AlphaFold2 — structure prediction from sequence

Paste a FASTA (monomer or multimer), get a predicted structure with pLDDT, PAE, and pTM/ipTM. ~5-10 min per run.

What it is for

Pick AF2 when you need the gold-standard structure prediction with calibrated pLDDT + PAE. For faster single-sequence folds use ESMFold (D4); for affinity-aware folds use Boltz-2 (D6).

AlphaFold2 (Jumper et al., Nature 2021) packaged via ColabFold (Mirdita et al., Nature Methods 2022). Standard MSA-backed structure prediction with calibrated pLDDT and PAE, monomer or multimer.

When it fits:

  • You need the gold-standard fold with full MSA + templates and calibrated confidence.
  • Your target is monomeric or a small multimer (2–4 chains).
  • You can wait ~5–10 min per run for MMseqs2 MSA fetch + 3 recycles.

Inputs

You will need:

  • Single-letter FASTA sequence(s). Multimers separated by : or pasted as multi-record FASTA.
  • A stable target topology — AF2 underperforms on intrinsically disordered or flexible regions.

Each run uses a preset that sets the scale and scope:

Standalone — your FASTA
Paste or upload FASTA (single chain or multimer). ColabFold MMseqs2 MSA + AF2. Up to 1500 AA total across chains. ~5-10 min on A100-80GB.
Batch — many fold targets
Fold many independent targets in one job (up to 50 records). Each record can be a monomer or a multimer (use ``:`` to separate chains inside a record). Per-design results stream into the job page as folds complete. Slowest of the structure-prediction tools — expect ~5-10 min per fold.

Parameters you set on the form:

Sequence
Paste FASTA. Use : as a chain separator for multimers (e.g. SEQ_A:SEQ_B).
Recycles
Number of model recycles. 3 is the AF2 default; lower is faster but trades a small amount of accuracy.

Typical runtime:

standalone
5–10 min

How to read the results

Predicted PDB with per-residue pLDDT, pairwise PAE, and pTM / ipTM (multimers). Download PDB or PAE matrix for downstream filtering and analysis.

Where a tool reports them, the scores mean:

ipTM
Predicted confidence in the binder to target interface. Higher is better. Aim above roughly 0.7 on a tractable target.
pLDDT
Per-residue confidence in the predicted fold. Higher means the model is more sure of that part of the structure.
i_pAE and pAE
Predicted alignment error, at the interface (i_pAE) or across the whole structure (pAE). Lower is better.

Try these examples

One-click sample inputs that load straight into the run form. Edit any field before submitting.

Ubiquitin (76 aa)
Tiny monomer benchmark. Quick MSA-backed AF2 fold; useful for confirming the pipeline end-to-end.
Top7 de novo design (93 aa)
The canonical de novo designed protein (Kuhlman et al. 2003). Shows AF2 on a designed fold rather than a natural sequence.

References

Jumper et al., Nature 2021 (AF2); Mirdita et al., Nature Methods 2022 (ColabFold)

Open the AlphaFold2 — structure prediction from sequence form All guides