Outliers in Laboratory Data: When Excluding a Value Is Justified
When can a laboratory legitimately exclude an outlier? Assignable causes, what Grubbs-type tests really show, and why criteria must come first.
Deamidation or Isotope Peak? Reading the 0.984 Da Shift in Peptide Data
Deamidation adds 0.984 Da to a peptide. Learn how to tell it from the carbon-13 isotope peak and how it shows up in LC-MS and HPLC data.
How a Peptide Stability Protocol Is Designed
What a peptide stability protocol specifies, how real-time and accelerated studies differ, what is tested at each pull, and how a date is derived.
TFA, Formic Acid and Other Mobile Phase Additives for Peptide HPLC
How TFA, formic acid, HFBA and high-pH buffers change peptide HPLC peak shape, retention and MS signal, and why labs often run two separate methods.
Gradient Elution in Peptide HPLC: Why Methods Ramp the Solvent
Why peptides are separated by gradient rather than isocratic HPLC, how slope and dwell volume shape retention, and why late-run baselines drift.
Karl Fischer Titration and Water in Lyophilized Peptides
Why freeze-dried peptides keep some water, how Karl Fischer titration measures it, and why loss on drying and HPLC purity figures tell you something else.
Cross-Contamination at the Peptide Bench: Finding and Closing the Routes
Where peptide cross-contamination happens on the lab bench, how to tell a stray peak from real chemistry, and bench habits that remove the risk by design.
Mass Accuracy in ppm: How Much a Peptide Mass Match Actually Proves
How mass accuracy in ppm works for peptide identity results, what a mass match really proves, and which isomers intact mass alone cannot distinguish.
Detection and Quantitation Limits for Peptide Impurities: Reading the Fine Print
LOD, LOQ and reporting thresholds explained: why a not detected line on a peptide HPLC report depends on wavelength, sample load and method cutoffs.
Technical vs Biological Replicates: Counting What Counts
Technical vs biological replicates explained: what each measures, why pseudoreplication shrinks error bars, and how to report n honestly.