Skip to content

zedrian/shkoma

Repository files navigation

shkoma

Protein data analysis.

Requirements

  • Python 3 with installed uniprot, numpy, biopython, rpy2 and pandas packages
  • R with installed Peptides package

Usage example

from shkoma import correlation

# 1. load main experiment data
main_data = correlation.load_main_data_from_csv('experiment-28.csv')

# 2. construct list of proteins using main_data
proteins = correlation.construct_proteins(main_data)
correlation.fill_protein_sequences(proteins)
correlation.save_proteins_to_csv(proteins, 'proteins.csv')

# 2. alternate possibility: load proteins from .csv
proteins = correlation.load_proteins_from_csv('proteins.csv')

# 3. construct list of protein records
protein_records = correlation.construct_protein_records(proteins, main_data)

# 4. fill missed peptide records (for each protein)
correlation.fill_missed_peptide_records(protein_records)

# 5. fill computational protein and peptide parameters (for each protein)
correlation.fill_protein_parameters(protein_records)
correlation.fill_peptide_parameters(protein_records)

# 6. save results to folder (separate file for each protein record)
correlation.save_protein_records_to_folder(protein_records, 'experiment-28')

About

Protein data analysis

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages