MHCII3D is a tool for predicting MHC-II binding peptides, based on structural scaffolds of MHC-II-peptide complexes and statistical scoring functions (SSFs).

MHCII3D provides two modes:

1. the scoring of peptides, and
2. the scan of a protein sequence for binding regions

Scoring a list of peptides

For all prediction modes, first a HLA-DRB allele needs to be selected.

Then  a list of peptides needs to be entered, either on peptide per line or in multi-fasta format.

Output

Column	Meaning
Allele	Beta chain allele name
Peptide	Input sequence
Binding Sequence	Binding Sequence in the 15-mer peptide scaffold. ‘-‘ characters indicate that the peptide is shorter than the scaffold or shifted to the left or to the right. For clarity, the 9-mer core is indicated here in magenta.
Binding Score	SSF based score, the more negative the better
Binding Rank	normalized rank of the score in a random background ranging from zero (optimal) to one (worse than lowest scoring random peptide).
IC50 predicted	Rank converted to a IC50 value by non-linear fit

Scan a protein sequence for binding regions

Here, only a single sequence may be given, either as plain sequence or in fasta format. The input sequence is cut into overlapping fragments and each fragment is subsequently scored.

The meaning of the respective columns is described in the table above.

In addition, three plots are given, showing the average of the binding score, rank and IC₅₀ per residue. Watch out for minima in these plots in order to locate binding regions (potential T-cell epitopes).