I am trying to compare iJO1366 and iAF1260(b), so have got downloads for them both from BiGG to avoid ID compatibility problems.
iAF1260 gives me a balanced accuracy with respect to gene essentiality data of 0.67 (accuracy 0.88), which is roughly the same as the model in supplementary material from the original publication.
However, iJO1366 gives balanced accuracy of 0.50 (accuracy 0.81), which is much lower than the model sourced from the publication. This seems to be due to a large number of essential genes that are not predicted by the model to be essential.
I'm not sure exactly what the problem is, nor am I sure that this is the correct place to put the issue, but it was the best place I could find to put it!
I am trying to compare iJO1366 and iAF1260(b), so have got downloads for them both from BiGG to avoid ID compatibility problems.
iAF1260 gives me a balanced accuracy with respect to gene essentiality data of 0.67 (accuracy 0.88), which is roughly the same as the model in supplementary material from the original publication.
However, iJO1366 gives balanced accuracy of 0.50 (accuracy 0.81), which is much lower than the model sourced from the publication. This seems to be due to a large number of essential genes that are not predicted by the model to be essential.
I'm not sure exactly what the problem is, nor am I sure that this is the correct place to put the issue, but it was the best place I could find to put it!