Work out a protein’s properties

Mass is the reliable one. Sum the residues, add a water, and the answer is right to the precision of the atomic weights — with one decision to make: average mass for ordinary use, monoisotopic for mass spectrometry, where the difference is a whole dalton on a small protein.

The extinction coefficient at 280 nm is next most reliable, and it is the number that gets you a concentration from a spectrophotometer. It counts tryptophans, tyrosines and disulfide bonds — bonds, not cysteines, which is why a reduced and an oxidised protein have different coefficients and why the tool has to ask. A protein with no tryptophan rests entirely on tyrosine and is worth a few per cent of doubt; one with neither does not absorb at 280 at all, and no amount of arithmetic will get a concentration out of that reading.

Isoelectric point is the least reliable, and not because the arithmetic is hard. It depends on which published pKa set you use, and the common sets disagree — which is why two calculators give different answers for the same sequence. Worse, it assumes every ionisable group is freely exposed, and in a folded protein many are not. Treat a computed pI as a starting point for choosing a buffer or an ion exchange column, not as a measurement.