How to Calculate Protein Isoelectric Point (pI)

The isoelectric point (pI) of a protein is the pH at which it carries zero net charge. It is a single number with outsized practical reach: at its pI a protein is least soluble and will not move in an electric field, which is exactly the principle behind isoelectric focusing, the starting condition for ion-exchange chromatography, and the pH you avoid when you want a protein to stay in solution. This guide builds pI from first principles — the ionizable groups, their pKa values, the Henderson–Hasselbalch charge of each, and the sum over the whole chain — works two examples by hand, and explains why a whole-protein pI must be found by iteration rather than by averaging two pKa values. The protein isoelectric point calculator runs the same titration and bisects to the zero-crossing so you can check each step.

What “net charge zero” actually means

A protein is not a single acid or base; it carries many ionizable groups at once — the two chain termini plus every acidic and basic side chain. Each one is either protonated or deprotonated depending on the pH, and each state carries a charge. At low pH almost everything is protonated and the protein is net positive; at high pH almost everything is deprotonated and it is net negative. Somewhere in between the positives and negatives balance and the sum is zero. That pH is the pI.

So pI is not a property of one group — it is the pH where a whole set of partial charges adds to zero, which is why it depends on the full sequence composition.

The ionizable groups and their pKa values

Only a subset of residues can gain or lose a proton in the normal pH range. Each has a characteristic pKa — the pH at which that group is half protonated. This tool uses the EMBOSS pKa set:

Group	Type	pKa
α-amino terminus (N-term)	basic	8.6
Lysine (K) side chain	basic	10.8
Arginine (R) side chain	basic	12.5
Histidine (H) side chain	basic	6.5
α-carboxyl terminus (C-term)	acidic	3.6
Aspartate (D) side chain	acidic	3.9
Glutamate (E) side chain	acidic	4.1
Cysteine (C) side chain	acidic	8.5
Tyrosine (Y) side chain	acidic	10.1

A basic group is positively charged when protonated (low pH) and neutral when it loses the proton. An acidic group is neutral when protonated and negative when it loses the proton. All the other residues (A, N, Q, G, I, L, M, F, P, S, T, V, W) are not ionizable in this range and contribute nothing to the charge.

The charge of one group: Henderson–Hasselbalch

For a single group, the fraction protonated follows from the Henderson–Hasselbalch relation. The charge each group contributes at a given pH is:

• Basic group (charge +1 when protonated): q = +1 / (1 + 10^(pH − pKa))
• Acidic group (charge −1 when deprotonated): q = −1 / (1 + 10^(pKa − pH))

At pH = pKa each formula gives ±0.5 — the group is half charged, as it should be. Far below its pKa a basic group is fully +1; far above, it is 0. The mirror holds for acidic groups.

Net charge of the whole chain

Sum the contribution of every ionizable group, weighting by how many of each the sequence contains:

Q(pH) = Σ(basic groups) +1/(1+10^(pH−pKa)) − Σ(acidic groups) 1/(1+10^(pKa−pH))

The N-terminus counts once as a basic group, the C-terminus once as acidic, and each K/R/H and D/E/C/Y counts once per occurrence. Q(pH) is a smooth, monotonically decreasing curve: it starts positive at low pH and ends negative at high pH, so it crosses zero exactly once. The pI is the pH where Q(pH) = 0.

Why you iterate, not average two pKa

A common textbook shortcut says “pI is the average of the two pKa values that straddle the neutral charge.” That is correct only for a single amino acid or a very simple molecule with one acidic and one basic group. For a real protein with dozens of ionizable groups, the zero-crossing is set by the collective sum of all of them — there is no single pair to average. The reliable method is to evaluate Q(pH) across the range and find where it changes sign, then narrow in. The calculator does this by bisection between pH 0 and 14 until the net charge is zero to within a thousandth. Any guide that teaches the two-pKa average for whole proteins is oversimplifying.

Worked example 1: glycine (the case where the shortcut is exact)

Glycine is a single residue, so it has just two ionizable groups: the α-amino terminus (pKa 8.6) and the α-carboxyl terminus (pKa 3.6), and no side chain. With one basic and one acidic group, the two-pKa average is exact:

pI = (pKa_N-term + pKa_C-term) / 2 = (8.6 + 3.6) / 2 = 12.2 / 2 = 6.10

The calculator returns pI = 6.10 for glycine. This is the one situation where averaging works — and it is precisely why the shortcut survives in textbooks despite failing for real proteins.

Worked example 2: net charge of ubiquitin at pH 7

Human ubiquitin (76 residues) contains the ionizable groups K 7, R 4, H 1 (basic) and D 5, E 6, C 0, Y 1 (acidic), plus one N-terminus and one C-terminus. Evaluate each group’s charge at pH 7:

Group	Count	Charge each at pH 7	Total
N-term (pKa 8.6)	1	+0.9755	+0.9755
K (pKa 10.8)	7	+0.9998	+6.9989
R (pKa 12.5)	4	+1.0000	+4.0000
H (pKa 6.5)	1	+0.2403	+0.2403
C-term (pKa 3.6)	1	−0.9996	−0.9996
D (pKa 3.9)	5	−0.9992	−4.9960
E (pKa 4.1)	6	−0.9987	−5.9925
Y (pKa 10.1)	1	−0.0008	−0.0008

Summing: positive groups give +12.2147, negative groups give −11.9889, so Q(pH 7) = +0.226. The calculator reports +0.23. Because the net charge is still slightly positive at pH 7, the pI must be a little above 7 — and indeed bisecting to Q = 0 gives pI = 7.54 for ubiquitin with this pKa set. Notice how the single histidine (only +0.24 charged at pH 7, since its pKa of 6.5 is close to 7) and the lone tyrosine barely register; the pI is dominated by the lysines and arginines against the aspartates and glutamates.

What pI means at the bench

• Solubility and precipitation. A protein is least soluble at its pI because with no net charge there is no electrostatic repulsion to keep molecules apart — they aggregate. This is how isoelectric precipitation works, and why you generally buffer away from the pI to keep a protein in solution.
• Isoelectric focusing (IEF) and 2-D gels. In a pH gradient under an electric field, a protein migrates until it reaches the pH equal to its pI, then stops. That focusing is the first dimension of 2-D proteomics.
• Ion-exchange chromatography. At a buffer pH below the pI the protein is net positive and binds a cation exchanger; above the pI it is net negative and binds an anion exchanger. Knowing the pI tells you which resin and starting pH to choose. The molarity calculator and dilution calculator then handle the buffer preparation.

Limitations: theoretical versus measured pI

The number here is a theoretical pI from a fixed pKa set, and it can differ from an experimental value by a few tenths of a pH unit. Reasons:

• pKa-set dependence. Different published sets (EMBOSS, Bjellqvist, DTASelect, Solomon) give different pI for the same sequence. This tool uses EMBOSS and says so.
• Context. Real pKa values shift with the local environment — a buried residue, a neighbouring charge, a bound metal or a disulfide all move a group’s pKa away from the model value.
• Post-translational modifications. Phosphorylation adds a negative charge (lowers pI); acetylation of a lysine removes a positive one (also lowers pI). The unmodified-sequence pI ignores these.
• Solution conditions. Ionic strength and temperature shift the effective pI from the dilute-solution model.

Treat the computed pI as a planning estimate — excellent for choosing a chromatography direction, not a substitute for a measured value.

Common mistakes

• Averaging two pKa for a whole protein — valid only for a single amino acid like glycine, wrong for anything with many ionizable groups.
• Counting non-ionizable residues (S, T, N, Q, etc.) as if they carried charge; they do not, in the normal pH range.
• Forgetting the termini — the N- and C-terminus each contribute one ionizable group.
• Assuming cysteine always titrates — a cysteine locked in a disulfide or bound to a metal does not behave like a free thiol.
• Comparing pI across tools and calling one wrong when they simply use different pKa sets.
• Ignoring PTMs when the purified protein is phosphorylated or acetylated.

Where this fits in the wider workflow

pI is a sequence-derived property, so it pairs naturally with the other two protein tools. Get the mass first with the protein molecular weight calculator (the how to calculate protein molecular weight guide derives it), use the pI to pick a purification or focusing condition, then quantify the purified protein with the protein concentration calculator — see the how to determine protein concentration from A280 guide. All three live in the protein calculators hub and share the same sequence input, so the numbers stay consistent end to end.