Appointing sample attributes as experimental factor

[floradanna]

In ISA-JSON compliant experiment, sample attribute(s) can be experimental factor(s). As researchers, I want to indicate which sample attribute(s) is the experimental factor in study and assay, so that this information can be processed by analysis pipeline

[kdp-cloud]

@floradanna Could you specify what the difference is with the parameter_value tag we have in SEEK? Can you also share if there are constraints to using this tag like:

• Number
• Naming
• Level of the Sample Type it can be used ...

[floradanna]

As defined in ISA Model (https://isa-specs.readthedocs.io/en/latest/isamodel.html#study), "A factor corresponds to an independent variable manipulated by the experimentalist with the intention to affect biological systems in a way that can be measured by an assay."

A factor:

• can be any existing variable in a sample type: a specific parameter value or a characteristics.
• can be specified at study and/or assay level
• it should be possible to indicate 0 or more experimental factors at study and/or assay level

[kdp-cloud]

can be any existing variable in a sample type: a specific parameter value or a characteristics.

If I understand correctly:

• It's an extra type of annotation for a sample attribute, and not just an extra ISA tag.
• Only a sample attribute with ISA tag parameter_value or characteristic can be an experimental factor
• Multiple sample attributes in a sample type can be labeled as 'experimental factor'

Also, I see 'Factor Name' and 'Factor type' in the specifications. Do you want be able to specify these? Or is it enough if you simply specify it's an experimental factor?

[floradanna]

• correct
• correct
• correct

it is enough to be able to simply specify that "this attribute" is an experimental factor