predict_counts: Predict counts of one protein and one haplotype

In richelbilderbeek/bbbq: The Bilderbeek Bianchi and Bogaart Question

Predict counts of one protein and one haplotype

Description

Predict the number of binders and the number of binders that overlap with at least one amino acid with a TMH for one protein.

Usage

predict_counts(
  protein_sequence,
  haplotype,
  peptide_length,
  percentile,
  verbose = FALSE,
  ic50_prediction_tool
)

Arguments

protein_sequence	a protein sequence
haplotype	one MHC-I or MHC-II haplotype
peptide_length	length of the peptide in amino acids
percentile	how low the IC50 must be for the protein to be considered a binder. For example, 0.02 denotes that the protein must have an IC50 in the lowest 2 percent range. The default value is returned by get_ic50_percentile_binder.
verbose	set to TRUE for more output
ic50_prediction_tool	tool to predict the IC50 from a peptide. Possible values are: mhcnuggetsr mhcnuggetsr, which uses MHCnuggets mhcnuggetsr netmhc2pan, which uses NetMHC2pam EpitopePrediction uses EpitopePrediction

Details

Use predict_counts_per_proteome to predict these counts for multiple proteins.

Value

a tibble with

1. n_binders number of epitopes that bind

2. n_binders_tmh number of epitopes that bind and have one amino acid overlapping with a TMH

3. n_spots number of spots for the n-mer

4. n_spots_tmh number of spots that have one amino acid overlapping with a TMH

The tibble will have one row (which matches the number of proteins).

Author(s)

Richèl J.C. Bilderbeek

Examples

if (pureseqtmr::is_pureseqtm_installed()
) {
  protein_sequence <- "MYSFVSEETGTLIVNSVLLFLAFV"

  predict_counts(
    protein_sequence = protein_sequence,
    haplotype = get_mhc1_haplotypes()[1],
    peptide_length = 9,
    percentile = 0.123,
    ic50_prediction_tool = "EpitopePrediction"
  )
}

richelbilderbeek/bbbq documentation built on July 27, 2023, 2:15 a.m.