Skip to content

Possible gene ID–gene symbol mismatch in ftINIT-reconstructed models #669

Description

@mfcesur

Hi,

I generated cell-specific metabolic models with ftINIT by setting "removeGenes = true". The reconstructed models appear to contain incorrect mappings between Ensembl gene IDs (model.genes) and gene symbols (model.geneShortNames).

To investigate this, I compared the reconstructed adipocyte model against both the original Human2 model and the Ensembl (release 109) annotations. Human2 includes the correct gene symbols, whereas the adipocyte model shows what looks like a systematic shift of gene symbols after a certain point in the gene list. Please find the attached table.

I did not have the opportunity to investigate this in detail, but I suspect it may be related to the way gene-associated fields are updated after gene removal. This is because, removeLowScoreGenes rebuilds the gene list using getGenesFromGrRules. Since getGenesFromGrRules reconstructs the gene list using the unique function (without the 'stable' option), I was wondering whether this could change the original order of "model.genes":

genes = unique(transpose([rxnGenes{nonEmpty}]));

If so, geneShortNames may potentially become misaligned. I would appreciate any thoughts on whether this could explain the behavior I observed, or whether there might be another step in the ftINIT pipeline where this mismatch could be introduced.

Thank you very much for your time.

Kind regards,
Müberra

geneSymbol_comparisons.xlsx

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions