In the past, I found it quite tedious to manually specify Relion's scratch directory every time I ran a protocol. Additionally, when running multiple Relion protocols simultaneously, each scratch directory had to be carefully chosen to be unique. Otherwise, concurrent executions would conflict with each other.
Luckily, the SCIPION_SCRATCH configuration variable automatizes both of these issues. In this PR, I am adding the functionality to use this Scipion feature within Relion protocols.
This will simplify running jobs locally. I still have to use the "custom" option for the cluster because the scratch is only defined at execution time.
In the past, I found it quite tedious to manually specify Relion's scratch directory every time I ran a protocol. Additionally, when running multiple Relion protocols simultaneously, each scratch directory had to be carefully chosen to be unique. Otherwise, concurrent executions would conflict with each other.
Luckily, the
SCIPION_SCRATCH
configuration variable automatizes both of these issues. In this PR, I am adding the functionality to use this Scipion feature within Relion protocols.