123 123Calculator

Isoelectric Point Calculator

Sort the pKa values rather than labelling them — tyrosine is the case where the usual shortcut fails.

Clear
Isoelectric point9.74Lysine
Net charge at pH 7.40.9719positive — it migrates to the cathode
Carboxyl pKa2.18
Amino pKa8.95
Side chain pKa10.53basic side chain
pKa values averaged8.95 and 10.53the two ionisations either side of the neutral form, found by sorting rather than by which group is which
Net charge at the isoelectric point2.7542e-8zero, by definition — this row is the check
A molecule is least soluble at its pI, because with no net charge nothing keeps neighbouring molecules apart. That is why casein precipitates out of milk at pH 4.6 and why isoelectric focusing separates proteins so sharply.

The formula

pI = average of the two pKa values flanking the neutral form

Which two pKa values

At the isoelectric point the molecule carries no net charge, and the pI is the average of the two ionisations either side of that neutral form. For an amino acid with an inert side chain those are simply the carboxyl and the amino group, giving glycine a pI of 5.97.

Sort the pKa values, do not label them

The usual shortcut says an acidic side chain pairs with the carboxyl and a basic one with the amino group. It works for aspartate, whose pI is (1.88 + 3.65)/2 = 2.77, and for lysine at (8.95 + 10.53)/2 = 9.74. It fails for tyrosine: its phenol side chain is acidic but its pKa of 10.07 sits ABOVE the amino group's 9.11, so it is the last proton to leave rather than the second. The shortcut gives 6.14 where the real value is 5.66.

Sorting works in every case. Count the groups that carry a positive charge when protonated — the amino group, plus a basic side chain if there is one. That is how many protons must come off to reach neutrality, and the pKa values either side of that step are the pair to average, whatever chemical class they belong to.

Why it is the number that matters

A protein is least soluble at its pI, because with no net charge there is nothing to keep molecules apart and they aggregate. That is the basis of isoelectric precipitation, and it is why casein drops out of milk at pH 4.6. It is also why electrophoresis works: at any other pH the molecule carries charge and moves in a field, in a direction the pI predicts.

Where these models stop

Michaelis-Menten assumes a single substrate and a steady state; Langmuir assumes one layer on identical sites with no interaction between them; Stokes-Einstein assumes a hard sphere much larger than the solvent molecules around it. Each is an idealisation that happens to describe real systems well over a useful range, and each fails predictably outside it. Knowing which assumption a number rests on is usually more valuable than the number.