Scrutinizing MHC-I Binding Peptides and Their Limits of Variation

Koch, Christian P.; Perna, Anna Maria; Pillong, Max; Todoroff, Nickolay; Wrede, Paul; Folkers, Gerd; Hiss, Jan A.; Schneider, Gisbert

doi:10.1371/journal.pcbi.1003088

Cited by 36 publications

(38 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Considerable efforts have been dedicated to the development of accurate methods for the prediction of peptide binding to MHC molecules, applying many different approaches including similarity matrices (Kim et al, 2009), linear regression (Wang et al, 2015) and artificial neural networks (Hoof et al, 2009;Koch et al, 2013;Kuksa et al, 2015), among others. Of these methods, NetMHC (Nielsen et al, 2003) has been shown in several benchmark studies to be a state-of-the-art predictor for peptide-MHC binding affinity (Lundegaard et al, 2008;Peters et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

Gapped sequence alignment using artificial neural networks: application to the MHC class I system

2015

View full text Add to dashboard Cite

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:

show abstract

Section: Introductionmentioning

confidence: 99%

Gapped sequence alignment using artificial neural networks: application to the MHC class I system

2015

View full text Add to dashboard Cite

show abstract

“…Due to the high sensitivity of modern fluorimeters, DSF is well suited to low volume, rapid measurements on multiple samples. DSF has found numerous uses in characterizing protein stability and ligand binding, and has been recently used to assess the binding of peptides to class I MHC proteins (Gras et al, 2012; Koch et al, 2013a; Koch et al, 2013b; Hassan et al, 2015), as well as the impact of peptides and other molecules binding to class II proteins (Günther et al, 2010; Rupp et al, 2011; Clayton et al, 2014), including HLA-DM (Álvaro-Benito et al, 2015). Although early applications of DSF required customized fluorimeters, the technique is easily adaptable to fluorescence-equipped RT-PCR instruments common to many molecular biology laboratories.…”

Section: Introductionmentioning

confidence: 99%

Differential scanning fluorimetry based assessments of the thermal and kinetic stability of peptide–MHC complexes

Hellman

Yin

Wang

et al. 2016

Journal of Immunological Methods

View full text Add to dashboard Cite

Measurements of thermal stability by circular dichroism (CD) spectroscopy have been widely used to assess the binding of peptides to MHC proteins, particularly within the structural immunology community. Although thermal stability assays offer advantages over other approaches such as IC50 measurements, CD-based stability measurements are hindered by large sample requirements and low throughput. Here we demonstrate that an alternative approach based on differential scanning fluorimetry (DSF) yields results comparable to those based on CD for both class I and class II complexes. As they require much less sample, DSF-based measurements reduce demands on protein production strategies and are amenable for high throughput studies. DSF can thus not only replace CD as a means to assess peptide/MHC thermal stability, but can complement other peptide-MHC binding assays used in screening, epitope discovery, and vaccine design. Due to the physical process probed, DSF can also uncover complexities not observed with other techniques. Lastly, we show that DSF can also be used to assess peptide/MHC kinetic stability, allowing a single experimental setup to probe both binding equilibria and kinetics.

show abstract

“…Luo et al proposed both a colored and non-colored bipartite networks [Luo et al, 2016]. Shallow and high-order artificial neural networks were proposed from various labs [Hoof et al, 2009, Koch et al, 2013, Kuksa, 2015, Nielsen et al, 2003. Of these approaches, NetMHC/NetMHCpan have been shown to achieve state-of-the-art for MHCpeptide binding prediction [Nielsen et al, 2016, Trolle et al, 2015.…”

Section: Introductionmentioning

confidence: 99%

HLA class I binding prediction via convolutional neural networks

Vang

Xie

2017

Bioinformatics

View full text Add to dashboard Cite

Many biological processes are governed by protein-ligand interactions. One such example is the recognition of self and nonself cells by the immune system. This immune response process is regulated by the major histocompatibility complex (MHC) protein which is encoded by the human leukocyte antigen (HLA) complex. Understanding the binding potential between MHC and peptides can lead to the design of more potent, peptide-based vaccines and immunotherapies for infectious autoimmune diseases.We apply machine learning techniques from the natural language processing (NLP) domain to address the task of MHC-peptide binding prediction. More specifically, we introduce a new distributed representation of amino acids, name HLA-Vec, that can be used for a variety of downstream proteomic machine learning tasks. We then propose a deep convolutional neural network architecture, name HLA-CNN, for the task of HLA class I-peptide binding prediction. Experimental results show combining the new distributed representation with our HLA-CNN architecture acheives state-of-the-art results in the majority of the latest two Immune Epitope Database (IEDB) weekly automated benchmark datasets. We further apply our model to predict binding on the human genome and identify 15 genes with potential for self binding. Codes are available at https://github.com/ucicbcl/HLA-bind.

show abstract

Scrutinizing MHC-I Binding Peptides and Their Limits of Variation

Cited by 36 publications

References 57 publications

Gapped sequence alignment using artificial neural networks: application to the MHC class I system

Gapped sequence alignment using artificial neural networks: application to the MHC class I system

Differential scanning fluorimetry based assessments of the thermal and kinetic stability of peptide–MHC complexes

HLA class I binding prediction via convolutional neural networks

Contact Info

Product

Resources

About