garnish_slim(dt, slimmer = TRUE)



Input data.table from garnish_affinity output.


Logical. Default is TRUE, set to false to return percentile/ranks of predictions from prediction tools.


A slimmed down data table from garnish_affinity output with minimal information for each peptide prediction. Contains at most the following columns:

  • sample_id

  • external_gene_name

  • protein_change

  • nmer

  • MHC

  • class: MHC class I or II

  • mhcflurry_prediction

  • mhcflurry_prediction_percentile

  • mhcnuggets_pred_lstm

  • mhcnuggets_pred_gru

  • affinity(nM)_netMHC

  • affinity(nM)_netMHCpan

  • \ \ Ensemble_score DAI: Differential agretopicity of variant to corresponding wild-type, see garnish_summary. min_DAI: The most conservative DAI value based on a global alignment to the wild-type proteome. iedb_score dissimilarity cl_proportion: clonal cell fraction composed by the variant antigen.garnish_class: epitope class, one of "Classic", "Alternative", "IEDB high", "high dissimilarity", see ?garnish_summary for thresholds. See garnish_affinity for more column descriptions. This function reduces garnish_affinity output to a more manageable table, dropping columns with less critical information and removing wild-type peptide rows. For sample-level summary statistics, see garnish_summary, and for highest priority epitopes per sample, see garnish_antigens. garnish_affinitygarnish_summarygarnish_antigens