pepMeld-Lite

• Simpler workflow to process Roche-NimbleGen arrays.
• This does not complete the additional processing if the phyical coordinates of the array is present.
  • Use pepMeld for better preprocessing if the physical coordinates of the array are exposed

Summary

• This python script will:
  • Open the data file,
    • Take the log base2 of each binding affinity value
    • Subtract the associate control with each value
    • Take the median of the repeated peptide sequences of each array, by MHC type.
  • Open a "meta data" file to
    • rename sequences to a human readable format.
    • Apply a qualitative positive/negative "Threshold" for each MHC-Type based on the distribution.
  • Open the peptide sequence to protein accession number lookup
    • Choose a unique accession number / peptide sequence
    • Join the sequences to the peptide data, for each MHC type.
  • Output a file ready to upload into iedb.org epitope databases.

Running main.py to agregate the data for IEDB.org

Creates IEDB.org ready files from Roche-NimbleGen Peptide Array datasets Arguments

argument	single_character	type	default	help	required
meta_filepath	m	string	nan	filepath of the meta data	t
meta_sep	e	string	\t	filepath of the meta data	f
pep_seq_protein_lookup_path	p	string	nan	Filepath of the protein lookup file	t
lookup_sep	l	string	\t	separator used in the pep_seq_protein_lookup file [ , \t ]	f
in_dir	i	string	nan	Location of the input data files	t
in_data_ext	x	string	Mamu.txt.gz	File names endswith to search for in data files of the in_dir	f
in_data_sep	d	string	\t	separator used in the in_data file(s) [ , \t ]	nan
sequence_lengths	s	string	8,9,10	comma delimited list of lengths to include [ 8,9,10 ]	f
out_dir	o	string	nan	output directory	t
out_prefix	f	string	mhc_ppma	prefix to use in the output files.	f

Example:

python3 ~/github/pepMeld-lite/main.py \
--meta_filepath='/Volumes/mhc_dataset/mhc_meta.tsv' \
--meta_sep='\t' \
--pep_seq_protein_lookup_path=/Volumes/mhc_dataset/SIV_SHIV_PEP_QX12_correspondence_key.txt.gz \
--lookup_sep='\t' \
--in_dir='/Volumes/mhc_dataset/input_data' \
--in_data_ext='Mamu.txt.gz' \
--in_data_sep='\t' \
--sequence_lengths='8,9,10' \
--out_dir='/Volumes/mhc_dataset/output_data' \
--out_prefix='mhc_ppma'

Raw data input data format

Required columns:

Column Name	Details
PROBE_SEQUENCE	Peptide sequence (all caps)
REPL	The corresponding Position of a Replicate (used to match control position)
SEQ_ID	Either MER_REDUNDANT or _

Meta data input data format

Required columns:

Column Name	Details
VENDOR_NAME	Data column name from the vendor provided data table
MHC Allele Name	This is the MHC Allele Name that will be uploaded to IEDB.org
CONTROL_SUBTRACT	This is the column to subtract the values by Peptide_Sequence & Replicate position
EXCLUDE	To exclude the column from analysis
PASSING_THRESHOLD	Control Subtracted threshold that is considered passing

Peptide Sequence to Protein Accession format

Required columns:

Column Name	Details
PEPTIDE_SEQUENCE	Peptide sequence (all caps)
ORIGINAL_SEQ_ID	Associated Virus Accession Number _ Protein Accession Number