• Q1. In QuantCompare option, what is the difference between the protein ratio from Linear Regression and the protein composite ratio?
    A. The protein ratio from linear regression is conventional Census ratio using linear regression of light and heavy peptides in chromatogram. This approach may miss singleton proteins, because the regression score can be low for singleton proteins, requiring a user to use separate singleton tool in the IP2. Protein composite ratio combines both singleton and non-singleton peptides to generate a single ratio. Instead of using linear regression filter (e.g. R^2>=0.5), it uses profile score (PARK, NM 5, 319-322) to filter noisy peptides. Then, it takes a median ratio, the 50th percentile as protein ratio.
  • Q2. How to re-submit only failed jobs manually in the cluster?
    A.Follow the below steps.
  1. Connect to IP2 server by terminal, putty client or Mac terminal
  2. Type jobstatus command to find the folder name of the search you want to re-submit. Typically jobstatus has lines like below. In this case, folder name is projects2016_06_27_12_99515 ProLuCID /ip2_data/test/test_2015_11_10_15_78104/search/projects2016_06_27_12_99515/ projects2016_06_27_12_99515/ 99515 ip2 33 test false
  3. Connect to cluster by ‘ssh username@your_cluster.domain.name’
  4. Cd /change/to/your/job/folder. Usually this job path is $HOME folder, but it can be different depending on your cluster file server setup (contact your IT, if unsure)
  5. cd search_folder_name. In this example, it is projects2016_06_27_12_99515
  6. ls *my*job. Job files listed are unfinished jobs (Important. Please run command to list currently running jobs [e.g. qstat], and make sure xxx_my.jobs are indeed failed jobs).
  7. Delete matching sqt files. For example, for test_my_01.job, delete test_my_01.sqt, if sqt exists. Command is ‘rm test_my_01.sqt’
  8. Submit failed job. Command is ‘qsub test_my_01.job’ This command qsub is based on PBS or SGE clusters. Your command can be different if you use different type of cluster.

Feedback

Was this helpful?

Yes No
You indicated this topic was not helpful to you ...
Could you please leave a comment telling us why? Thank you!
Thanks for your feedback.

Post your comment on this topic.

Post Comment