netmhc2pan

Branch	GitHub Actions
master
develop

R package for NetMHCIIpan [1].

Please note that this page is intended for academic users only. Other users are requested to contact the Software Package Manager at [email protected]

• YouTube video or download video

Install

Use usethis (or devtools) to install the netmhc2pan package:

usethis::install_github("richelbilderbeek/netmhc2pan")

Load the netmhc2pan library:

library(netmhc2pan)

Install NetMHCIIpan to a default folder using netmhc2pan and a download URL. The download URL can be obtained from the NetMHCIIpan website at https://services.healthtech.dtu.dk/service.php?NetMHCIIpan-3.2 and expires after four hours.

install_netmhc2pan("https://services.healthtech.dtu.dk/download/fb4dee93-ddbb-4639-ab21-45fdfd0ed23e")

Installation troubleshooting

URL is expired

Request a new download link

cannot open URL 'https://services.healthtech.dtu.dk/download/[number]/netMHCIIpan-3.2.Linux.tar.gz'

From a fresh download link, you can download the file, yet not install using the install_netmhc2pan function.

In that case, use the install_netmhc2pan_bin_from_file function:

install_netmhc2pan_bin_from_file("netMHCIIpan-3.2.Linux.tar.gz")

This function assumes that the file netMHCIIpan-3.2.Linux.tar.gz is in the working directory.

After this, install like normal:

install_netmhc2pan()

cannot open URL 'https://www.cbs.dtu.dk/services/NetMHCIIpan-3.2/data.Linux.tar.gz'

This happens when the university website is down (yes, this has happened!).

In that case, use the install_netmhc2pan_data_from_file function:

install_netmhc2pan_data_from_file("data.Linux.tar.gz")

This function assumes that the file data.Linux.tar.gz is in the working directory.

After this, install like normal:

install_netmhc2pan()

Usage

We start from a FASTA file with one or more protein sequences:

fasta_filename <- system.file(
  "extdata", "example.fasta", package = "netmhc2pan"
)

This is how to content of such a FASTA file looks like:

message(readLines(fasta_filename), sep = "\n")

>sp|L0T550|RIP2_MYCTU_158_180
AVLWAGVAFLAFLQLTALVLNLL

MHC2 has multiple alleles that bind differently to protein parts. NetMHCIIpan supports more than 5000 alleles:

all_alleles <- get_netmhc2pan_alleles()
testit::assert(length(all_alleles) > 5000)

We simply pick the first two alleles:

alleles <- all_alleles[1:2]

Running the NetMHCIIpan analysis:

bindings <- run_netmhc2pan(
  fasta_filename = fasta_filename,
  alleles = alleles
)

The data frame bindings contains the results:

knitr::kable(bindings)

The strongest binding is between the MHC2 allele DRB1_0101 at the 7th (to and including 21st) amino acid of the sp|L0T550|RIP2_MYCTU_158_180 protein, as it is in the top 6.5% of best binders.

FAQ

Under which operating systems does netmhc2pan work?

netmhc2pan can only work on the set of operating systems NetMHCIIpan works on. Currently, only Linux is supported:

How to solve Can't locate Env.pm in @INC?

Install the Perl Env package with:

/usr/bin/perl -MCPAN -e 'install Env'

See notes at this Issue.

Are there similar packages?

These are the ones I use:

• mhcnuggetsr: predicts IC50 values for MHC-I and MHC-II haplotypes
• epitope-prediction: predicts IC50 values for MHC-I haplotypes

References

• [1] Improved methods for predicting peptide binding affinity to MHC class II molecules. Jensen KK, Andreatta M, Marcatili P, Buus S, Greenbaum JA, Yan Z, Sette A, Peters B, Nielsen M. Immunology. 2018 Jan 6. doi: 10.1111/imm.12889. PubMed: 29315598