In back-to-back runs of Comet, the "sprank" value has been observed to change. This needs to be addressed such that output from the same run are exactly the same.
This is due to threading and how results are stored. Peptides are only stored if they score greater than current lowest score. Depending on which threads finish first, the peptide list can change between runs, thus affective the sp scores and sp rank. This is an unfortunate artifact that is not going to be addressed.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Jimmy,
We employ Comet in a clinical Proteomics pipeline at Mayo.
We would like deterministic results to adhere to great patient quality.
We would be happy to submit a pull request in order to facilitate this, if you are too busy.
Can you provide some hints on the affected variables / threads?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This is due to threading and how results are stored. Peptides are only stored if they score greater than current lowest score. Depending on which threads finish first, the peptide list can change between runs, thus affective the sp scores and sp rank. This is an unfortunate artifact that is not going to be addressed.
Jimmy,
We employ Comet in a clinical Proteomics pipeline at Mayo.
We would like deterministic results to adhere to great patient quality.
We would be happy to submit a pull request in order to facilitate this, if you are too busy.
Can you provide some hints on the affected variables / threads?