BarcodeBabel
Python script for generation of peptide barcodes for electrospray LC-MS/MS
Overview
BarcodeBabel is an algorithm to generate libraries of peptide barcodes with user-defined ranges of features for optimal detectability by nanoscale liquid chromatography tandem mass spectrometry (LC-MS/MS). User-selected rules include lack of presence in a user-provided proteome, m/z range, presence of homopolymers, hydrophobicity range, enzyme cleavage sites, residue frequencies, and library size. Sample human proteome with common contaminants is provided.
Manifest
NoteBooks: BarcodeBabel.ipynb - Jupiter notebook with BarcodeBabel script
BarcodeBabel: barcodes.py - BarcodeBabel script
Data: Sample human proteome and common contaminants, and example 48 test barcodes
Dependencies
matplotlib
numpy
scipy
sci-kitlearn
pandas
biopython
tqdm (progress bars)
xlrd
lxml
pyteomics (preferred way to install is pip Python package manager)
Usage
- Select enzyme cleavage sites: def append_enzyme_cleavage_sites
• Defaults: GRA on both ends, for trypsin cleavage
• User may choose to specify separately for beginning and end
• If the default is changed, edit enzyme function def trypsinize(peptide) to reflect cleavage
- User defines frequencies of amino acids:
• Defaults: Frequencies will be 0 for: K, R, H, C, M, P, I
• Rationale: K, R, H may cause trypsin cleavage within sequences. M, C may be oxidized. P skews fragmentation. I and L cannot be distinguished. Q and N easily deamidated.
- User defines m/z range at user-defined pH
• Defaults: pH = 3 (for standard LC-MS at low pH), m/z range: 550-850
• Peptide properties calculated for the enzyme-cleaved barcode
- User defines maximum length of allowed repeats
• Default: 2 repeats
- User defines hydrophobicity range
• Based on the maximally hydrophobic 5-residue in candidate peptides
• Default: -0.5 to 2.5, determined by examining library of tryptic peptides with favorable range of retention
• Hydrophobocity is calculated for the enzyme-cleaved barcode
- User selects reference proteome, to disallow overlap:
• Provided proteome: human reference proteome and common contaminants
- User defines library size:
• Default: 2000
• Generates .csv file
- Default: BarcodeBabel also reports 5'-3' DNA sequence for synthesis, 3'-5' reverse complement, no overhangs, out-of-frame start / stop codons in DNA sequence
• Included in the generated csv file
Authors
Elliot Eton | elliot.eton@gmail.com
Josh Fass | josh.fass@choderalab.org
Nicole McNeer | mcneern@mskcc.org
Alex Kentsis | kentsisresearchgroup@gmail.com