01Identify which protein the sequence can represent

The original study by Stergachis and colleagues on selecting peptides for targeted proteomics describes useful proteotypic peptides in terms of protein uniqueness and suitable mass-spectrometry behaviour. A readily measured sequence must also distinguish the intended protein in the relevant background.Stergachis and colleagues — Rapid empirical discovery of optimal peptides for targeted proteomics (opens in a new tab)

Look for the sequence and protein identifier together. A protein name alone may hide several forms, and a sequence found in more than one candidate cannot resolve their individual contributions simply because its analytical peak is clear.

The table is a reasoning example, not a database search result. In an actual paper, check the sequence database, relevant protein forms and assignment rules before accepting the word unique.

A shared peptide can still answer a deliberately combined question. The problem arises when a combined measurement is described as though it belongs exclusively to one protein.

02Keep peptide recovery separate from starting protein amount

Hoofnagle and colleagues’ recommendations explain that incomplete digestion, peptide loss and protein-form differences can complicate inference from liberated peptides to endogenous protein amount. Measuring the resulting peptide well does not erase those preceding steps.Hoofnagle and colleagues — Recommendations for the generation, quantification, storage and handling of peptides used for mass spectrometry-based assays (opens in a new tab)

Consider an original count illustration. Two samples each begin with 100 copies of a protein, but the measurable surrogate yield is 60 copies in one and 90 in the other. A 1.5-fold peptide difference would exist despite equal starting protein counts.

This simplified illustration does not model a specific digestion or instrument. It makes the missing link visible: the study needs evidence that the peptide comparison appropriately represents the original material.

Read what was evaluated about processing consistency and recovery. If the report only validates instrument response after digestion, keep that boundary in the protein-level conclusion.

03Locate the standard on the sample timeline

Draw a short timeline from starting sample to measurement and mark where the standard was added. A reference introduced after a stage cannot directly experience losses that occurred before its introduction.

For example, a peptide standard added to a finished digest can be informative about its later measurement. Its presence alone does not reveal how much parent protein was lost during an earlier extraction.

A protein reference introduced earlier raises a different question: whether it behaves sufficiently like the endogenous material through processing. Do not assume that any material carrying the same protein name automatically has equivalent recovery.

Also retain whether the output is a relative comparison or a calibrated amount. A ratio to an internal standard is part of a calculation; its biological unit should come from the documented method.

04Inspect agreement between the chosen surrogates

When several peptides represent one protein, look at whether they support the same comparison. Agreement can be informative, while disagreement deserves explanation before a single protein-level number is repeated.

In a fictional dataset, one surrogate rises twofold and another is unchanged. Averaging them into a confident increase would conceal the conflict. The useful next question concerns the sequence assignments, processing and evidence behind each measurement.

Report the protein target, surrogate sequences, standard type and addition stage alongside the supported quantitative conclusion. This gives a reader enough information to recognise whether the result concerns a particular protein form, a broader group or recovered peptide material.