scripts includes all the scripts for analyzing the mass spectrometry data, and doing the Monte Carlo estimates, and making all the figures.

These scripts assume the following file structure

/scripts/
/data/
     /tara_ocean_smags/
     /antiox_gene_name_lists/
     /protein_expression_data/
                             /nunn_data/
                             /phytoplankton_genomes/
                                                   /frag_genome/
                                                   /phaeo_genome/
                             /mass_spec_data/
                                            /frag_data/
                                                      /mzml_converted/
                                                      /fdr_idxml/
                                            /phaeo_data_3/
                                                      /mzml_converted/
                                                      /fdr_idxml/

Data

For the frag_data, I got data from PRIDE project with the ID: PXD007098. I specifically used these data files:

D05_1.raw
D120_1.raw
D1_2.raw
L120_3.raw
T0_3.raw

For the phaeo_data_3, I got data from the PRIDE project with the ID: PXD014877. I used the script download_phaeo_proteome_3.sh for this.

Data for the phytoplankton genomes for protein stoichiometry were downloaded from this awesome paper by Delmont et al (2021) specifically with:

# annotation files
wget https://www.genoscope.cns.fr/tara/localdata/data/SMAGs-v1/SMAGs_v1_EggNog.tar.gz

# protein sequences
wget https://www.genoscope.cns.fr/tara/localdata/data/SMAGs-v1/SMAGs_v1_concat.faa.tar.gz

Other files ('Table_S03_statistics_nr_SMAGs_METdb.xlsx') were manually downloaded from https://www.genoscope.cns.fr/tara/, and the ferritin sequence used is from https://www.uniprot.org/uniprot/B6DMH6.fasta .

Genomic Data Processing

converting_tara_to_single_line_fasta.sh

Converts the large file of SMAGs to a single line fasta file.

tara_oceans_antioxidant.R

Script that gets all the antioxidant protein sequence IDs and makes txt files of lists, for eventually subsetting the large fasta file.

antioxidant_stoich_from_seqs.ipynb

Jupyter notebook that calculates stoichiometric composition of various phytoplankton proteins. The plots for these proteins is in plotting_antiox_stoich.R.

Proteomic Data Processing

Once the raw MS data have been downloaded, they need to be converted to mzML. They were converted with convert_raw_to_mzml.sh, frag_converting_file.sh, and phaeo_converting_file.sh.

Then they are searched using the corresponding genomes with database-searching-openms.sh, database-searching-frag.sh, and database-searching-phaeo.sh. Before they are searched the genomes are appended with a contaminant database (CRAP), with adding_crap_to_genomes.sh.

Peptides are quantified using the FeatureFinderIdentification method (Weisser et al 2017) with the script feature-finder-general.sh, feature-finder-phyto-proteomes.sh, and feature-finder-phyto-proteomes-phaeo.sh.

The output of this is then converted to csv with protein-quant-openms.sh, protein-quant-frag.sh, and protein-quant-phaeo.sh.

Monte Carlo Estimation

We first need to determine some of the parameters, specifically the distribution of fold-changes for protein expression. Three scripts separately analyze the fold-change distribution from three different data sets: nunn_protein_expression_fold_change.R (Nunn et al 2013, PLoS ONE); li_dist_vals.R (Li et al 2014, Cell); schmidt_dist_vals.R (Schmidt et al 2015, Nature Biotechnology). Each of these outputs a parameter file with the appropriate parameters of the fitted log-normal distribution.

Monte Carlo method is done using monte_carlo_antioxidants_both.R.