Skip to main content

Table 2 Identification of peptide binding cores.

From: NN-align. An artificial neural network-based alignment algorithm for MHC class II peptide binding prediction

Allele

Peptide

Core

TEPITOPE

SMM-align

NetMHCIIPan

NN-W-P1

NN-W

DRB1*0101

AGFKGEQGPKGEPG

FKGEQGPKG

FKGEQGPKG

FKGEQGPKG

FKGEQGPKG

FKGEQGPKG

FKGEQGPKG

DRB1*0101

GELIGILNAAKVPAD

IGILNAAKV

IGILNAAKV

IGILNAAKV

IGILNAAKV

IGILNAAKV

IGILNAAKV

DRB1*0101

PEVIPMFSALSEGATP

VIPMFSALS

VIPMFSALS

VIPMFSALS

VIPMFSALS

VIPMFSALS

VIPMFSALS

DRB1*0101

PKYVKQNTLKLAT

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

DRB1*0101

VGSDWRFLRGYHQYA

WRFLRGYHQ

WRFLRGYHQ

WRFLRGYHQ

WRFLRGYHQ

WRFLRGYHQ

WRFLRGYHQ

DRB1*0101

AFVKQNAAALA

FVKQNAAAL

FVKQNAAAL

VKQNAAALA

FVKQNAAAL

FVKQNAAAL

FVKQNAAAL

DRB1*0101

AAYSDQATPLLLSPR

YSDQATPLL

YSDQATPLL

YSDQATPLL

YSDQATPLL

YSDQATPLL

YSDQATPLL

DRB1*0301

PVSKMRMATPLLMQA

MRMATPLLM

MRMATPLLM

MRMATPLLM

MRMATPLLM

MRMATPLLM

VSKMRMATP

DRB1*0401

AYMRADAAAGGA

MRADAAAGG

MRADAAAGG

MRADAAAGG

YMRADAAAG

YMRADAAAG

AYMRADAAA

DRB1*0401

PKYVKQNTLKLAT

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

YVKQNTLKL

KYVKQNTLK

DRB1*1501

ENPVVHFFKNIVTPR

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VVHFFKNIV

DRB1*1501

ENPVVHFFKNIVTPRGGSGGGGG

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VHFFKNIVT

VVHFFKNIV

DRB5*0101

GGVYHFVKKHVHES

YHFVKKHVH

YHFVKKHVH

YHFVKKHVH

YHFVKKHVH

YHFVKKHVH

YHFVKKHVH

DRB5*0101

NPVVHFFKNIVTPRTPPPSQ

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

DRB5*0101

VHFFKNIVTPRTPGG

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

FKNIVTPRT

Correct

  

15/15

14/15

14/15

14/15

10/15

  1. 15 HLA-DR restricted peptides compiled from the protein database [32]. The columns in the table state the HLA restriction, the peptide sequence, the binding core as determined from the protein structure, followed by the binding core as predicted by TEPITOPE, SMM-align, NetMHCIIpan, NN-W-P1 (the NN-based method including data redundancy step-size rescaling and P1-PSSM encoding), and NN-W (the NN-based method including data redundancy step-size rescaling), respectively. Erroneous predictions are underlined.