Quick summary
Proteins can survive in some ancient materials after DNA has degraded, but rarely as complete molecules. Scientists extract fragments, identify their amino-acid patterns with mass spectrometry, compare them with reference proteins and infer the most plausible sequence or family. They recover evidence, then reconstruct what is missing; they do not read an intact original.
Why proteins can persist
DNA and proteins both decay, but mineral binding and protected environments can preserve protein fragments for long periods. Teeth, bone, eggshell and dental calculus may retain molecules. Heat, moisture, acidity and microbial activity accelerate damage, so age alone does not predict preservation.
From sample to peptide spectrum
- Sampling: a small amount is removed with archaeological context recorded.
- Cleaning and extraction: surface contamination is reduced and proteins are released.
- Digestion: enzymes may cut proteins into peptides with predictable endpoints.
- Mass spectrometry: instruments measure peptide masses and fragment them, producing spectra related to amino-acid order.
- Matching: software compares spectra with reference databases or derives candidate sequences directly.
How missing pieces are inferred
Observed peptides constrain sections of the protein. Homologous proteins from living and extinct relatives provide alternative residues for gaps. Evolutionary trees and statistical models estimate which sequence best explains the fragments and relationships among species. Structural models can test whether a proposed sequence plausibly folds, but structure prediction cannot create direct evidence for an unobserved residue.
Contamination and chemical damage
Modern human proteins, laboratory reagents and microbes can enter a sample. Researchers use blank controls, replicate extraction and independent laboratories where possible. Ancient proteins also undergo characteristic modifications, but damage is not a perfect timestamp. A convincing claim combines molecular patterns with sample context and contamination controls.
What confidence means
Confidence may apply at several levels: a peptide-spectrum match, a particular amino acid, a protein identity or a species assignment. These are not interchangeable. Closely related species can share most sequences, and short peptides may match many proteins. Good studies publish spectra, search settings, false-discovery controls and alternative interpretations.
Reality check
A reconstructed ancient protein is not necessarily a complete molecule recovered from the specimen. Some positions are directly observed, some are strongly inferred and others remain unknown. A visually complete sequence can conceal those different evidence levels unless uncertainty is marked position by position.
Why the method matters
Palaeoproteomics can help identify species, study diets and relationships, and extend molecular evidence beyond environments where ancient DNA survives. Its strength is greatest when it complements anatomy, archaeology and DNA rather than replacing them. Better databases and transparent uncertainty will improve reconstructions, but cannot restore information destroyed by decay.
