Files in the data folder#

The data folder of a GECKO 4 project holds the input data of the reconstruction, the kcat source files and the files of the kcat predictors. GECKO (MATLAB) and geckopy (Python) read and write the same files, so a data folder made with one toolbox is used by the other. Build an ecModel describes the layout of a project and Project format 4 the folders outside data.

Databases and gene identifiers#

File

Content

Functions

Page

uniprot.tsv

UniProt data for the organism: protein identifier, gene identifier, EC number, MW and sequence. loadDatabases downloads it when it is missing

loadDatabases (Python: geckopy uniprot-download) writes; makeEcModel, getStandardKcat and calculateFfactor read

Build an ecModel, Choose the ecModel structure

kegg.tsv

KEGG data for the organism: UniProt protein ID, gene, KEGG gene ID, EC number, MW, pathway and sequence. loadDatabases downloads it when it is missing

loadDatabases (Python: geckopy kegg-download) writes; makeEcModel (make_ec_model) reads

Build an ecModel, Choose the kcat sources

uniprotConversion.tsv

Conversion table for models whose gene identifiers match no UniProt field. A header row, then the gene identifier and the UniProt identifier. A gene that is not listed keeps its identifier in geckopy and gets an empty UniProt identifier in MATLAB. Written by the user

ModelAdapter.getUniprotIDsFromTable (Python: get_uniprot_ids_from_table)

Build an ecModel

pseudoRxns.tsv

Optional. Reaction identifiers in the first column and reaction names, for reference, in the second. These reactions lose their gene association in makeEcModel (make_ec_model) and receive no standard kcat in getStandardKcat (assign_standard_kcat). Written by the user

makeEcModel, getStandardKcat

Choose the ecModel structure, Choose the kcat sources

ComplexPortal.json

Enzyme complex information from the Complex Portal

getComplexData (Python: get_complex_data) writes; applyComplexData (apply_complex_data) reads

Choose the ecModel structure

smilesDB.tsv

SMILES for the metabolites of the ecModel, a cache of the PubChem queries

findMetSmiles (Python: find_met_smiles)

Choose the kcat sources

Kcat source files and BRENDA data#

Path

Content

Functions

Page

kcats/<source>.tsv

One kcat source file per source: custom, brenda, ec_tree, homologs, dlkcat and okp_<method>. The format is described in Kcat source files

kcatsFromBrenda, kcatsFromECTree, kcatsFromHomologs, kcatsFromDLKcat and kcatsFromOKP write; loadKcatSource and collectKcats read. custom.tsv is written by hand or by convertCustomKcats (convert_custom_kcats)

Choose the kcat sources

brenda/

Optional. The BRENDA files of the project: kcat.tsv, sa.tsv, mw.tsv, measurements.tsv and the other files of the BRENDA release. Without the folder, the files of the toolbox-assets release are downloaded once

ModelAdapter.getBrendaDBFolder (Python: get_brenda_db_folder); geckopy brenda-refresh writes kcat.tsv, sa.tsv and mw.tsv to the folder, by default data/brenda of the current directory; it does not produce measurements.tsv, sequences.fasta or substrates.tsv, and kcats_from_ec_tree then falls back to the asset

Choose the kcat sources

customKcats.tsv

Curated kcats of a GECKO 3 project (columns proteins, genes, gene_name, kcat, rxns, notes, stoicho). Without kcats/custom.tsv, collectKcats raises collectKcats:legacyCustomKcats

convertCustomKcats (Python: convert_custom_kcats) converts it into kcats/custom.tsv

Set custom kcats

PhylDist.mat

KEGG phylogenetic distances, which geckopy uses to match kcats of other organisms. Downloaded when it is missing. GECKO reads the corresponding file of RAVEN

kcats_from_brenda

Choose the kcat sources

Predictor files#

File

Content

Functions

Page

DLKcat.tsv

Input of a local DLKcat run in Docker and, after the run, its output

kcatsFromDLKcat (Python: kcats_from_dlkcat) with via docker writes; kcatsFromDLKcat with via file reads

Choose the kcat sources

DLKcat_run.txt

The number of reactions that the last DLKcat run covered, when it covered a subset of the reactions

kcatsFromDLKcat (Python: kcats_from_dlkcat)

Choose the kcat sources

DLKcatIgnoreMets.tsv, DLKcatCurrencyMets.tsv

Optional project copies of the metabolites left out of the DLKcat and OKP input and the homolog search (names with SMILES; pairs of currency metabolites). Default: the toolbox files

kcatsFromOKP, kcatsFromDLKcat, kcatsFromHomologs

Choose the kcat sources

OKP.csv

Input of the last OpenKineticsPredictor job

kcatsFromOKP (Python: kcats_from_okp)

Choose the kcat sources

OKP_job_<method>.txt, OKP_output_<method>.csv

The job and the result of one OKP method

kcatsFromOKP (Python: kcats_from_okp)

Choose the kcat sources

OKP_job.txt, OKP_output.csv

The job and the result of a job, written by earlier versions of GECKO. kcatsFromOKP adopts them for its method when no job of that method exists

kcatsFromOKP (Python: kcats_from_okp)

Choose the kcat sources

okpApiKey.txt

The API key of OpenKineticsPredictor, read when the environment variable OKP_API_KEY is not set

kcatsFromOKP, kcatsFromDLKcat (Python: kcats_from_okp, kcats_from_dlkcat)

Choose the kcat sources

Proteomics, flux data and tuning data#

File

Content

Functions

Page

paxDB.tsv

Protein abundance data from PaxDB for the f factor (the mass fraction of proteins accounted for in the ecModel). Header lines start with #, followed by three tab-separated columns: an internal identifier, the gene and the abundance. Without the file, calculateFfactor returns 0.5

calculateFfactor (Python: load_pax_db, calculate_f_factor)

Set the protein pool

proteomics.tsv

Measured protein levels (mg protein/gDCW) from one or more experiments: UniProt identifiers in the first column, protein levels from individual replicates in each subsequent column

loadProtData (Python: load_prot_data)

Constrain enzymes with proteomics

fluxData.tsv

Experimental growth rate, protein content and carbon-source uptake, that constrain the ecModel during proteomics integration and in flux-variability comparisons. Exchange fluxes in mmol/gDCWh; uptake negative, excretion positive. Not required for the initial reconstruction

loadFluxData (Python: load_flux_data), constrainFluxData

Constrain enzymes with proteomics, Analyse an ecModel

evotuneFluxData.tsv, evotuneMaxGrowth.tsv, evotuneZeroExch.tsv

Optional data of the kcat tuning with CMA-ES: flux data, maximum growth rates and exchange reactions with zero flux

loadEvotuneData

Tune kcats against experimental data

Spontaneous reactions#

The file that [spontaneous] file in model_adapter.toml names is a TSV with a header row. It lists reaction identifiers in the first column and, optionally, a second column in which 1 or true marks the spontaneous reactions. The path is relative to the project folder and the file name is free. See Model adapter parameters.

See also#