PANDORA v2.0: Benchmarking peptide-MHC II models and software improvements.

Area of Science:

• Immunology
• Structural Biology
• Computational Biology

Background:

• T-cell receptor (TCR) recognition of peptide:MHC (pMHC) complexes is vital for distinguishing self from non-self.
• Experimental structures of pMHC complexes are limited, necessitating computational modeling approaches.
• pMHC class II (pMHC-II) complexes are key players in the cancer immune response.

Purpose of the Study:

• To present an updated version of the PANDORA software for modeling 3D pMHC-II structures.
• To systematically evaluate the performance of PANDORA in modeling pMHC-II complexes.
• To provide an efficient and reliable computational tool for pMHC-II structure prediction.

Main Methods:

• Benchmarking PANDORA against 136 experimentally determined pMHC-II structures.
• Utilizing a pipeline that restrains peptide residues within the MHC-II binding groove.
• Evaluating model accuracy using Ligand-Root Mean Squared Deviation (L-RMSD).

Main Results:

• PANDORA achieved a median backbone L-RMSD of 0.42 Å on the binding core and 0.88 Å on the whole peptide.
• The software was enhanced to function as a pan-allele framework with improved usability.
• The computational efficiency allows for rapid generation of 3D models.

Conclusions:

• The updated PANDORA software provides accurate and efficient 3D modeling of pMHC-II complexes.
• These models can supplement experimental data and serve as starting points for further computational analyses.
• PANDORA facilitates research in cancer immunology by enabling the identification of MHC-binding peptides.