Evaluation

[tluquez]

Hello! :)
There are two issues and 1 question:

1. Question: In the line 125, I am trying to replace all unauthorised characters in the reactions and update the reaction column of each file:
   temp_var <- mget(setdiff(ls(pattern = "react"), lsf.str())) sapply(temp_var[[i]][, 3], function(i) { assign(temp_var[[i]][, 3], mgsub( c("-->", '<==>', '->', "\\(|\\)"), c("=>", '<=>', "=>", ""), temp_var[[i]][, 3] ), envir = .GlobalEnv) })
   However, even though the code successfully replace the desired characters, the assign() section is not writing the new column to the file in the workspace. How can I accomplish the above mentioned goal?

2. Issue: When running the gapfill() function from g2f there is an error raised, which stands:
   1050 Orphan reactants found Error in reactionList[validateSyntax(reactionList)] : invalid subscript type 'list'
   So I checked where does this list subscript come from:
   class(validateSyntax(stm_react$equation)) "logical"
   So now I don't know why is this error popping up.

3. Issue: When executing the gap fill from cobrapy (line 205), there is this error:
   Infeasible: gapfilling optimization failed (infeasible)
   And, even after look at it in the web there is no useful option to try.

Thank you in advance for your time.

Kind regards.

[dosorio]

library(sybilSBML)
library(minval)
library(g2f)

Solving the question 1

model <- readSBMLmod("MODEL1507180017.xml")
modelS <- as.matrix(model@S)
rownames(modelS) <- model@met_name
modelR <- minval:::rearmReactions(S = modelS, reversible = model@react_rev)
modelR <- gsub("-Ureidoglycolate", "Ureidoglycolate", modelR)
orphanMetabolites(modelR)

Solving the issue 1

selectedR <- sample(seq_along(modelR), size = 100, replace = FALSE)
query <- modelR[selectedR]
#gapFill(reactionList = query, reference = modelR)

Subset Model to generate new XML

filteredModel <- rmReact(model = model, react = model@react_id[!seq_along(modelR) %in% selectedR])
writeSBML(morg = filteredModel,filename = "filteredModel.xml")

In Python

import cobra
import libsbml

filteredModel = cobra.io.read_sbml_model('filteredModel.xml')
model = cobra.io.read_sbml_model('MODEL1507180017.xml')
gapfill(filteredModel, model, demand_reactions=False, iterations=1)

[tluquez]

The question

Hi Dani, thanks for your help. May be I did not express myself clear enough with the question. I understand your code, and thanks at all for the use of the '@' operator, I had not realised that it can be used to extract data from .modelorg objects. I can update the syntax of the reactions' formulae with this code:

stm_react$equation <- mgsub(c("-->", '<==>', '->', "\\(|\\)"), c("=>", '<=>', "=>", ""), stm_react$equation) 

But, inasmuch as I am planning to supplement several especies_react files, I want to do that assignment operation sequentially, like this:

temp_var <- mget(setdiff(ls(pattern = "react"), lsf.str()))  # Retrieves a list with all the species_react.csv files 

 sapply(temp_var[[i]][, 3], function(i) {
  assign(temp_var[[i]][, 3], mgsub(
    c("-->", '<==>', '->', "\\(|\\)"),
    c("=>", '<=>', "=>", ""),
    temp_var[[i]][, 3]
  ), envir = .GlobalEnv)
})

However it just does not update the object especie_react$equation in the global environment. So, basically, what I am trying to do is update an object in the global environment using the loop, however it is not updated to the global environment. Do you have any idea of how to do that? I have been searching in the web but nothing seems to work.

First issue

I still get the same error. Did you were able to recreate the error? The data files can be downloaded with the first chunk in the warming up section (line 97).

Second issue

I am still working on it. As soon as I have some update I will be telling you.

[tluquez]

@pot53