In Silico Protease Digestion and Peptide Mass Fingerprinting Explained
6 min read · Updated July 10, 2026
Before you run a protein through a mass spectrometer, or before you order a construct with a tag you plan to cleave off, it helps to know exactly where a protease will cut and what size fragments come out the other side. In silico protease digestion predicts those cut sites and peptide masses directly from the sequence, using the same cleavage rules the real enzyme follows. This guide covers how trypsin, Lys-C and chymotrypsin choose their cut sites, why missed cleavages matter, and how peptide mass is calculated once you have the fragments.
How trypsin chooses its cut sites
Trypsin is the default protease in proteomics because its specificity is simple and well behaved: it cleaves the peptide bond on the C-terminal (carboxyl) side of lysine (K) or arginine (R). That single rule is what makes an in silico digest predictable — scan the sequence for every K and R, and each one is a candidate cut site.
There is one important exception. Trypsin essentially does not cleave a K/R bond when the next residue is proline (P). A K-P or R-P bond resists cleavage strongly enough that it's treated as a non-cut site, not just a slow one. A digestion prediction that ignores this rule will report peptides that never actually appear in a real digest, and will miss the longer peptide that spans across the K-P or R-P junction instead.
Lys-C and chymotrypsin: different rules, different fragments
Swapping the protease changes both where the cuts fall and how many peptides you get, which is useful when trypsin's fragments aren't the right size for your experiment.
Lys-C cleaves only C-terminal to lysine — arginine is left alone entirely. Because it recognizes only one of the two residues trypsin targets, Lys-C produces fewer, longer peptides from the same protein, which can be helpful for resolving regions where a dense cluster of K and R residues would otherwise generate a run of very short tryptic peptides.
Chymotrypsin works on a different chemical logic: it cleaves C-terminal to bulky aromatic or hydrophobic residues, mainly tyrosine (Y), tryptophan (W) and phenylalanine (F), with weaker, less consistent cleavage after leucine (L) or methionine (M). Because its preferred residues are far less common than K and R, a chymotryptic digest tends to leave larger gaps between cut sites in K/R-poor stretches, and it's often used specifically to cover regions a tryptic digest leaves as one long, unhelpful peptide.
Missed cleavages: why a real digest is never fully cut
A naive prediction that cuts at every possible site and stops there doesn't match reality. Digestions are rarely, if ever, 100% complete — some fraction of the enzyme's recognized sites survive intact in any given molecule, usually because of steric hindrance, incomplete digestion time, or a cut site sitting too close to another cut site or the protein's end.
That means a useful prediction has to report more than the fully-cleaved peptide list. It should also generate peptides that retain one or more missed cleavage sites — a peptide spanning two tryptic fragments joined at an uncut K or R, for instance. An actual experimental digest is a mixture of fully- and partially-cleaved products, and matching observed masses against only the fully-cleaved list will leave real peaks in your spectrum unexplained.
Calculating peptide mass
Once the cut sites are decided, each resulting fragment's mass is calculated exactly the way whole-protein molecular weight is calculated — just applied to a shorter stretch of sequence. Sum the residue masses of every amino acid in the peptide, then add the mass of one water molecule (the peptide's free N-terminal amine and C-terminal carboxyl group account for that extra water compared with a mid-chain residue).
Because this is the same arithmetic used for full-length molecular weight, the same average-mass tables apply. It's just done once per predicted fragment instead of once for the whole protein, which is what turns a list of cut positions into a list of expected masses.
Putting predicted peptides to work
The main use for this kind of prediction is peptide mass fingerprinting: you digest a protein experimentally, measure the resulting peptide masses by mass spectrometry, and match those observed masses back against a predicted digest to confirm identity or localize where in the sequence a modification, mutation or missing peptide sits. A predicted mass list — including missed-cleavage variants — is what you match the spectrum against.
Choosing a protease is also a practical, upfront decision. A digest full of very short peptides (a few residues) is hard to detect and identify by MS, and a digest full of very long peptides is hard to ionize and fragment well; predicting the fragment sizes for trypsin, Lys-C and chymotrypsin before you commit lab time lets you pick whichever produces peptides in an easier-to-work-with size range for your protein.
It's also worth checking a construct this way before you order it. If you're appending a fusion tag or a linker, digesting the full fusion sequence in silico shows whether the protease you plan to use during sample prep will cut inside the tag, inside the linker, or leave the tag attached to your protein of interest when you didn't intend either outcome.
Where Protease Digestion fits in
SeqBench's Protease Digestion tool applies these cleavage rules for you — trypsin, Lys-C, chymotrypsin and other common proteases — to a pasted protein sequence, returning the resulting peptides and their masses, rather than requiring you to work out K-P exceptions or sum residue masses by hand.
Before or after digesting, Protein Properties gives you the full-length molecular weight and composition for the same sequence, which is useful context when you're deciding whether a digest's peptide sizes make sense. If you're evaluating a construct with a tag or linker, the Hydrophobicity Plot can help you spot which stretches of the fusion are likely to sit at the surface versus be buried, which is relevant alongside cleavage site location when judging accessibility to the protease.
Frequently asked questions
Why does trypsin sometimes not cut after a lysine or arginine?
Trypsin fails to cleave a K/R-proline bond even though proline itself isn't the residue trypsin recognizes. Any tool predicting a tryptic digest should treat K-P and R-P as non-cut sites, or it will report peptides that don't actually exist in the real digest.
How do I predict the peptide masses I'd see in a mass spec run?
Digest the protein sequence in silico with the protease you actually used (usually trypsin), then sum each peptide's residue masses plus one water molecule to get its mass. Matching those predicted masses against your observed spectrum is the basis of peptide mass fingerprinting.
What's the difference between trypsin and Lys-C digestion?
Trypsin cuts after both lysine and arginine, while Lys-C cuts only after lysine. Because it skips every arginine site that trypsin would cut, Lys-C produces fewer, longer peptides from the same protein.
Do I need to account for missed cleavages when predicting peptide masses?
Yes. Real digests are never 100% complete, so alongside fully-cleaved peptides you should also predict peptides with one or more missed cleavage sites, since both will be present in an actual experimental sample.
Related references
Common affinity, epitope and solubility protein tags with amino acid sequences, mass, purification/detection and protease cleavage.
One-letter and three-letter amino acid codes with key properties.
Side-chain and terminal pKa values and how they set the isoelectric point.