Understanding scFASTCORMICS

[cjrodriguez98]

Hello!

I am very interested in learning how to correctly use scFASTCORMICS. I managed to run the example Keratinocytes2024. But I am not understanding the output.

From my understanding of the paper I was expecting a reconstruction of each cluster (to see the metabolism of each cell type) and how are they all connected in an enviroment (the comunity).

My goal is to be able to pin point the metabolic profile of each cell type (cluster) and maybe see if we see changes when taking into account the community.

But from now I only see a multi_cell_population_model and ExpandedInputModel.

Thanks,

Carlos

[thomassauter]

Hi Carlos,
did you see driverAnalysisMP_TS.m?
This script is showing a bit how the analysis of the reconstructed models can be started.
Best,
Thomas

[cjrodriguez98]

Yes! I was able to generate some plots.

But I was interested in extracting the contextualized GEMs from the comunity model, is that possible?

[thomassauter]

This is how you could extract the contextualized GEMs in two variants:

clearvars -except solverOK, clc, close all force

load('Results09-Jul-2024.mat')
Results_keep=Results
Results_keep.multi_cell_population=Results_keep.multi_cell_population_model;

model=Results_keep(1).multi_cell_population_model

modelToExtract=1 %which cluster to extract

% Extract individual GEM
% This will only include the reactions of the desired cluster,
% but the resulting model will not be consistent.
% For consistency reactions of the other clusters will be necessary.
% See below.
idx = strfind(model.rxns, ['_' num2str(modelToExtract)]);
idxToKeep = find(not(cellfun('isempty', idx)));
numel(idxToKeep)
modelInd = removeRxns(model,model.rxns(setdiff(1:numel(model.rxns),idxToKeep)));
modelInd.description='Reactions Cluster 1'

% Extract consistent individual GEM
% This will include some reactions of other submodels to ensure consistency
A = fastcore_4_rfastcormics(idxToKeep, model, 1e-4, []);
modelIndCons = removeRxns(model,model.rxns(setdiff(1:numel(model.rxns),A)));
modelIndCons.description='Model Cluster 1'

A2 = fastcc_4_rfastcormics(modelIndCons, 1e-4, 1);
numel(A2)