The thought here is that when choosing to look at a single/few genes it is more helpful to see either the raw or normalized counts for multiple groups (e.g. control, low ferritin, high ferritin) as opposed to a single measure of difference between two groups (still reasonable when looking at many genes within a pathway). Moving this to the Datasets tab will allow for more than two group selections. It also makes sense as it won't be tied to a specific contrast as with pathways.
The thought here is that when choosing to look at a single/few genes it is more helpful to see either the raw or normalized counts for multiple groups (e.g. control, low ferritin, high ferritin) as opposed to a single measure of difference between two groups (still reasonable when looking at many genes within a pathway). Moving this to the Datasets tab will allow for more than two group selections. It also makes sense as it won't be tied to a specific contrast as with pathways.