Running DOCK

From DISI
Revision as of 20:32, 8 October 2012 by Therese (talk | contribs) (7 revisions)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Running DOCK

  • modify $mud/INDOCK and set up the desired directory structure – either manually or by running 'md4db.csh bysubset N1 N2 Type', where N1 is the identifier of the library (1: lead-like; 2: fragment-like), N2 is the number of chunks (i.e., jobs you can run in parallel), and Type is the category of library (i.e., bysubset, byvendor, etc).
  • if it hasn't been generated by a script, create the file dirlist, which conatins the list of the directories (i.e., chunks of the database) that you want to dock.
  • if you plan to use any of John's scripts in the downstream processing, leave the output file prefixes at test..
  • take care that the paths to the .db.gz files in split_database_index do not get too long. If they do, go via links.
  • submit the calculations to the cluster with $mud/submit.csh from the directory in which your data (most importantly, dirlist) resides. See MUD - Michael's Utilities for Docking for setting the $mud variable.