Massimo Andreatta1, Morten Nielsen2. 1. Instituto de Investigaciones Biotecnológicas, Universidad Nacional de San Martín, Buenos Aires, Argentina and. 2. Instituto de Investigaciones Biotecnológicas, Universidad Nacional de San Martín, Buenos Aires, Argentina and Center for Biological Sequence Analysis, Technical University of Denmark, Kgs. Lyngby, Denmark.
Abstract
MOTIVATION: Many biological processes are guided by receptor interactions with linear ligands of variable length. One such receptor is the MHC class I molecule. The length preferences vary depending on the MHC allele, but are generally limited to peptides of length 8-11 amino acids. On this relatively simple system, we developed a sequence alignment method based on artificial neural networks that allows insertions and deletions in the alignment. RESULTS: We show that prediction methods based on alignments that include insertions and deletions have significantly higher performance than methods trained on peptides of single lengths. Also, we illustrate how the location of deletions can aid the interpretation of the modes of binding of the peptide-MHC, as in the case of long peptides bulging out of the MHC groove or protruding at either terminus. Finally, we demonstrate that the method can learn the length profile of different MHC molecules, and quantified the reduction of the experimental effort required to identify potential epitopes using our prediction algorithm. AVAILABILITY AND IMPLEMENTATION: The NetMHC-4.0 method for the prediction of peptide-MHC class I binding affinity using gapped sequence alignment is publicly available at: http://www.cbs.dtu.dk/services/NetMHC-4.0.
MOTIVATION: Many biological processes are guided by receptor interactions with linear ligands of variable length. One such receptor is the MHC class I molecule. The length preferences vary depending on the MHC allele, but are generally limited to peptides of length 8-11 amino acids. On this relatively simple system, we developed a sequence alignment method based on artificial neural networks that allows insertions and deletions in the alignment. RESULTS: We show that prediction methods based on alignments that include insertions and deletions have significantly higher performance than methods trained on peptides of single lengths. Also, we illustrate how the location of deletions can aid the interpretation of the modes of binding of the peptide-MHC, as in the case of long peptides bulging out of the MHC groove or protruding at either terminus. Finally, we demonstrate that the method can learn the length profile of different MHC molecules, and quantified the reduction of the experimental effort required to identify potential epitopes using our prediction algorithm. AVAILABILITY AND IMPLEMENTATION: The NetMHC-4.0 method for the prediction of peptide-MHC class I binding affinity using gapped sequence alignment is publicly available at: http://www.cbs.dtu.dk/services/NetMHC-4.0.
Authors: Siwen Hu-Lieskovan; Srabani Bhaumik; Kavita Dhodapkar; Jean-Charles J B Grivel; Sumati Gupta; Brent A Hanks; Sylvia Janetzki; Thomas O Kleen; Yoshinobu Koguchi; Amanda W Lund; Cristina Maccalli; Yolanda D Mahnke; Ruslan D Novosiadly; Senthamil R Selvan; Tasha Sims; Yingdong Zhao; Holden T Maecker Journal: J Immunother Cancer Date: 2020-12 Impact factor: 13.751
Authors: Christian Garde; Sri H Ramarathinam; Emma C Jappe; Morten Nielsen; Jens V Kringelum; Thomas Trolle; Anthony W Purcell Journal: Immunogenetics Date: 2019-06-10 Impact factor: 2.846
Authors: Saori Sakabe; Brian M Sullivan; Jessica N Hartnett; Refugio Robles-Sikisaka; Karthik Gangavarapu; Beatrice Cubitt; Brian C Ware; Dylan Kotliar; Luis M Branco; Augustine Goba; Mambu Momoh; John Demby Sandi; Lansana Kanneh; Donald S Grant; Robert F Garry; Kristian G Andersen; Juan Carlos de la Torre; Pardis C Sabeti; John S Schieffelin; Michael B A Oldstone Journal: Proc Natl Acad Sci U S A Date: 2018-07-23 Impact factor: 11.205