The isoelectric point calculator above reads a peptide or protein sequence in one-letter code, counts every ionisable group it contains, and solves for the pH at which the positive and negative charges cancel exactly. It also reports the net charge at any pH you choose, which in practice is the number people actually need: whether a molecule will bind a cation exchanger at pH 5, or run towards the anode at pH 8.5.
Arb Digital builds free calculators that show their working rather than hiding it. A pI is not a physical constant you can look up to two decimal places; it is the output of a model built on a set of assumed pKa values, and different published sets give different answers. This page lets you switch between two standard sets, or type in your own, so the sensitivity of the answer is visible instead of implied.
What This Isoelectric Point Calculator Does
Seven things in a protein carry charge: the free amino terminus, the free carboxyl terminus, and the side chains of aspartate, glutamate, cysteine, tyrosine, histidine, lysine and arginine. Everything else is neutral across the useful pH range. The calculator counts those groups in your sequence, applies the Henderson–Hasselbalch relation to each one to get its fractional protonation at a given pH, sums the result, and searches for the pH where the total is zero.
The output panel gives the pI as the headline, the net charge at your chosen pH, the residue count, and how many basic and acidic groups the sequence contains. The bar breakdown shows net charge at five reference pH values so you can see the shape of the titration curve at a glance rather than having to plot it.
There is a clean boundary with the two adjacent tools on this site. The protein molecular weight calculator turns a sequence into a mass in daltons and an extinction coefficient at 280 nm, and it states outright that it does not compute a pI because that needs a full charge-state model. This page is that model. Meanwhile the protein solubility calculator handles ammonium sulfate salting out, which is a different route to the same goal of getting a protein out of solution. The pI tells you where charge-based methods stop working; salting out is what you reach for when they do.
How to Use It
- Paste the sequence in one-letter code. FASTA headers, numbering, spaces and line breaks are stripped automatically, so you can paste straight from a database record.
- Pick a pKa set. The classic textbook set and the EMBOSS set are both in wide use and both defensible. Selecting Custom lets you override any individual value.
- Set the reporting pH to whatever your buffer actually is. This is the number that predicts binding and migration behaviour.
- Read the net charge sign. Positive means the molecule binds a cation exchanger and moves towards the cathode; negative means the opposite.
- Switch pKa sets and watch the pI move. If it moves a lot, the answer is model-limited and should be treated as a range.
The Formula and How It Is Calculated
Each basic group contributes a positive charge equal to its protonated fraction, 1 / (1 + 10pH − pKa). Each acidic group contributes a negative charge equal to its deprotonated fraction, −1 / (1 + 10pKa − pH). The net charge is the sum over every group, with the N-terminus treated as basic and the C-terminus as acidic. The pI is the root of that function, found here by bisection over the range pH 0 to 14 with two hundred iterations, which converges far below the precision the underlying pKa values justify.
For a single amino acid with no ionisable side chain the answer collapses to the familiar textbook shortcut: the pI is the mean of the two flanking pKa values. Alanine with a carboxyl pKa of 2.4 and an amino pKa of 9.6 gives a pI of 6.00, which is the value the calculator returns and which sits alongside the published figure of about 6.01 in the OpenStax treatment of amino acids and isoelectric points. For anything with charged side chains the shortcut breaks down and the full sum is required, which is what this page does.
The default example, the octapeptide DRVYIHPF, has one aspartate, one arginine, one tyrosine and one histidine plus the two termini. Its charges do not cancel at any simple midpoint, and the calculator returns a pI just below 7.8 with the classic set. Change to the EMBOSS set and it drops by about two tenths of a unit, which is a fair illustration of how much confidence a quoted pI deserves.
Why Two pKa Sets Give Two Answers
The pKa values in every published set are averages. They were measured on free amino acids or on short model peptides, in dilute aqueous solution, at a stated temperature and ionic strength. A residue buried in a folded protein experiences none of those conditions. Its effective pKa is shifted by the local dielectric environment, by hydrogen bonding, and above all by neighbouring charges: an aspartate sitting next to a lysine is harder to deprotonate than a free one, sometimes by more than a full pH unit.
This is why an experimentally measured pI, from isoelectric focusing, frequently differs from a calculated one by half a unit or more, and occasionally by much more for proteins with unusual charge clustering. The calculated value is a sequence-based prediction that ignores structure entirely. It is a good starting point for method development and a poor substitute for a measurement when the answer matters.
The practical response is not to chase precision but to use the number as a range. If two pKa sets put a protein between 6.4 and 6.9, plan your ion exchange step at pH 5 or pH 8.5, not at pH 6.6. Working close to a predicted pI is where prediction error costs the most, because that is exactly where the net charge is changing fastest with pH.
Reading the Net Charge Rather Than the pI
For most laboratory decisions the net charge at your working pH is the more useful output, and it is the one people skip past. The rule is simple. Below the pI the molecule is net positive; above it, net negative. A cation exchange resin binds positive species, so you load below the pI. An anion exchanger binds negative species, so you load above it. In electrophoresis a net negative molecule migrates towards the anode.
What the sign does not tell you is how strongly. A protein one unit from its pI may carry a fractional net charge and bind weakly and reversibly; two units away it may carry several charges and bind tightly enough to need a salt gradient to release. The magnitude column in the bar breakdown is there to make that comparison quick.
The other subtlety is that binding depends on local surface charge patches, not only on the net total. A protein with a net charge of zero can still bind an ion exchanger if one face of it is strongly positive. Net charge is a good first predictor and a poor guarantee, which is why chromatography method development is empirical. Once you have a buffer in mind, the pH calculator and the Henderson-Hasselbalch calculator handle its preparation, and the ionic strength calculator tells you how much electrostatic screening the salt in it will supply.
Free Amino Acids, Peptides and Whole Proteins
The same model covers all three, but the reliability changes sharply. For a free amino acid the calculation is essentially exact, because there is no structure to perturb the pKa values and the measured constants are the ones being used. For a short unstructured peptide it remains good, with errors of a tenth or two. For a folded globular protein it becomes an estimate.
Two sequence-level effects also matter. Blocked termini are common: many peptides are synthesised with an acetylated N-terminus or an amidated C-terminus, which removes that charge entirely. If your molecule is blocked, set the corresponding terminal pKa far outside the range, for example an N-terminal value of −5 or a C-terminal value of 25, so its contribution vanishes. Post-translational modifications do the same thing at side chains: phosphorylation adds a strongly acidic group that this page cannot see from the sequence alone.
Disulfide bonds matter too. Two cysteines joined in a disulfide are no longer ionisable, so a protein with four disulfides has eight fewer acidic groups than its sequence suggests. In a reducing buffer they are free again. That single distinction can move a calculated pI noticeably in cysteine-rich proteins, and it is entirely invisible to a sequence-only calculation.
Where the pI Actually Gets Used
Isoelectric focusing separates proteins by driving them through a pH gradient until each stops at its own pI, which is the highest-resolution charge-based separation available and the first dimension of classical two-dimensional gel electrophoresis. A calculated pI tells you roughly where in the strip to expect a band.
Isoelectric precipitation exploits the solubility minimum at the pI. Casein precipitating from milk at pH 4.6 is the everyday example, and it is the same principle used to crash a recombinant protein out of a lysate before further purification. Whether that is useful depends on the protein tolerating the pH swing, which many do not.
Formulation is the third area. A therapeutic protein stored at a pH close to its pI aggregates readily, because the electrostatic repulsion that keeps molecules apart has been removed. Formulation buffers are therefore chosen deliberately away from the pI. When you are working out how much of a purified protein you have in hand, the protein concentration calculator converts an A280 reading or a standard curve into mg/mL, and the free ExPASy ProtParam tool is the reference implementation most groups compare their numbers against.
Arb Digital publishes hundreds of free calculators across chemistry, maths, finance and marketing — no sign-up, no limits. If something you need is missing, tell us and we will look at building it.
Browse All Free Tools Suggest a ToolCommon Mistakes to Avoid
- Quoting a calculated pI to two decimal places — the underlying pKa values are averages, and a measured pI often differs by half a unit or more.
- Forgetting blocked termini — an acetylated N-terminus or an amidated C-terminus removes a charge the sequence still implies.
- Counting disulfide-bonded cysteines as ionisable — they are not, unless the sample is in a reducing buffer.
- Ignoring post-translational modifications — phosphorylation and sialylation both add acidic groups that no sequence-only calculation can detect.
- Choosing a working buffer at the predicted pI — it is the point of minimum solubility and maximum sensitivity to prediction error.
Related Free Tools From Arb Digital
Use the protein molecular weight calculator for mass and extinction coefficient from the same sequence, and the protein concentration calculator to turn an absorbance into mg/mL. The protein solubility calculator covers ammonium sulfate precipitation, the Henderson-Hasselbalch calculator and pH calculator handle buffer preparation, and the ionic strength calculator quantifies the salt environment. The full free online tools hub lists everything else.
Frequently Asked Questions
The isoelectric point, or pI, is the pH at which a molecule carries no net electrical charge because its positive and negative ionised groups exactly cancel. At that pH it does not migrate in an electric field and its solubility is usually at a minimum.
Each ionisable group is given a fractional charge from the Henderson-Hasselbalch relation at a trial pH, the fractions are summed, and the pH where the total reaches zero is found numerically. Only the two termini and the D, E, C, Y, H, K and R side chains contribute.
Because they use different published pKa sets, which disagree by a few tenths of a unit. Switching sets on this page shows the size of that disagreement directly, which is why a calculated pI should be treated as a range rather than a fixed constant.
It is a sequence-based estimate that ignores structure. Buried residues and neighbouring charges shift real pKa values, sometimes by more than a full unit, so a measured pI from isoelectric focusing often differs by half a unit or more.
Below the pI a molecule is net positive and binds a cation exchanger; above the pI it is net negative and binds an anion exchanger. The magnitude indicates binding strength, though surface charge patches mean net charge is a first predictor rather than a guarantee.
Select the custom pKa set and move that terminal value far outside the pH range, such as minus five for a blocked N-terminus or twenty-five for a blocked C-terminus. Its contribution then falls to effectively zero across the whole calculation.
Yes. Cysteines paired in a disulfide are no longer ionisable, so a protein with several disulfides has fewer acidic groups than its sequence implies. In a reducing buffer they become free again and the sequence-based count is correct.
Because net charge is what keeps molecules electrostatically repelling one another. Remove it and they associate more readily, which is the basis of isoelectric precipitation and also the reason formulation buffers are deliberately chosen away from the pI.
This calculator is provided for education and general reference. It describes how an isoelectric point is computed from sequence and is not laboratory, analytical or safety guidance; follow the methods and risk assessments issued by your own institution.
