marcottelab/MS_grouped_lookup

Name: MS_grouped_lookup

Owner: The Marcotte Lab

Description: null

Created: 2016-12-07 18:50:00.0

Updated: 2017-08-24 16:03:44.0

Pushed: 2017-08-24 19:38:08.0

Homepage: null

Size: 14108

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

MS_grouped_lookup

This is a folder for looking up peptides from fractionation-MS experiments using custom grouping for uniqueness

One part of a larger scheme that goes:

1. Convert .RAW thermo files in /MS/submit to mzXML in /MS/processed using local Windows MSConvert
2. Run MSblender on TACC (ls5) $scratch using the https://github.com/marcottelab/run_msblender setup (using branch https://github.com/marcottelab/MSblender/tree/msblender_restructure)
   • Download output ot /MS/processed
3. Format proteome for peptide lookup
4. Lookup peptides

Analysis steps

Define grouping of peptides

1. Run eggnog-mapper to assign proteins to groups and format output

Running hmmer straight on a full protein will take several days to process

$ nohup python /project/eggnog-mapper-0.99.2/emapper.py -i /project/cmcwhite/orthology_proteomics/proteomes/human/uniprot-reviewed%3Ayes+AND+proteome%3Aup000005640.fasta --output human_hmmer_euNOG -d euNOG --override --scratch_dir /project/cmcwhite/orthology_proteomics/proteomes/human/ -m hmmer --output_dir /project/cmcwhite/orthology_proteomics/eggnog_mapper  &> /project/cmcwhite/orthology_proteomics/logs/nohup_human_euNOG.txt &


Instead, break up the proteome into chunks and process in parallel

Make a file with the species ID of one or more species

spec_list.txt
    human
    mouse

This script currently assumes a proteomes directory structure of 
   /proteomes/[speciesID]/working_proteome/[single fasta]

Break up a proteome and create a list of commands for each chunk
$ bash scripts/create_emapper_commands.sh spec_list.txt /project/eggnog-mapper-0.99.2/emapper.py euNOG hmmer        
Run each command in parallel
$ cat human_euNOG_hmmer_COMMANDS.txt | parallel -j10

Combine the outputs
$ cat proteomes/human/output/*emapper.annotations > eggnog_mapper/human_hmmer_euNOG.emapper.annotations


The output from the eggnog mapper need to be formatted 
$ format_emapper_output.R -f human_hmmer_euNOG.emapper.annotations -o human_hmmer_euNOG.mapping -s hmmer -l euNOG

creates file with format:
ProteinID   ID

2. Do an artificial trypsin digest on a proteome

$ python scripts/trypsin.py –input proteomes/human/uniprot-proteome%3AUP000005640.fasta –output proteomes/human/uniprot-proteome%3AUP000005640_peptides.csv –miss 2

3. Get protein-unique peptides

Identify proteins from peptides that are unique to single proteins

$ python scripts/define_grouping.py --spec human --grouping_type protein  --peptides proteomes/human/working_proteome/uniprot-proteome_human_reviewed_peptides.csv --output_dir proteomes/human/working_proteome/

4. Get group-unique peptides

Identify groups of proteins from peptides that are unique to the proteins in a group

$ python scripts/define_grouping.py --spec human --grouping_type euNOG --grouping eggnog_mapper/human_hmmer_euk.mapping  --peptides proteomes/human/working_proteome/uniprot-proteome_human_reviewed_peptides.csv --output_dir proteomes/human/working_proteome/

Identify proteins in an experiment

1. Consolidate identified peptides from multiple experiments into a single file

$ bash scripts/consolidate_MSblender_output.sh /MS/processed/Fusion_data/ExperimentA/output ExperimentA elutions/

/MS/processed/Fusion_data/ExperimentA/output

  fraction1_pep_count_FDR001
     ACDER 1
     ETIAJR 2

  fraction2_pep_count_FDR001
     GFEAR 1
     AYTQWER 3

—>

ExperimentA_elution.csv

    ExperimentA,fraction1,ACDER,1
    ExperimentA,fraction1,ETIAJR,2
    ExperimentA,fraction2,GFEAR,1
    ExperimentA,fraction2,AYTQWER,3

These formatted files are stored in the elutions/ folder

2. Lookup peptides by protein

Do the look up $ python scripts/get_elution_profiles.py human protein ExperimentA elutions/ExperimentA_elution.csv proteomes/human/working_proteome/unique_peptides_human_protein.csv proteomes/contam/contam_benzo_peptides.csv

$ python scripts/get_wideform_prot.py identified_elutions/human/ExperimentA_elution_human_protein.csv

Transform columns to a wide table. ex.

   "tidy elution format"
   ExperimentA,fraction1,protein1,10
   ExperimentA,fraction2,protein1,30
   ExperimentA,fraction1,protein2,3
   ExperimentA,fraction2,protein2,2

   --> 
   "wide elution format"

   ID,fraction1,fraction1
   protein1,10,30
   protein2,3,2

3. Lookup peptides according to a grouping of proteins

$ python scripts/get_elution_profiles.py human euNOG ExperimentA elutions/ExperimentA_elution.csv proteomes/human/working_proteome/unique_peptides_human_euNOG.csv proteomes/contam/contam_benzo_peptides.csv

$ python scripts/get_wideform_group.py identified_elutions/human/ExperimentA_elution_human_euNOG.csv eggnog_mapper/human_hmmer_euNOG.mapping annotation_files/all_annotations.csv

Going to be removed, not very useful extra format Similar to get_wideform_prot.py, but also creates an alternate format that shows the proteins in a group

ID,proteinIDs,fraction1,fraction2