answer different questions. What bracketing assumes about monotonicity and matrixing costs in power to detect a slope difference, how the shape of the degradation curve decides where pull points can be removed, and why an activation energy that changes with the condition means a 40 °C ranking is not a 5 °C ranking.
Read moreRepeatability and the tolerance you commit to fix a floor below which a weighed mass carries an offset you cannot see. What static charge, buoyancy and ambient water do to a milligram of dry powder, and how that error reappears unchanged in every dilution.
Read moreCyclisation narrows the conformational distribution an NMR experiment averages over, but it does not guarantee a single conformer. What cis/trans exchange data, amide temperature coefficients, NOEs and residual dipolar couplings each license, and where a single fitted structure becomes a shape that is never populated.
Read moreA benchmark AUC describes the ordering of one curated set under one choice of negatives; it is not a probability that your designed sequence has the property. What one-hot encodings, physicochemical descriptors and protein language model embeddings can each carry for a 12-mer, why random splits leak through sequence similarity, what the negative-sampling result does to published AMP benchmark analyses, and how to report an applicability-domain distance so a score can be read at all.
Read moreA deletion peptide is one residue short and chromatographically close to the target, which is what makes it the hardest impurity to remove. Capping converts that failure into a truncated fragment that stops growing and, given a purification handle, is far easier to remove - at a cost in yield and in acylation chemistry applied to every chain on the resin.
Read moreWhen a coupling stalls because the resin-bound chains have formed beta-sheet, a lithium or potassium salt, a different solvent or a hot coupling can each move the outcome. What the published data support, and what a rescued crude profile does not prove.
Read moreAn electrophoretic light scattering instrument measures mobility; the zeta potential it prints is what a chosen model makes of that mobility. Which model applies depends on kappa-a, and the two common limits differ by a factor of 1.5 in the potential returned from the same mobility.
Read morePolysorbate 20 and 80 and poloxamer 188 inhibit surface adsorption by different mechanisms - one by occupying the interface, the other by complexing the solute in bulk. What the critical micelle concentration does and does not license you to conclude, why hydrolysis and oxidation of the polysorbate ester have opposite consequences for interfacial protection, why quantifying polysorbate against a protein background is hard, and why SDS belongs in a structure experiment rather than a formulation.
Read moreAn 18.2 megohm reading is an ionic measurement, not a statement of cleanliness. What water purification grades, solvent grades, leachables and trace metals actually control in peptide work — and what each specification does not cover.
Read moreA failing result is data whose referent is still unknown: it may describe the material or the measurement. How the laboratory phase of an investigation is bounded, why invalidation needs an identified cause rather than a plausible one, what separates a retest from a resample, why averaging a failure with passing repeats conceals rather than resolves, and the narrow claim an outlier test actually supports.
Read morePermeability in this class tracks how many backbone donors are still exposed at the membrane rather than how many the molecule owns, though no single count settles it. The size and polarity boundaries, the intramolecular hydrogen bonding that shields donors in a low dielectric, how the two conformations are demonstrated by NMR in polar and apolar solvent, N-methylation as the designed alternative, and why PAMPA and cell monolayers disagree.
Read moreSequence-based predictors return a ranked list of scissile bonds, not a verdict. What the subsite matrices encode, why coverage in the underlying cleavage databases is so uneven, where structural accessibility breaks a sequence-only score, and how LC-MS of the actual digest settles the question.
Read moreA C-terminal hydrazide survives Fmoc synthesis where a thioester does not, then unmasks in the same tube: nitrite at -20 to -10 C gives the acyl azide, MPAA gives the thioester, and native chemical ligation consumes it. Conditions, resins, junction limits, and what the nitrite step really damages.
Read moreThree cysteines and one tris-electrophile close two fused macrocycles in a single aqueous step. What the reaction demands, what competes with it, and why a correct product mass is not proof of the intended connectivity.
Read moreA contaminated Fmoc amino acid does not spread its damage across a synthesis; it puts a defect at one residue, every time. D-enantiomer, dipeptide, beta-alanine and acetic acid each fail differently, and a high RP-HPLC purity figure is compatible with all four.
Read moreLight obscuration and flow imaging return different particle counts from the same tube, and the gap is systematic rather than random. Where each method is blind, why refractive index drives the disagreement, how sample handling manufactures the particles you are measuring, and why a subvisible count and an SEC aggregate peak describe different populations.
Read moreCounter-current chromatography holds one liquid phase in the column by centrifugal force and pumps the other over it, so there is no solid surface to bind a peptide irreversibly. What that buys, what governs the solvent system, and where it fits beside preparative reversed-phase.
Read moreA mass shift tells you an oxygen was added or an amide became an acid, not where or to what. This walks through MS/MS localisation, chromatographic corroboration, and the orthogonal tests that turn a candidate degradant into an assigned structure.
Read moreA container closure integrity result is a statement about a leak path, not about sterility. This guide traces how the maximum allowable leakage limit was derived from helium leak rate and microbial ingress data, compares helium, vacuum decay, high-voltage leak detection and laser headspace analysis on what each actually measures, and works through why positive controls, residual seal force and low temperature complicate every number - ending on the dried peptide, where a breach can leave the routine release attributes unchanged.
Read moreContact counts are conditional on the cutoffs that produced them, so they do not transfer between programs or between papers. Why hydrogen bond and salt bridge definitions disagree by construction, what trajectory occupancy adds and what it inherits from the force field, why a buried charge pair can be stabilising or destabilising depending on desolvation cost, and what amide temperature coefficients, hydrogen exchange and double-mutant cycles each actually license you to conclude.
Read moreSelenocysteine has a side-chain pKa of 5.2, so the selenolate dominates at neutral pH. Paired with a peptide selenoester it gives ligations that finish in minutes at junctions where a thioester stalls, and a deselenisation step that leaves cysteine thiols alone.
Read more